Evaluation of the Health Effects
From Exposure to Gasoline and
Gasoline Vapors
Final Report
NESCAUM
Northeast States for Coordinated
Air Use Management
Air Toxics Committee
August 1989
-------
Evaluation of the Health Effects
From Exposure to Gasoline and
Gasoline Vapors
Final Report
August 1989
Prepared by:
Northeast States for Coordinated Air Use Management
(NESCAUM)
Air Toxics Committee
Technical Project Directors:
Margaret Round
Northeast States for Coordinated Air Use Management
Norman Anderson
U.S. Environmental Protection Agency, Region I
David Brown
Connecticut Department of Health Services
Project Manager:
Michael J. Bradley
Northeast States for Coordinated Air Use Management
-------
Second Printing, October 1989
Third Printing, November 1989
-------
TABLE OF CONTENTS
page
List of Tables xi
List of Figures xv
Acronyms xvi
Acknowledgements xvii
Peer Reviewers xix
Executive Summary xxi
1. Introduction 1-1
2. Methods 2-1
2.1 Introduction 2-1
2.2 Selection of Major Indicator Components of Gasoline 2-2
2.3 Literature Search 2-5
2.4 Summary 2-8
3. Substance Identification 3-1
3.1 Gasoline 3-1
3.1.1 Composition of Liquid Gasoline 3-4
3.2.1. Composition of Gasoline Vapors 3-8
3.2 Physical and Chemical Properties of Gasoline and Benzene,
Toluene, and Xylene 3-13
3.2.1 Gasoline 3-13
3.2.2. Benzene 3-15
3.2.3 Toluene 3-16
3.2.4 Xylene 3-17
3.3 Summary 3-18
4. Ecosystem Considerations 4-1
4.1 Introduction 4-1
4.2 Aquatic Vertebrates 4-3
4.3 Aquatic Invertebrates 4-5
4.4 Aquatic Flora and Microorganisms 4-6
4.5 Bioconcentration 4-7
111
-------
4.6 Biodegradation 4-9
4.7 Summary 4-10
5. Exposure Assessment 5-1
5.1 Sources and Fate of Gasoline and Gasoline Vapors 5-1
5.2 Exposure Assessment Scenarios 5-2
5.3 Exposure Assessment Methodology 5-3
5.3.1 Criteria for Selecting Monitoring Studies 5-4
5.3.2 Monitoring Methods Used in the Studies 5-5
5.4 Estimating Vapor Exposure At and Near Service Stations (Scenarios 1-3 5-6
5.4.1 Scenario 1 - Exposure of Self-Service Customers 5-7
5.4.2 Scenario 2 - Exposure of Service Station Attendants 5-8
5.4.3 Scenario 3 - Exposure of Residents in Neighborhoods
Near Service Stations 5-8
5.5 Estimating Exposure from Leaking Underground Tanks (Scenarios 4-6 5-13
5.5.1 Scenario 4 - Exposure of Residents Near Leaking Underground
Tanks 5-13
5.5.2 Scenario 5 - Exposure to Gasoline-Contaminated Water 5-14
5.5.3 Scenario 6 - Non-Ingestion Exposure to Contaminated Water 5-16
5.5.3.1 Estimation of Reasonable Upper Bound of Non-Ingestion
Exposure 5-19
5.6 Calculation of Body Burdens in Humans from Exposure to Gasoline,
Benzene, Toluene, and Xylene 5-22
5.6.1 Scenario 1 5-23
5.6.2 Scenario 2 5-27
5.6.3 Scenarios 5-28
5.6.4 Scenario 4 5-30
5.6.5 Scenarios 5-32
5.6.6 Scenario 6 5-32
5.7 Summary 5-33
6. Pharmacokinetics 6-1
6.1 Introduction 6-1
6.2 Pharmacokinetics of Gasoline 6-1
6.3 Pharmacokinetics of Benzene, Toluene, and Xylene 6-10
6.3.1 Absorption 6-10
IV
-------
6.3.1.1 Benzene 6-10
6.3.1.2 Toluene 6-11
6.3.1.3 Xylene 6-12
6.3.2 Distribution 6-13
6.3.2.1 Benzene 6-13
6.3.2.2 Toluene 6-13
6.3.2.3 Xylene 6-13
6.3.3 Metabolism 6-14
6.3.3.1 Benzene 6-14
6.3.3.2 Toluene 6-14
6.3.3.3 Xylene 6-16
6.3.4 Excretion 6-18
6.3.4.1 Benzene 6-18
6.3.4.2 Toluene 6-20
6.3.4.3 Xylene 6-20
6.4 Summary 6-21
7. General Toxicity of Gasoline and Specific Gasoline Components 7-1
7.1 Gasoline 7-4
7.1.1 Effects of Acute Exposure 7-4
7.1.1.1 Human Studies 7-4
7.1.1.2 Animal Studies 7-7
7.2.1 Effects of Subacute and Subchronic Exposure to Gasoline 7-12
7.1.2.1 Human Studies7-12
7.1.2.2 Animal Studies 7-12
7.1.3 Effects of Chronic Exposure to Gasoline 7-37
7.1.3.1 Human Studies 7-37
7.1.3.2 Animal Studies 7-41
7.2 Benzene 7-42
7.2.1 Effects of Acute Exposure to Benzene 7-42
7.2.1.1 Human Studies 7-42
7.2.1.2 Animal Studies 7-43
7.2.2 Effects of Subacute and Subchronic Exposure to Benzene 7-45
7.2.2.1 Human Studies 7-45
7.2.2.2 Animal Studies 7-46
7.2.3 Effects of Chronic Exposure to Benzene 7-52
-------
7.2.3.1 Human Studies 7-52
7.2.3.2 Animal Studies 7-57
7.3 Toluene 7-58
7.3.1 Effects of Acute Exposure to Toluene 7-58
7.3.1.1 Human Studies 7-58
7.3.1.2 Animal Studies 7-60
7.3.2 Effects of Subacute and Subchronic Exposure to Toluene 7-62
7.3.2.1 Human Studies 7-62
7.3.2.2 Animal Studies 7-62
7.3.3 Effects of Chronic Exposure to Toluene 7-65
7.3.3.1 Human Studies 7-65
7.3.3.2 Animal Studies 7-67
7.4 Xylene 7-68
7.4.1 Effects of Acute Exposure to Xylene 7-68
7.4.1.1 Human Studies 7-68
7.4.1.2 Animal Studies 7-69
7.4.2 Effects of Subacute and Subchronic Exposure to Xylene 7-70
7.4.2.1 Human Studies 7-70
7.4.2.2 Animal Studies 7-70
7.4.3 Effects of Chronic Exposure to Xylene 7-71
7.4.3.1 Human Studies 7-71
7.4.3.2 Animal Studies 7-72
7.5 Summary 7-73
8. Reproductive and Developmental Toxicity 8-1
8.1 Introduction 8-1
8.2 Gasoline 8-2
8.2.1 Reproductive Effects in Humans 8-2
8.2.2 Developmental Effects in Humans 8-2
8.2.3 Reproductive Effects in Animals 8-3
8.2.4 Developmental Effects in Animals 8-4
8.3 Benzene 8-5
8.3.1 Reproductive and Developmental Effects in Humans 8-5
8.3.2 Reproductive Effects in Animals 8-5
8.3.3 Developmental Effects in Animals 8-6
8.4 Toluene 8-11
VI
-------
8.4.1 Reproductive Effects in Humans 8-11
8.4.2 Developmental Effects in Humans 8-11
8.4.3 Reproductive Effects in Animals 8-12
8.4.4 Developmental Effects in Animals 8-12
8.5 Xylene 8-15
8.5.1 Reproductive and Developmental Effects in Humans 8-15
8.5.2 Reproductive Effects in Animals 8-16
8.5.3 Developmental Effects in Animals 8-16
8.6 Summary 8-22
9. Genetic Toxicity 9-1
9.1 Gasoline 9-1
9.1.1 Gene Mutation Studies 9-1
9.1.1.1 Bacterial Cells 9-1
9.1.1.2 Mammalian Cells 9-7
9.1.2 Studies of Chromosomal Aberrations 9-8
9.1.2.1 Mammalian Cells 9-8
9.1.3 Studies of Chemical Alternations of DNA 9-10
9.1.3.1 Mammalian Cells 9-10
9.2 Benzene 9-12
9.2.1 Gene Mutation Studies 9-12
9.2.1.1 Bacterial Cells 9-12
9.2.2 Studies of Chromosomal Aberrations 9-12
9.2.2.1 Mammalian Cells 9-12
9.2.2.2 Mammalian Cell Transformation 9-18
9.3 Toluene 9-18
9.3.1 Gene Mutation Studies 9-18
9.3.1.1 Bacterial Cells 9-18
9.3.1.2 Mammalian Cells 9-18
9.3.2 Studies of Chromosomal Aberrations 9-19
9.3.2.1 Mammalian Cells 9-19
9.4 Xylene 9-19
9.4.1 Gene Mutation Studies 9-19
9.4.1.1 Bacterial Cells 9-19
9.4.2 Studies of Chromosomal Aberrations 9-19
9.4.2.1 Mammalian Cells 9-19
vn
-------
9.6 Summary 9-22
10. Carcinogenicity 10-1
10.1 Epidemiologic Evidence of Human Carcinogenicity 10-1
10.1.1 Epidemiological Studies of Workers in the Gasoline
Service Industry 10-1
10.1.1.1 Case Control Studies 10-9
10.1.1.2 Cohort Studies 10-14
10.1.2 Epidemiological Studies of Workers in the Petroleum Refining
Industry and Other Petroleum-Based Industries 10-16
10.1.2.1 Case Control Studies 10-17
10.1.2.2 Cohort Studies 10-26
10.1.3 Summary of Epidemiological Studies in Progress 10-30
10.1.4 Benzene Carcinogenicity 10-31
10.1.5 Summary of Human Carcinogenicity Studies 10-34
Genito-Urinary Tract Tumors 10-34
Liver Tumors 10-35
Pulmonary Tumors 10-35
Hematopoieric Tumors 10-36
Skin 10-36
10.1.6 Summary 10-37
10.2 Animal Carcinogenicity 10-38
10.2.1 Introduction 10-38
10.2.2 Lifetime Inhalation Bioassays in Rats and Mice 10-40
10.2.2.1 Uncertainties in the Carcinogenicity Studies 10-42
10.2.2.2 Fuel Blend Issue 10-42
10.2.2.3 Renal Toxicity Issue 10-45
10.2.3 Skin Painting Study 10-47
10.2.4 Carcinogenicity Studies for Benzene, Toluene, and Xylene 10-47
10.2.4.1 Benzene 10-47
10.2.4.2 Toluene 10-48
10.2.4.3 Xylene 10-49
10.3 Qualitative Assessment of Cancer Risks of Gasoline 10-50
10.3.1 Corroboration 10-51
10.3.2 Multiple Routes of Exposure 10-51
10.3.3 Multiple Species 10-51
vui
-------
10.3.4 Carcinogenicity in Mammals 10-51
10.3.5 Causation of Malignant Tumors 10-52
10.3.6 Diversity of Tumor Types 10-52
10.3.7 Diversity of Primary Tumor Locations 10-52
10.3.8 Multiplicity of Tumors 10-52
10.3.9 Early Tumor Onset 10-53
10.3.10 Evidence of Genotoxicity 10-53
10.3.11 Structure Activity Relationship 10-53
10.4 Conclusions 10-53
10.5 Summary 10-55
11. Risk Assessment and Risk Characterization 11-1
11.1 Determination of Equivalent Doses 11-1
11.2 Derivation of Quantitative Risk Assessment Criteria 11-12
11.2.1 Derivation of Quantitative Health Criteria 11-12
11.3 Risk Characterization 11-23
11.3.1 Quantitative and Qualitative Assessment of Carcinogenic Risk 11-23
11.3.1.1 Cancer Risk Research Needs 11-30
11.3.2 Quantitative and Qualitative Assessment of Non-Cancer Risks 11-31
11.3.2.1 Non-Cancer Risk Comparisons 11-31
11.3.2.2 Non-Cancer Risk Research Needs 11-34
11.4 Uncertainties 11-34
Appendix A
Solubilities of the Principal Components of Gasoline
Appendix B
Review of the Literature on Gasoline Vapor Measurements at and Near Service Stations
Appendix C
ISCLT Dispersion Model Output
Appendix D
Derivation of Quantitative Risk Assessment Criteria
References
IX
-------
-------
LIST OF TABLES
page
2-1 Data employed in selection of indicator chemicals 2-3
2-2 Number of studies cited in RTECS for each gasoline component identified
in Table 2-1 2-6
2-3 Data bases searched 2-7
3-1 Major additives used in gasoline 3-3
3-2 Variations in composition of liquid gasoline 3-5
3-3 Major compounds of the principal hydrocarbon groups of unleaded gasoline 3-6
3-4 Refinery streams and components in an unleaded gasoline blended
specifically for hazard evaluation 3-7
3-5 Vapor composition of gasoline 3-9
3-6 Hydrocarbon composition (HC) in gasoline vapors as compared with liquid gasoline ..3-10
3-7 Principal components of the gasoline vapor phase 3-11
3-8 Twenty most soluble hydrocarbon components of gasoline 3-12
3-9 Selected physical properties of o-, m-, and p-xylenes, toluene, and benzene 3-14
4-1 The octanol-water partition coefficients for major components in gasoline 4-8
5-1 Gasoline vapor measurement studies considered in scenarios 1 and 2 5-8
5-2 Estimated exposure levels for scenarios 1 and 2 5-10
5-3 Estimated exposure levels for scenario 3 5-12
5-4 Estimated exposure levels for scenario 5 and 6 5-17
5-5 Summary of ambient concentrations and exposure doses associated with exposure
to gasoline and selected indicator constituents 5-24
6-1 Human blood/gas partition coefficients at 37°C for selected compounds in gasoline 6-2
6-2 Amounts of penetration of five aromatic hydrocarbons through rat skin 6-4
6-3 Amounts of penetration of five aliphatic hydrocarbons through rat skin 6-5
6-4 Human tissue-gas partition coefficients at 37°C for selected compounds in gasoline 6-8
7-1 Normal values for the cellular elements in human blood 7-3
7-2 Summary of the acute effects of gasoline in laboratory animals 7-8
XI
-------
7-3 Subchronic toxicity of leaded and unleaded gasoline 7-13
7-4 Summary of the composition and boiling ranges of the test materials
used in Haider et al. (1984) 7-19
7-5 Nephrotoxic effects in male rats following a 21-day inhalation exposure to an
unleaded gasoline blend 7-20
7-6 Nephrotoxic effects in male rats following a repeat 21-day inhalation exposure
to full-range alkylate naphtha 7-22
7-7 Nephrotoxic effects in male rats following a 21-day inhalation exposure to
thermal-cracked naphtha 7-23
7-8 Nephrotoxic effects in male rats following a 21-day inhalation exposure to
heavy catalytic-reformed naphtha 7-24
7-9 Nephrotoxic effects in rats following a 90-day inhalation exposure to an
unleaded gasoline blend 7-25
7-10 Grading system used to evaluate the severity of nephrotoxic responses 7-27
7-11 Average nephrotoxicity scores of tested gasoline components 7-28
7-12 Occurrence of pancytopenic period preceding leukemia in 42 cases with
long-term exposure to benzene 7-55
7-13 Dose-response relationship for 8-hour exposures to toluene 7-59
7-14 Mean concentrations of organic solvents in the breathing zone of 40 car painters 7-67
8-1 Summary of benzene inhalation teratology 8*7
8-2 Review of benzene inhalation teratology studies 8-8
8-3 Data of the fetuses of onho-, meta-, and pora-xylene treated CFY rats 8-17
9-1 Genotoxicity studies of unleaded gasoline 9-2
9-2 Genotoxic effects in humans following occupational exposure to gasoline 9-9
9-3 Genetic toxicology of benzene 9-13
9-4 Mutagenicity testing of xylene 9-20
10-1 Summary of epidemiological studies reviewed 10-2
10-2 Summary of site-specific statistically significant cancer findings from case-control
studies that evaluated employment in the gasoline service industry as a risk factor 10-10
10-3 Summary of site-specific statistically significant cancer findings from Milham's
1983 Washington state PMR study, by study population exposed 10-15
xu
-------
10-4 Summary of site-specific statistically significant cancer findings from PMR studies
that evaluated employment in the petroleum refinery industry as a risk factor 10-18
10-5 Summary of site-specific statistically significant cancer findings from cohort mortality
studies that evaluated employment in the petroleum refinery industry as a risk factor 10-19
10-6 Summary of site-specific statistically significant cancer findings from nested
case-control studies that evaluated different occupational categories in the
petroleum industry as risk factors 10-23
10-7 Summary of site-specific statistically significant cancer findings from case-control
studies that evaluated employment in the petroleum industry as a risk factor 10-25
10-8 Distribution of 2,2,4-trimethylpentane in Fischer 344 rats 10-39
10-9 Chronic gasoline inhalation study specifications of unleaded motor gasoline 10-41
10-10 Kidney tumor incidence in male Fischer-344 (F/344N) rats from chronic exposure
to vaporized unleaded gasoline in the MacFarland et al. study 10-43
10-11 Incidence of liver tumors in B6C3Fi mice exposed to vaporized gasoline 10-44
11-1 Summary of selected animal LOAELs and NOAELs derived from gasoline health
effects studies 11-8
11-2 Summary of reference air criteria for gasoline and gasoline components 11-16
11-3 Summary of estimated air concentrations corresponding to lifetime cancer
risks for gasoline 11 -20
11 -4 Non-cancer reference doses for gasoline and selected indicator constituents 11 -22
11-5 Potential cancer risks associated with exposure to gasoline and selected
indicator constituents 11 -24
11-6 Relative carcinogenic potencies of chemicals evaluated as known animal
carcinogens, or suspected of known human carcinogens 11 -26
11-7 Qualitative carcinogenicity categories established by the U.S. Environmental
Protection Agency 11-29
11-8 Potential non-cancer risks associated with exposure to gasoline and
selected indicator constituents 11-32
A-l Solubilities of gasoline components A-3
B-1 Gasoline vapor concentrations measured during refueling with a self-serve
type pump nozzle B-5
B-2 Gasoline vapor concentrations in the breathing zone measured during refueling B-7
B-3 Gasoline vapor concentrations measured during refueling at a service
station in Philadelphia B-9
Xlll
-------
C-1 Gasoline vapor mass emission rates for the Northeast states C-4
D-1 Summary of studies which examined the effects of benzene at concentrations
up to 50 ppm D-24
D-2a Summary of animal LOAELs and NOAELs derived from gasoline health
effects studies D-54
D-2b Summary of animal LOAELs and NOAELs derived from benzene health
effects studies D-57
D-2c Summary of animal LOAELs and NOAELs derived from toluene health
effects studies D-59
D-2d Summary of animal LOAELs and NOAELs derived from the xylene health
effects studies D-61
D-3 Summary of human LOAELs and NOAELs derived from studies on gasoline,
benzene, toluene, and xylene D-62
D-4a Summary of human equivalent doses and equivalent air concentrations
for LOAELs and NOAELs derived for gasoline D-63
D-4b Summary of human equivalent doses and equivalent air concentrations
for LOAELs and NOAELs derived for benzene D-65
D-4c Summary of human equivalent doses and equivalent air concentrations
for LOAELs and NOAELs derived for toluene D-67
D-4d Summary of human equivalent doses and equivalent air concentrations
for LOAELs and NOAELs derived for xylene D-69
D-5a Summary of reference air criteria for gasoline D-70
D-5b Summary of reference air criteria for benzene D-72
D-5c Summary of reference air criteria for toluene D-73
D-5d Summary of reference air criteria for xylene D-75
xiv
-------
LIST OF FIGURES
page
3-1 Components of a typical crude oil refining process 3-2
6-1 Skin penetration for five aromatic hydrocarbons 6-7
6-2 Pathways of benzene metabolism 6-15
6-3 Toluene metabolism in humans and animals 6-17
6-4 Proposed metabolic pathways of xylenes in animals and humans 6-19
xv
-------
ACRONYMS
ASTM
BCF
BTX
BZ
GARB
CI
cnr
CNS
ECoG
EEC
HCCN
IRDC
ISCLT
IU
LDH
LOAEL
LUSTs
MBTE
MON
NAG
NIOSH
NOAEL
NOEL
OCT
OR
PAHs
PCE
PMR
PNH
RDS
RON
SCE
SMR
STP
THC
TLV
TMP
TWA
UG
U.S. EPA
USTs
VER
American Society of Testing and Materials
Bioconcentration factor
Benzene, toluene, and xylene
Breathing zone
California Air Resources Board
Confidence interval
Chemical Industry Institute for Toxicology
Central nervous system
Electrocorticogram
Electroencephalogram
Heavy catalytic-cracked naphtha
International Research and Development Corporation
Industrial Source Complex Long-Term Dispersion Model
International units
Lactic dehydrogenase
Lowest observed adverse effect level
Leaking underground storage tanks
Methyl-butyl tertiary ether
Motor octane number
N-acetyl-beta-D-glucosaminadose
National Institute of Occupational Safety and Health
No observed adverse effect level
No observed effect level
Omithine carbamyl transferase
Odds ratio
Polycyclic aromatic hydrocarbons
Polychromatic erythrocytes
Proportionate mortality ratio
Paroxysmal nocturnal hemoglobinuria
Respiratory distress syndrome
Research octane number
Sister chromatid exchange
Standardized mortality ratio
Standard temperature and pressure
Total hydrocarbons
Threshold limit value
Trimethylpentane
Time-weighted average
Unleaded gasoline
U.S. Environmental Protection Agency
Underground storage tanks
Visual evoked response
xvi
-------
ACKNOWLEDGEMENTS
NESCAUM is grateful to the members of the Air Toxics Committee who made this
document possible by writing chapters, providing information, and reviewing drafts of the
document
Authors and reviewers of this document include:
Northeast States for Coordinated Air Use Management - Michael J. Bradley, Frank
Di Genova, Margaret M. Round, Nancy Seidman
Connecticut Department of Environmental Protection, Air Compliance Unit - Kim
Grossman, John Cove, Teresa McKinley, Robert Rubino, Bharati Vajjhala
Connecticut Department of Health Services, Toxics Hazard Section - David Brown,
Carolyn Jean Dupuy, Hari Rao
Maine Bureau of Health - Norman Anderson, Ann Melville
Maine Department of Environmental Protection, Bureau of Air Quality Control - Judy
Cohen, Ron Severance
Massachusetts Department of Environmental Protection, Office of Research and Standards -
Donna Bishop, Diane Manganaro, Carol Rowan
Massachusetts Department of Environmental Protection, Division of Air Quality Control -
James Neely
New Hampshire Department of Environmental Services, Air Resources Division - Richard
Andrews
New Hampshire Department of Public Health Services, Environmental Health Risk
Assessment - Amy Juchatz, Timothy Markey
New Jersey Department of Environmental Protection, Division of Environmental Quality -
Joann Held, Andy Opperman
New Jersey Department of Health, Environmental Health Program - Michael Berry, Cathy
Cunningham
New York Department of Environmental Conservation, Division of Air, Bureau of Air
Toxics - Robert Majewski, Carlos Monies, Virginia Rest, Moises Riano
New York Department of Health - Charles L. Ambrose, Judith S. Schreiber
Rhode Island Department of Environmental Management, Division of Air Resources -
Barbara Morin
Vermont Department of Environmental Conservation, Air Pollution Control Program -
Brian Fitzgerald, Harold Garabedian
U.S. Environmental Protection Agency, Air Management Division, Region I • Norman
Anderson, Barbara Beck, Sarah Levinson, Ann Walsh
xvu
-------
U.S. Environmental Protection Agency, Air Management Division, Region n - Robert
Kelly
U.S. Environmental Protection Agency - Richard Valentinetti
In addition, NESCAUM appreciates the comments of Dr. Perry Conn on the
exposure assessment and epidemiology chapters, Dr. Hari Rao on the reproductive and
developmental toxicity chapter, Ann Melville on the ecosystem considerations chapter, and
Dr. Barbara Callahan (Gradiant Corporation) on the risk assessment chapter. NESCAUM
also thanks Dr. David Brown and Norm Anderson for their valuable contributions to the
risk assessment and risk characterization chapters of the document and Barbara Smith-
Mandell, technical editor for this document
NESCAUM gratefully asknowledges the contributions of Peter Guldberg of Tech
Environmental who consulted with NESCAUM on the exposure assessment chapter and
Robert A. Michaels of RAM TRAC Corporation who consulted with NESCAUM on the
risk assessment and risk characterization chapters of this document.
The completion of this document would not have been possible without the
comments and continuing support of Norman Anderson, David Brown, and Barbara
Smith-Mandell.
The mention of commercial products, their source, or their use in connection with
material reported herein is not intended to be an actual or implied endorsement.
xvm
-------
PEER REVIEWERS
The background information for this document was developed by Dynamac
Corporation under the supervision of Dr. Norbert Page. The document submitted by
Dynamac Corporation was peer reviewed by the NESCAUM Air Toxics Committee and the
following individuals.
Robert Drew, Ph.D.
Health and Environmental Sciences Department
American Petroleum Institute
1220 L Street, NW
Washington, DC 20005
Myron Mehlman
Toxics Division
Mobil Oil
Pennington Rocky Hill Road
Pennington, NJ 08534
Lawrence Slimak
Director of Technical Affairs Division
Motor Vehicle Manufacturers Association of the U.S., Inc.
300 New Center Building
Detroit, MI 48202
John Holmes
Director of Research
California Air Resources Board
1102 Q Street
Sacramento, CA 95814
John Batchelder
Health Assessment Section
Research Division
California Air Resources Board
1102 Q Street
Sacramento, CA 95814
Joan Wiersema
Texas Air Resources Board
6330 Highway 290 East
Austin, TX 78723
S. Thomas Dydek, Ph.D.
Effect Evaluations Section
Research Division
Texas Air Resources Board
6330 Highway 290 East
Austin, TX 78723
xix
-------
Dr. Robert Bellies
Carcinogen Assessment Group
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
Comments submitted during the peer review process were reviewed by the Air
Toxics Committee. Based on the background information and these comments, the final
document was prepared under the technical direction of Dr. David Brown, Norman
Anderson, and Margaret Round. The final document was accepted by the Air Toxics
Committee in July 1989.
xx
-------
EXECUTIVE SUMMARY
Introduction
In response to a 1985 request by the Directors of the Northeast States for
Coordinated Air Use Management (NESCAUM), this document was prepared to assess the
public health risks of occupational and non-occupational exposure to unleaded gasoline.
Public health risks depend upon the presence of two necessary conditions: lexicological
potency and the potential for exposure.
This assessment estimates exposure of service station attendants, self-service
customers, and nearby residents to gasoline vapor emissions associated with service station
operations. Estimates are also determined for residents exposed to gasoline contaminated
drinking water and gasoline vapors which have migrated below ground into their homes
from leaking underground storage tanks. A review of the lexicological data is also
conducted in order to assess the health significance of these exposures.
Over the past several years, several of the Northeast states have developed expertise
in toxicology and risk assessment Although there is agreement among the states
concerning several areas of risk assessment, the NESCAUM states recognize that risk
assessments completed by agencies may differ from one another in their interpretation of
lexicological or exposure data. In addition, our understanding of gasoline-induced health
effects is likely to increase as more health data become available on the mixture, and as
improved techniques are developed for assessing these data. Given these considerations,
this document has been developed as a tool for assisting states in their management of
gasoline-related health risks, and with the expectation that its findings may be modified to
accommodate the various risk assessment and management approaches that exist among the
states within the region.
Approach
Standard risk assessment methods were adopted in this assessment, although it was
necessary to modify these methods because of limited lexicological data on the mixture and
ihe variability of the composition of the mixture after it is released into the environment.
This assessment of gasoline includes studies on the entire mixture, particular fractions of
the mixture, and specific components (benzene, toluene, and xylene) which are considered
to have the greatest health impacts.
Literature searches of available lexicological databases for gasoline, benzene,
toluene, and xylene were conducted. In addition, secondary literature, primarily developed
by the U.S EPA, was reviewed and utilized. Abstracts of human and animal studies are
xxi
-------
organized according to acute, subacute and subchronic and chronic toxicity; reproductive
and developmental toxicities; genetic toxicity, and carcinogenicity. Potential effects on
ecosystems are also considered
Exposure Assessment
Gasoline is a complex mixture of hydrocarbons and additives; the relative
concentration of gasoline components is dependent upon the crude oil source, refinery
process and product lines. Gasoline consists principally of paraffins (66 to 69 percent),
aromatics (24 to 27 percent), and olefins (6 to 8 percent). Chemicals are added to improve
engine performance. Gasoline exists in the environment in four states: as a free-moving
liquid, adsorbed into soil, in groundwater, and as an aerosol or vapor. Gasoline
components partition in environmental media according to vapor pressure, water solubility
and partition coefficients. Because benzene, toluene and xylene have high vapor pressure
and water solubility, they may exist in both the vapor phase and water soluble fraction of
gasoline.
The six exposure scenarios selected for analysis are:
Scenario 1: a self-service customer at a service station inhaling gasoline vapors.
Scenario 2: a full-time service station attendant inhaling gasoline vapors.
Scenario 3: an individual residing downwind of a nearby service station inhaling
gasoline vapors arising from gasoline pumps.
Scenario 4: an individual residing near a service station inhaling gasoline vapors
arising from a leaking underground storage tank.
Scenario 5: an individual residing near a service station ingesting gasoline
contaminated well water.
Scenario 6: an individual residing near a service station inhaling vapors which
arise indoors from use of contaminated well water and dermal contact with
contaminated water (e.g., showering).
The primary source of data for Scenarios 1 and 2 are monitoring studies of self-
service customers and service station attendants during refueling operations. Sources of
xxii
-------
gasoline vapors at service stations include losses from underground tanks, displacement
vapor losses from filler pipes during refueling, fuel spillage and evaporative and tailpipe
emissions from motor vehicles. Therefore, self-service customers are exposed to gasoline
vapors both during refueling operations and other time spent at the service station. Service
station attendants spend a full workday exposed to gasoline vapors from these sources.
Residents living downwind of a service station may be continuously exposed to these
emissions, but at significantly lower concentrations. In the case of residential exposure,
dispersion modelling is used to estimate ambient concentrations from service station
emissions. Because conditions associated with leaking underground storage tanks can vary
significantly from case-to-case, quantification of exposures is based on limited case study
information. Estimates for exposure and associated risks for any given site need to be
determined on a site-specific basis.
Hazard Identification
General exposure parameters were employed for adult and infant weight, and
corresponding ingestion and inhalation rates. Exposures associated with various exposure
durations, expressed as mg/kg/day, are calculated for each exposure scenario in order to
compare these exposures to lexicological criteria. Doses are calculated based on total
ventilation for pulmonary effects and alveolar ventilation for systemic effects. Using
alveolar ventilation as the basis for calculating systemic doses assumes that only the two-
thirds of the inhaled dose that reaches the alveoli is absorbed into the systemic circulation.
Non-ingestion exposure to gasoline contaminated water suggests a 2:1 non-ingestion to
ingestion ratio; exposure from non-ingestion sources (e.g., showering, dish-washing) are
comparable to and may be twice as much as the dose received from ingestion of 2 liters of
contaminated water per day.
The available toxicokinetic data on gasoline, while limited, show that gasoline is
absorbed from all exposure routes, including perinatal. The dermal route appears to be
slower than oral and inhalation routes. Some gasoline components are absorbed more
rapidly than others. For example, aromatic compounds (e.g., benzene, toluene, and
xylene) which have both high blood/air partition coefficients and skin penetration rates, are
absorbed more rapidly than other gasoline components. Metabolic pathways for benzene,
toluene and xylene are defined, but the toxic metabolites are not well understood.
Interactive effects at ambient exposure concentrations have not been characterized for these
compounds.
xxm
-------
General Toxicitv
The toxicity of gasoline, benzene, toluene and xylene has been investigated in both
short- and long-term exposure studies. The human studies generally involve acute
environmental (accidental and deliberate inhalation) and acute and chronic occupational
exposures to gasoline or to a mixture of gasoline components (particularly the aromatic
compounds). Studies on laboratory animals have focused on the subacute and subchronic
effects from exposure to gasoline and its major constituents (benzene, toluene, and xylene).
Acute exposure to gasoline and benzene, toluene, and xylene has been associated
with skin and sensory irritation, central nervous system depression, and effects on the
respiratory system. Prolonged exposures to these compounds also effects these organs as
well as the kidney, liver and blood systems. In general, the effects that have been identified
following gasoline exposure have also been identified for one or more of the specific
components of gasoline evaluated in this assessment For example, all substances have
been shown to be neurotoxic and studies that indicate that gasoline is hemotoxic are
supported by the abundant literature on benzene hematotoxicity.
The primary effects reported in several animal studies after protracted exposure to
gasoline vapors are pulmonary toxicity and nephrotoxicity. The studies investigating the
nephrotoxic effects in rats suggest a sex and species specific effect primarily from exposure
to the branched alkanes (e.g., trimethylpentane); however, renal changes have occasionally
been reported in female rats and mice exposed to certain distillate fractions of gasoline.
Thus, the nephrotoxic response observed in rodents may be influenced by several factors
including the exposure mixture, test protocol and/or the preferential distribution of the
gasoline components to the kidney. The exposure related lesions consist of increase foci of
regenerative epithelium in the renal cortex and dilated tubules at the corticomedullary
junction.
Reproductive and Developmental Effects
Reproductive and developmental effects are among the most sensitive non-cancer
toxic endpoints for benzene, toluene, and xylene exposures. These effects include
increased resorptions, reduced fetal body weight, and delayed skeletal development, and in
the case of benzene, induced bone marrow suppression in offspring. Benzene and xylene
have been shown to be teratogenic in rats at maternally toxic doses after inhalation and oral
exposure. Cleft palates in mice were also observed after oral exposure to xylene. In the
only reported teratogenicity study in animals exposed to gasoline vapors, reduced size of
fetuses in the high dose group was reported. Anecdotal data link chronic gasoline vapor
xxiv
-------
exposure of pregnant mothers to congenital central nervous system effects in their children.
Menstrual disorders in female workers exposed to gasoline vapors have also been reported.
Genetic Toxicity
Unleaded gasoline, benzene, toluene and xylene have been evaluated for genotoxic
effects in a variety of test systems. Generally, unleaded gasoline is not mutagenic in
bacterial systems while positive results have been recorded for sex-linked mutations in
Drosophilia malangaster, forward mutations with mouse lymphoma cells and induction of
unscheduled DNA synthesis. Chromosomal aberrations were observed in humans exposed
to gasoline vapors, although additional exposures confound the results of this study.
Benzene is clastogenic, particularly in mammalian cells. Toluene and xylene studies are
judged to be equivocal with negative evidence for mutagenicity and evidence of
chromosomal aberrations in rat bone marrow cells and workers exposed to these gasoline
components.
Carcinogenicity
One adequate carcinogen bioassay has been conducted with gasoline vapors. In
that study, statistically significant increases in kidney tumors in male Fischer 344 rats and
hepatocellular tumors in B6C3F1 mice were observed Major uncertainties are (1) the
vapor composition in this study was different from the ambient human environment and (2)
the kidney tumors observed in male rats may be the result of a mechanism specific to the
male rat and not female rats or other species. The male rat appears to selectively distribute
the hydrocarbons (e.g., 2,2,4-trimethylpentane) believed responsible for the nephrotoxicity
to the kidney. The female rat distributes significantly less of the hydrocarbon dose to the
kidney as the male rat. This may account for the higher sensitivity in the male rat and
weaker response in the female rat The development of renal tumors as a result of
nephrotoxicity has not been demonstrated in the rat. Thus, insufficient data are available on
the mechanism of male rat kidney tumors to discount the positive carcinogenicity data from
this bioassay. Carcinogenicity has been corroborated in more than one study, by multiple
routes of exposure, and in at least two species of laboratory animals. Therefore, it is the
finding of this assessment that animal bioassays provide sufficient basis for presuming
gasoline to be a probable human carcinogen.
An association between benzene exposure and hematopoietic tumors has been
found in human epidemiological studies. Hematopoietic system neoplasms, mainly
leukemias and lymphomas, are associated with rodents exposed to benzene via the oral and
inhalation routes. Other neoplastic lesions associated with exposure to benzene include
xxv
-------
carcinomas of the mammary glands, Zymbal gland, skin, oral cavity, nasal cavity, lungs,
and preputial gland; adenomas of the harderian gland and lungs; papillomas of the skin and
oral cavity; and tumors of the forestomach, liver, lungs and ovaries. Available data are
inadequate to determine the carcinogenicity of toluene or xylene.
Epidemiologic studies of unleaded gasoline are not available because insufficient
time has elapsed since its introduction in the mid-1970s. The results of epidemiologic
studies of typical gasoline exposure provide limited evidence for carcinogenicity in
humans. In general, these studies were limited by deficiencies in quantitative exposure data
and multiple exposure to other petroleum products and chemicals.
Risk Assessment
lexicological potency of gasoline and the indicator compounds were evaluated with
respect both to cancer and non-cancer effects. Assumption of non-threshold and low dose
linearity for cancer risks from exposure to gasoline or benzene were characterized based
upon the adoption of U.S EPA cancer potency values for unleaded gasoline and benzene.
This assessment has drawn a distinction between interspecies differences in toxicity
and in delivered dose. When assessing the systemic effects observed in animal studies,
human equivalent doses have been estimated by scaling the mg/kg/day animal doses by a
factor based on metabolism (body weight raised to the three-quarters power). Further
uncertainty factors were then applied to this dose in order to estimate a no effect level for
threshold effects in sensitive human populations.
In the quantitative assessment of non-cancer effects, critical studies were identified
from the general toxicity, reproductive and developmental toxicity, and genetic toxicity
chapters. Based on this evaluation, the most sensitive health effects associated with
gasoline, benzene, toluene, and xylene were determined. These health effect endpoints
included kidney toxicity (gasoline); genetic, hematopoietic, and developmental effects
(benzene); neurotoxic effects (toluene); and reproductive and fetotoxic effects (xylene).
Comparisons of human equivalent doses for the lowest effect levels observed in
animal studies show that sensitive toxicity endpoints for each substance reviewed in this
assessment (gasoline, benzene, toluene, and xylene) are associated with fairly definable
dose ranges. For gasoline, kidney toxicity is associated with human equivalent doses in
the 2 to 4 mg/kg/day dose range. For benzene, hematotoxicity occurs in the dose range of
0.1 to 1.0 mg/kg/day. For toluene, thresholds for sensitive neurobehavioral,
hematological, and immunological effects occur in the dose range of 0.5 to 1.5 mg/kg/day.
Following the identification of the most sensitive lexicological responses, the
studies most appropriate for risk assessment based on study design and toxicological
xxvi
-------
relevance were selected for quantitative risk assessment For-non-cancer threshold effects,
uncertainty factors were applied to account for interspecies differences between humans
and laboratory animals, exposure durations and sensitive populations. The non-cancer
reference doses for the most sensitive endpoints for gasoline, benzene, toluene, and xylene
are presented in Table 1.
Risk Characterization
Estimated cancer and non-cancer risks for each of the six scenarios are based upon
comparison of health criteria with estimated exposure doses. The health criteria for cancer
effects are cancer potency values; the health criteria for non-cancer effects are reference
doses (RfDs). Several uncertainties are associated with quantifying cancer and non-cancer
risks to humans based upon data from animal bioassays and epidemiological studies.
These uncertainties include: (1) estimates may exclude gasoline components of potential
concern; (2) inaccuracies in the assumptions about the intensity and duration of exposure;
(3) lack of information on interactive effects among constituents in the complex mixture,
and (4) uncertainties associated with exposure of sensitive individuals, including pregnant
women, the very young, and the old or infirm, as well as individuals who may suffer from
chronic respiratory, immunological, or other predisposing illnesses. These and other
uncertainties warrant the adoption of conservative assumptions, when possible, so that
errors are made on the side of caution. A reflection of these uncertainties is provided by
both average and upper limit exposure doses and health criteria.
Non-cancer health risks associated with gasoline exposure are presented in Table 2.
The reference doses are: gasoline - 0.003 mg/kg/day; benzene - 0.004 mg/kg/d; toluene -
0.0014 mg/kg/d and xylene - 0.034 mg/kg/day. Both mean and worst-case exposure
assumptions yield estimates of exposure doses that are greater than reference doses derived
in this assessment Some margins of safety, however, exist with regard to specific
indicator substances under all scenarios.
Potential individual lifetime (70 years) cancer risks associated with exposure to
unleaded gasoline and benzene are presented in Table 3. These cancer risks are based on a
cancer potency value of 0.0035 per mg/kg/day for gasoline and 0.026 mg/kg/day for
benzene. The exposure doses corresponding to one in a million cancer risk for gasoline
and benzene are estimated to be 2.8 x 10"4 mg/kg/day and 3.8 x 10'5 mg/kg/day,
respectively. Based upon an evaluation of available data, toluene and xylene are assigned
cancer potency values of zero. Maximum individual lifetime cancer risks associated with
gasoline and/or benzene are estimated to be 3.5 x 10~4 under scenario 1, 3.6 x 10~3 under
xx vu
-------
TABLE 1
NON-CANCER REFERENCE DOSES FOR GASOLINE AND SELECTED
INDICATOR CONSTITUENTS
substance
(toxic
endpoint)
Reference
air levels
(ug/m3)
adult infant
Reference
dose*
(mg/kg/d)
Exposure
Interval
Reference
oral dose**
(mg/L)
gasoline
(kidney effects)
15
11
0.003
Subchronic 0.10
benzene 19
(developmental effects)
0.004
Subacute
0.10
toluene
(neurotoxicity)
68 52 0.0014
Subchronic 0.05
xylene
(reproductive effects)
165
0.034
Subacute
1.2
* based upon assumed weights and pulmonary ventilation rates (when applicable), as
follows:
mouse: 0.025 kg, 0.029 cu M/day
monkey: 5 kg, 1.7 cu M/day
rat: 0.25kg, 0.14 cu M/day
human: 70 kg, 21.6 cu M/day (pulmonary)
14.4 cu M/day (systemic)
infant: 10 kg; 4.0 cu M/day (pulmonary)
2.7 cu M/day (systemic)
**
Oral references doses based on inhalation doses and consumption of 2 L water/day.
xxviu
-------
TABLE 2
POTENTIAL NON-CANCER RISKS ASSOCIATED WITH EXPOSURE TO
GASOLINE AND SELECTED INDICATOR CONSTITUENTS
Exposure Scenario
Estimated exposure3 (mg/kg/day)
mean maximum
Reference dose (RfD)b
fme/ke/d)
margin of safety (RfD/exp. dose)
mean maximum
scenario 1:
gasoline
benzene
toluene
xylenes
scenario 2:
gasoline
benzene
toluene
xylenes
scenario 3:
gasoline
benzene
toluene
xylenes
scenario 4:
gasoline
benzene
toluene
xylenes
scenario 5:
gasoline
benzene
toluene
xylenes
self-service customer at gas station exposed
9.4xlO-3 1.0 xlO'1
7.3 x ID'5 7.2 x lO'4
5.7 x ID'5 4.9 x 10'4
2.2 x 10'5 2.6 x 10'4
gas station attendant exposed via inhalation1
1.8
2.1 x ID'2 1.4 x 10'1
3.8 x ID'2
1.5 x lO'2
via inhalation 1>3
0.003
0.004
0.0014
0.034
,3
0.003
0.004
0.0014
0.034
0.32
55
25
1545
0.002
0.19
0.04
2
0.03
5
3
131
.
0.03
_
-
resident living downwind of gas station exposed via inhalation 1»3
3.1xlO-3 1.6xlO-2
2.6xlO-5 l.lxlO"4
6.2 x 10'5 2.9 x 10'4
2.7 x 10'5 1.3 x 10'4
resident inhaling vapors from nearby leaking
.
3.6x10-' 1.9
6.2 x 10- l 5.9
4.2X10'1 3.6
resident exposed to gasoline via ingestion of
1.7X10'1 2.9
1.4x10-2 7.0xlO-2
8.0 x ID'3 5.0 x ID'2
8.6 x ID'3 4.0 x ID'2
0.003
0.004
0.0014
0.034
underground storage tank*»*
0.003
0.004
0.0014
0.034
contaminated well water2'4
0.003
0.004
0.0014
0.034
0.97
154
23
1260
.
0.01
0.002
0.08
0.02
0.29
0.18
4
0.19
36
5
262
.
0.002
0.0002
0.0009
0.001
0.06
0.03
0.85
-------
TABLE 2
(CONTINUED)
Exposure Scenario Estimated exposure3 (mg/kg/day) Reference dose (RfD)b margin of safety (RfD/exp. dose)
mean maximum fmg/kg/d) mean maximum
scenario 6: resident exposed via inhalation and dermal contact during showering1'4'5
gasoline l.TxlO'1 3.4X10"1 0.003 0.02 0.009
benzene 1.4xlO'2 2.8 xlO'2 0.004 0.29 0.14
toluene 8.0 x 10'3 1.6 x 10'2 0.0014 0.18 0.09
xylenes 8.6 xlO'3 1.7 x UK2 0.034 4 2
a refer to Chapter 5
b refer to Table 11-4
1 assumes inhalation of 14.4 cu M/d, 24 h/d
2 assumes ingestion of 2 L water/day
3 based upon arithmetic means of monitoring studies described in "Exposure Assessment"
4 based upon limited case-study information. Estimated risks for any given site need to be determined on a site-specific basis.
5 assumes mean values equal mean drinking water exposures, and upper limits equal twice drinking water maxima
-------
TABLE 3
POTENTIAL CANCER RISKS ASSOCIATED WITH EXPOSURE TO
GASOLINE AND SELECTED INDICATOR CONSTITUENTS
Exposure Scenario
scenario 1: self-service
gasoline
benzene
toluene
xylenes
Estimated exposure3 (mg/kg/day)
mean maximum
customer at gas station exposed
9.4xlO-3 LOxlO'1
7.3 x 10'5 7.2 x lO'4
5.7 x 10'5 4.9 x 10'4
2.2 x 10'5 2.6 x 10'4
Cancer Potency b
fper mp/kpAtt
via inhalation^ >3
0.0035
0.026
Estimated lifetime cancer risk?
mean maximum
3.3 x 10'5 3.5 x lO'4
1.9xlO'6 1 .9xlO-5
scenario 2: gas station attendant exposed via inhalation**3
gasoline 1.8
benzene 2.1 x 10'2 1.4 x 10'1
toluene 3.8 x 10'2
xylenes l.SxlO'2
0.0035
0.026
scenario 3: resident living downwind of gas station exposed via inhalation1!3
gasoline 3.1 x 10'3 1.6xlO'2 0.0035
benzene 2.6 xlO'5 1.1 x 10'4 0.026
toluene 6.2 x 10'5 2.9 x 10'4
xylenes 2.7 xlO'5 1.3xlO'4
scenario 4: resident inhaling vapors from nearby leaking underground storage tank1*4
gasoline - - 0.0035
benzene 3.6X10'1 1.9 0.026
toluene 6.2 x 10'1 5.9
xylenes 4.2X10'1 3.6
scenario 5: resident exposed to gasoline via ingestion of contaminated well water2'4
gasoline 1.7 x 10'1 2.9 0.0035
benzene 1.4xlO'2 7.0 x 10'2 0.026
toluene 8.0 x 10'3 5.0 x 10'2
xylenes 8.6 x 10'3 4.0 x 10'2
6.3 x 10'3
5.5 x 1C'4 3.6 x ID'3
1.1 xlO'5 5.6 xlO'5
6.8 x 10'7 2.9 x 10'6
9.4 x ID'3 4.9 x ID'2
6.0 xlO'4 l.OxlO'2
3.6 xlO'4 1.8x10-3
-------
TABLE 3
(CONTINUED)
Exposure Scenario Estimated exposure3 (mg/kg/day) Cancer Potencyb Estimated lifetime cancer risk0
mean maximum (per mg/kg/d1) mean maximum
scenario 6: resident exposed via inhalation and dermal contact during showering*'4*^
gasoline l.TxlO'1 3.4x10-' 0.0035 6.0 xlO'4 1.1 x 10'3
benzene 1.4 x 10'2 2.8 xlO'2 0.026 3.6 x 10'4 7.3 x 10'4
toluene 8.0 xlO'3 1.6xlO'2
xylenes 8.6 xlO'3 1.7xHT2
a refer to Chapter 5
b U.S EPA Cancer Potency Values
x c estimated lifetime (70 years) cancer risk = (estimated exposure dose) x (assumed cancer risk)
x ' assumes inhalation of 14.4 cu M/d. 24 h/d
2 assumes ingestion of 2 L water/day
3 based upon arithmetic means of monitoring studies described in "Exposure Assessment"
4 based upon limited case-study information. Estimated risks for any given site need to be determined on a site-specific basis.
5 assumes mean values equal mean drinking water exposures, and upper limits equal twice drinking water maxima
-------
scenario 2; 5.6 x 10'5 under scenario 3; 4.9 x 10'^ under scenario 4; 1.0 x 10'2 under
scenario 5, and 1.1 x 10*3 under scenario 6.
It should be noted that although the exposure doses for scenarios 4,5, and 6 are
based on data from limited case studies, significant risks may be associated with such
exposures. Estimated risks for any given site, however, need to be determined on a site-
specific basis.
It is concluded in this assessment that gasoline and at least one of its major
constituents (benzene) are presumed human carcinogens. Exposure to gasoline and its
components is also associated with other adverse health effects such as toxicity to the
hematopoietic, kidney, liver, reproductive/developmental and nervous systems.
Comparison of cancer and non-cancer health criteria show that non-cancer reference doses
for gasoline and benzene correspond approximately to one cancer risk in one hundred
thousand.
xxxui
-------
1. INTRODUCTION
In 1985, at the request of its Directors, the Northeast States for Coordinated Air
Use Management (NESCAUM) began developing coordinated multi-state assessments of
specific toxic air pollutants. Each of the NESCAUM states (Connecticut, Massachusetts,
Maine, New Hampshire, New Jersey, New York, Rhode Island, and Vermont) has or is
developing an air toxic control program based on ambient air quality guidelines or on the
application of state-of-the-art control technology. The states recognized the need to develop
regional risk assessment documents to support regulatory decisions for major toxic air
pollutants. The NESCAUM states established the Air Toxics Committee to develop
comprehensive health assessments for specific toxic air pollutants. The NESCAUM Air
Toxics Committee has completed comprehensive health assessments for tetrachloroethylene
and trichloroethylene. This document is an assessment of the health impacts associated
with exposures to gasoline and three of its major components: benzene, toluene, and
xylene.
The known and potential health effects of gasoline, especially as related to
exposures from gasoline service station operations and from leaking underground storage
tanks (LUSTs), have been identified as a major concern of NESCAUM. The health effects
of gasoline are also a major concern of the Office of Underground Storage Tanks of the
U.S. Environmental Protection Agency (U.S. EPA), which has joined with NESCAUM to
conduct this assessment of the potential risk to the general public and service station
employees from such gasoline exposures.
Air releases of volatile gasoline components from service stations result in
exposures of employees, customers, and nearby residents. These exposures warrant
evaluations with regard to their potential health risks. These risks may be associated with
the complex hydrocarbon mixture as a whole or with specific toxic components of the
mixture (e.g., benzene). The State of California recently finalized regulations to require
vehicle refueling controls at all service stations with a monthly gasoline throughput of over
240,000 gallons per year. This action is being taken to reduce ambient benzene
concentrations (CARB, 1987).
Environmental contamination from gasoline leaking from underground storage
tanks (USTs) and associated piping also present a significant public health concern.
Gasoline may enter nearby homes as the plume passes near or through basements or as it is
transported into homes through contaminated ground water. Significant exposures to
gasoline and gasoline vapors may result from these leaks.
1-1
-------
According to the U.S. EPA (1985J), 1.4 million underground storage tank systems
are owned or operated at over 500,000 facilities; of these, 95 percent store petroleum
products. Approximately half of the petroleum storage tanks are used for retail motor fuel
sales. Eighty percent of the storage tanks are constructed of unprotected bare steel. U.S.
EPA estimates that 50,000 to 200,000 of these tanks may be leaking, although industry-
sponsored studies have tended to yield lower estimates. Regardless of the actual
percentage of USTs that are now leaking, there is a trend toward increased leaks as the
USTs age. Several major releases of gasoline from USTs have been documented (Dowd,
1984). For example, Maine Geological Survey reported that over 215 wells have been
polluted with petroleum products in Maine from 1980 to 1985, primarily as a result of
leaking underground storage tanks (Scudder and Anderson, 1986).
In addition to its direct impacts on health, gasoline contributes to the formation of
photochemical oxidant air pollution in the lower atmosphere. Gasoline vapors participate in
chemical reactions that lead to the buildup of ozone and other organic oxidants in the air.
These pollutants are irritants to the eyes and respiratory system. Chronic exposures to
gasoline vapors may also produce irreversible changes in the lung, leading to chronic
obstructive or restrictive lung diseases.
A discussion of the impact of gasoline vapor emissions on photochemical oxidant
formation is beyond the scope of this assessment. Such an analysis would require more
sophisticated exposure assessment techniques than are used in this report Yet, because
oxidant control strategies are based on the overall reductions of these precursor pollutants,
the direct and indirect impacts of gasoline vapors may be considered similarly in regulatory
control strategies designed to reduce emissions of volatile organic compounds.
In response to Clean Air Act ozone reduction requirements, many states have
implemented gasoline marketing control measures which have resulted in significant
reductions of gasoline vapor emissions. These measures include controls on gasoline
transfer operations at refineries, at bulk storage facilities, and from filling gasoline storage
tanks at service stations. Several states have implemented or are implementing vapor
control requirements for vehicle refueling at service stations (Stage 13 vapor recovery).
Furthermore, the Northeast states are in the process of implementing a regional program to
reduce the volatility of gasoline during the warmer months of the year, resulting in
substantial reductions in gasoline vapor emissions. Additional regional ozone control
strategies are currently being evaluated by the NESCAUM states. These include more
restrictive vehicle tailpipe standards and lowering the reactivity of exterior paints.
In this report, the health effects of gasoline and gasoline vapors are reviewed and
analyzed. There are limited data regarding the health effects from exposure to complex
1-2
-------
mixtures such as gasoline. Consequently, this analysis is also based on information on the
toxicity of three major components of gasoline: benzene, toluene, and xylene. These
toxicity data are then evaluated relative to the estimated human exposures that result from
six typical scenarios associated with environmental releases of gasoline in order to estimate
the corresponding public health impact.
This analysis has relied upon the assessment guidelines proposed by the U.S.
Environmental Protection Agency. The primary U.S. EPA guidance documents used are
U.S. EPA Guidelines of the Health Risk Assessment of Chemical Mixtures (1987a), U.S.
EPA Guidelines for Exposure Assessment (1987a), and U.S. EPA Guidelines for
Carcinogen Risk Assessment (1987a).
The NESCAUM Air Toxics Committee, via a designated Steering Committee,
developed the plans for this project and provided general directions for this assessment.
This assessment was subject to peer review by scientists from state and federal agencies,
from automobile and petroleum industries, and from academia. This report is intended to
provide state environmental protection agencies with a basis from which to assess the risks
associated with exposures to gasoline and gasoline vapors released into the ambient air and
from gasoline leaking from underground storage tanks.
1-3
-------
2. APPROACH
2.1 INTRODUCTION
The methods by which the health risks of gasoline and gasoline components are
evaluated in this document are consistent with generally accepted risk assessment
procedures for individual chemicals (U.S EPA, 1987a; OSTP, 1985). The assessment of
gasoline, however, requires special considerations in addition to those associated with
traditional chemical-specific risk assessments. Because the substance is a chemical
mixture, interactions among the individual components may produce a different
lexicological profile than would be predicted from an analysis of the individual
components. Yet, lexicological studies on chemical mixtures are not routinely conducted.
Even when mixture studies are available, the composition of the mixtures may vary among
the studies. Also, since the composition of gasoline changes after it is released into the
environment, it is difficult to correlate the findings based on exposure to the pure mixture
with the effects that may occur in actual exposure situations. Mixtures such as gasoline
also present significant analytical problems, particularly with regard to the quantification of
exposure.
An alternative approach to the assessment of chemical mixtures is the analysis of
specific components that are considered to have the greatest health impacts. This approach
has the advantage of reducing the uncertainties in the exposure assessment and of providing
a more quantifiable dose-response relationship. If it is assumed that antagonistic
interactions among chemicals do not occur at low doses (i.e., doses well below those
associated with metabolic saturation), the risks of the gasoline mixture should be at least as
great as those estimated for the indicator compounds.
This assessment recognizes the limitations of both approaches. Thus, while it
follows the preferred course of evaluating the health impacts of gasoline or gasoline vapor
as a mixture, the assessment also evaluates selected gasoline components in order to
provide a more complete lexicological profile. The result is the derivation of various health
criteria that are relevant to the evaluation of gasoline exposure.
Because of the scientific limitations associated with the identification of gasoline in
the assessment, no rigorous definition of the mixture is attempted. Instead the assessment
of gasoline as a mixture includes studies in which there is exposure to multiple
hydrocarbons within the general distillate fraction of gasoline. This exposure may range
from wholly vaporized gasoline to volatilized gasoline components under ambient
conditions, and may include exposure to specific combinations of chemicals (such as
benzene, toluene, and xylene) that are contained in gasoline. Although results from
2-1
-------
different mixture studies may not be directly correlated with actual exposures, this report
presents these findings to satisfy a basic risk assessment objective of considering all the
relevant information regarding exposures to gasoline before considering individual
components.
2.2 SELECTION OF MAJOR INDICATOR COMPONENTS OF GASOLINE
For an analysis based on specific components of the gasoline mixture, a preferred
alternative would be to select indicator compounds from specific subclasses of compounds
in the mixture based on shared structure-activity relationships. The toxicity of the mixture
could then be inferred from the lexicological profiles of indicator compounds selected from
each subclass. The lack of adequate lexicological information on many of gasoline's
components, however, makes this approach difficult to conduct
A more practical approach to evaluating such a complex mixture is to select a subset
of those chemicals that are generally present in gasoline and that present the greatest
potential pubb'c health risk. An evaluation of the total mixture is thus replaced by an
evaluation of the specific components that, considered individually, are assumed to
represent the greatest lexicological concerns. While such an approach may not adequately
address health concerns associated with exposure to the various gasoline mixtures in the
environment, it does represent the best practical alternative given the limitations of our
current knowledge.
The first step in the selection process is to consider gasoline components in
concentrations of greater than or equal to 1 percent of the mixture in liquid gasoline or in
gasoline vapors, and those gasoline components that are moderately soluble. This step
satisfies criteria for exposure and mobility in the atmosphere and groundwater.
Table 2-1 lists 16 chemicals that are generally found in liquid gasoline at
concentrations greater than or equal to 1 percent (Domask, 1984), and 12 which are present
in gasoline vapors (CEC, 1983; Domask, 1984; Tironi et al., 1986). The joint set
represents 21 chemicals. Those substances in gasoline that are moderately soluble (200
mg/L or more) include six additional chemicals (all olefins). Thus, a total of 27 chemicals
are evaluated for selection as indicator compounds.
The second step in the selection process is to estimate the available lexicological
database for each of the gasoline components identified. Preliminary lexicological
evaluation for selection of major components of gasoline is limited by the availability of
information regarding the toxicity of the compound. A survey of the Registry of Toxic
Effects of Chemical Substances (RTECS, 1987) is used to estimate the number of
published studies for each chemical. This survey is used as an indicator of the adequacy of
2-2
-------
TABLE 2-1
DATA EMPLOYED IN SELECTION OF INDICATOR CHEMICALS
Compound
High
Concentration
in Liquid
Gasoline
(>1%)
High
Concentration
in Vapor
Phase
Moderately
Soluble
(>200 mg/L)
n-Paraffins
Propane
n-Butane
n-Pentane
n-Hexane
n-Heptane
X
X
X
X
X
X
X
Isoparaffins
Isobutane
Isopentane X
2-Methylpentane X
3-Methylpentane X
2,2,4-Trimethylpentane X
2,3,4-Trimethylpentane X
X
X
X
X
Cvcloparaffins
Methylcyclopentane
Olefins
Trans-2-Butene
Cis-2-Butene
2-Methyl-l-Butene
Trans-2-Pentene
2-Methyl-2-Butene
Cyclopentene
Cis-2-Pentene
1,3-Butadiene
1,4-Pentadiene
2-Methyl-l, 1,3-Butadiene
~ Propylene
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2-3
-------
TABLE 2-1
(continued)
Compound
High
Concentration
in Liquid
Gasoline
(>1%)
High
Concentration
in Vapor
Phase
(>1%)
Moderately
Soluble
(>200 mg/L)
Aromarics
Benzene X X
Toluene X X
Xylenes X X
Ethylbenzene X
2-4
-------
the database for subsequent toxicological evaluation, and not as an indicator of the toxic
hazards associated with each chemical. Indeed, chemicals or mixtures of chemicals without
adequate databases may present a health toncem, but that possibility cannot be addressed in
this assessment. The size estimates of the databases are presented in Table 2-2. The
results show that there are more than two health effects studies for only six compounds:
benzene, toluene, xylene, n-hexane, ethylbenzene, and 1,3-butadiene.
Further screening finds benzene, toluene, and xylene to be important both in terms
of exposure and toxicity. N-hexane is a health concern because of its neurotoxic potential
and its presence in liquid gasoline. It is rejected as an indicator compound because: (1) it is
not present in gasoline vapors in 1 percent or greater, (2) it is virtually insoluble, thus
would not be expected to be a substantial groundwater contaminant; (3) it has been shown
to be toxic only after long-term exposure at 100 ppm or greater or at acute exposures in
excess of 2000 ppm for 10 minutes or more; and (4) it has not been shown to be
carcinogenic, mutagenic, or teratogenic. Thus, while n-hexane is a significant component
of gasoline, for the purposes of this assessment, this compound is assumed not to be of
concern for the relevant exposure situations.
Ethylbenzene is considered a major gasoline component since it is an aromatic
chemical and is present in liquid gasoline, often in excess of 1 percent. It is not very
soluble in water (152 mg/L), appears in vapors in less than 0.5 percent, and has a low
vapor pressure (10 mmHg). Thus, inhalation exposure of vapors or ingestion in drinking
water is likely to be minimal. Ethylbenzene has low toxicity compared to the other
aromatics in gasoline. Mutagenicity and teratogenicity testing produced negative results.
Based on the minimal exposure expected from ethylbenzene and its low toxicity,
ethylbenzene was not selected as a indicator compound for further assessment.
The considerations of 1,3-butadiene reflect its potential carcinogenicity and
mutagenicity, as well as its water solubility. However, exposure to this chemical has not
been documented in gasoline-contaminated water supplies. Furthermore, its exposure
potential appears to be much more clearly associated with gasoline exhaust emissions than
with gasoline itself. Based mainly on its low exposure potential, therefore, 1,3-butadiene
was not selected for further evaluation. The results of this analysis indicate that the
chemicals of greatest concern in gasoline are benzene, toluene, and xylene.
2.3 LITERATURE SEARCH
Computerized literature searches of the available toxicological data for both gasoline
and its major components (benzene, toluene, and xylene) were conducted using the
databases listed in Table 2-3. Additional health effects and exposure data were provided by
2-5
-------
TABLE 2-2
NUMBER OF STUDIES CITED IN RTECS FOR EACH GASOLINE
COMPONENT IDENTIFIED IN TABLE 2-1
Chemical Number of Studies
Propane 0
n-Butane 2
n-Pentane 1
n-Hexane 10
n-Heptane 2
Isobutane 0
Isopentane 0
2-Methylpentane 0
3-Methylpentane 0
2,2,4-Trimethylpentane 0
2,3,4-Trimethy Ipentane
Methylcyclopemane 1
Trans-2-Butene
Cis-2-Butene
2-Methyl-l-Butene 0
t-2-Pentene 0
2-Methyl-2-Butene 0
Cyclopentene 2
Cis-2-Pentene
1,3-Butadiene 14
1,4-Pentadiene 1
2-Methyl-1,3-Butadiene
'Propylene 0
Benzene 107
Toluene 37
Xylenes 24
Ethylbenzene 14
2-6
-------
Data
TABLE 2-3
DATA BASES SEARCHED
Time Coverage
MEDLARS
MEDLJNE
CHEMLINE
RTECS
HSDB
TOXLINE
BOX 74
TOX65
CANCERLFT
CANCERPROJ
OTHER
CIS/SANSS/OHMTADS/AQUIRE
EMIC
TSCA Inventory
TSCAT
1966-presem
1977-present
1979-present
1974-1978
1965-1973
1967-present
1967-present
DIALOG
AGRICOLA 1970-present
APTIC 1966-1978
AQUALINE 1974-present
AQUATIC SCIENCES & FISHERIES ABSTRACTS 1978-present
BIOSIS PREVIEWS 1969-present
CAB ABSTRACTS 1973-present
CA SEARCH 1967-present
CIN (Chemical Indust. Notes) 1974-present
CLAIMS/ U.S. PATENT ABSTRACTS 1971-present
CONFERENCE PAPERS INDEX 1973-present
CRGS (Chemical Regulations and Guidelines Systems) 1981-present
CRIS/USDA 1974-present
ENVIROLINE 1971-present
ENVIRONMENTAL BIBLIOGRAPHY 1974-present
EXCERPTA MEDICA 1974-present
IRL (Life Sciences Collection) 1978-present
NTIS 1970-present
OCEANIC ABSTRACTS 1964-prcsent
PTS PROMT 1972-present
PTS PREDALERT 1972-present
SCISEARCH 1974-present
SSIE (Current Research) 1979-present
1940-present
2-7
-------
General Motors Research Laboratory, Ford Motor Company, Chemical Industry Institute
of Toxicology (CUT), the U.S. EPA Mobile Source Emissions Branch, SAE Technical
Paper Services, and the U.S. EPA Air RISC. Documents prepared by the U.S. EPA on
the health effects of gasoline and benzene, toluene, and xylene were also reviewed in this
document Based on these computer and manual searches, pertinent studies were obtained
from published literature including scientific journals, industry reports, and government
publications.
Abstracts of human and animal toxicological studies for gasoline, benzene, toluene
and xylene are organized under the following toxic endpoints: acute, subacute and
subchronic, and chronic toxicity; genetic toxiciry; reproductive and developmental effects;
and carcinogenicity. It should be noted that toxicological studies of hydrocarbon mixtures
contained in gasoline are presented in the gasoline section.
The methodology and approach used in the exposure assessment is presented in the
introduction of Chapter 5.
This document was peer reviewed by representatives of state and federal
governments, industry, and academia. Comments and supplemental data received by the
reviewers were carefully reviewed and incorporated into the document as deemed
appropriate by the NESCAUM Air Toxics Committee.
2.4 SUMMARY
Gasoline is a mixture of hydrocarbons that have a potential for interaction in
biological systems. Assessment of this mixture required modification of the standard U.S.
EPA risk assessment process since the nature of the mixture varies from time to time and
the actions of each of the possible mixtures has not been determined. In order to
accomplish this, the typical mixture and components were evaluated separately. Three
indicator components (benzene, toluene, and xylene) were selected based on the potential
for exposure, their concentrations in gasoline, and the status of the scientific health study
data base. Information on the selected chemicals and gasoline as a mixture was assembled
from various sources, including computerized literature sources and documents submitted
by public and private institutions. All documents were read in detail by the risk assessment
working group and the overall document was peer reviewed.
2-8
-------
3. SUBSTANCE IDENTIFICATION
3.1 GASOLINE
Unleaded gasoline is a refined product of crude oil (petroleum) composed of a
complex mixture of hydrocarbons, additives, and blending agents. A typical crude oil
refining process is illustrated in Figure 3-1. Several methods are available to obtain and
blend the preferred gasoline streams:
Distillation of crude petroleum into naphtha, kerosene, gas oils and reduced crude
petroleum
Catalytic cracking of heavier petroleum fractions to produce simpler and lighter
molecules
Polymerization of low-molecular weight olefinic gases to form high-molecular
weight olefinic liquids for later use in blending
Hvdrocracking of heavy petroleum fractions into lighter fractions by cracking in the
presence of added hydrogen
Reforming of low-octane gasoline range hydrocarbons into high-octane branched or
aromatic hydrocarbons
Alkvlation to create branched alkane derivatives (e.g., 2,2,4-trimethylpentane) with
high-octane characteristics from smaller ones
Isomerization to convert straight chain hydrocarbons into branched chain
hydrocarbons of the same molecular weight; these hydrocarbons are used as
feedstocks for alkylation (Kirk-Othmer, 1980)
Gasoline is formulated from several refinery streams to achieve industry
specifications for physical properties, including boiling range and vapor pressure, and
desired seasonal performance standards, such as octane rating, to minimize pre-ignition or
knock. In the United States, gasolines are blended primarily from four refinery streams:
light catalytic cracked naphtha, heavy catalytic cracked naphtha, light catalytic reformate,
and light alkylate naphtha. After the various gasoline streams have been blended, sulfur
compounds are removed or converted to less odoriferous forms to reduce the odor in the
finished product.
In addition to blending refinery streams, various additives and blending agents are
used to improve the performance and stability of gasoline. Table 3-1 lists the types of
compounds comprising these groups of additives and presents a brief description of their
function. These additional compounds are classified according to their function and include
the following: anti-knock agents, anti-icing agents, anti-oxidants, upper-cylinder
3-1
-------
Figure 3-1
Components of a typical crude oil refining process
LICK' S»l 3H.
-!»tir»n»<» i .*a«u-«<.|
i-
|3
iW
— LH_H
CAr*t**iC
s
*
1
*
32:
:-
!
"
u
L»«'
1
m
t
5
4
«
i
nix •L»iO»*IO ««»»r«. i
I
rn*r
SOURCE: Adapted from Domask, 1984.
3-2
-------
TABLE 3-1
MAJOR ADDITIVES USED IN GASOLINE
Type of additive Specific compounds
Function
Anti-knock
Anti-oxidants
Meial deactivators
Lead scavengers
Anti-rust agents
Anti-icing agents
Upper-cylinder
lubricants
Detergents
Dyes
2,2,4-Trimethylpentane
Tetraethyl lead
Tetramethyl lead
Tert-butyl alcohol
Methyl tert-butyl ether
Ortho-alkylated phenols combined
with phenylene diamine
p-Phenylenediamine
Aminophenols
2,6-di-Tert-butyl-p-cresol
N^'-Disalicylidene-U-
Diaminopropane
Elhylene dibromide
Ethylenedichlonde
Fauy acids amines
Sulfonates
Alcohols
Glycols
Amides
Amines
Organophosphate salts
Cycloparaffinic distillates
Amino hydroxy amide
(RED): Alkyl derivatives of azeobenzene-
4-azo-2-naphthoI
(ORANGE): Benzene-azo-2-naphthol
(YELLOW): p-Diethyl-aminoazobenzene
(BLUE): 1,4-di-Isopropylaminoanthraquinone
Retard combustion to
preclude premature
combustion in the
engine cylinder.
Impart good storage
characteristics, prevent
gum formation.
Capture trace amounts of
copper derived from fueling
systems to prevent catalytic
oxidation.
Performance enhancer for
unleaded gasoline.
Coal metal surfaces and
eliminate corrosion due
to water.
Minimize stalls before
engine is hot
Reduce engine wear.
Prevent dirt buildup.
Identify product grade.
SOURCE: Adapted from Kirk-Othmer, 1980.
3-3
-------
lubricants, metal deactivators, detergents, lead scavengers, dyes, and anti-rust agents. The
U.S. Department of Energy (U.S. DOE, 1986) has a bibliography of 332 citations
regarding effects, evaluation, development, and production of gasoline additives.
Blending agents, such as tert-butyl alcohol, methyl-butyl tertiary ether (MBTE),
and other alcohols and ethers are used in unleaded gasoline as replacements for
organometallic anti-knock agents such as tertaethyl lead. Trace concentrations of various
elements including manganese, chromium, zinc, copper, iron, boron, magnesium, lead,
and sulfur are also present in most gasolines (lungers et al., 1975). For the most part,
these metals are native to the crude oil prior to refining.
3.1.1 Composition of Liquid Gasoline
Gasoline is a complex mixture of hydrocarbons comprised of paraffins (alkanes),
olefins (alkenes), and aromatics. Compounds containing sulfur, nitrogen, and oxygen are
also present in the gasoline refinery streams. Comparison of composition data from a
variety of gasoline blends indicates that the liquid gasoline generally consist of 66 to 69
percent paraffins, 24 to 27 percent aromatics, and 6 to 8 percent olefins (Battelle, 1985).
The composition of typical gasoline blends produced in the U.S. is presented in Table 3-2.
The relative concentrations of gasoline components may vary considerably
depending upon the crude oil source, refinery process, and product line. The major
compounds in the principle hydrocarbon groups are presented in Table 3-3. Typical
gasoline product lines may contain more than 150 separate compounds (Domask, 1984),
although as many as 1200 compounds have been identified in some blends (Whittemore,
1979). Appendix A presents a detailed listing of over 240 compounds that represent the
principle components of gasoline, including information on solubilities, boiling points, and
chemical formulas (API, 1985c).
It is important to note that most of the lexicological data on gasoline are based on
studies using a general blend of unleaded gasoline known as "PS-6". This blend may
differ from the commercial gasoline blends produced by a particular petroleum company.
The composition of the PS-6 gasoline blend used in a lexicological study sponsored by the
American Petroleum Institute (API) is presented in Table 3-4.
Unleaded gasoline contains higher concentrations of isoparaffins and aromatics than
does leaded gasoline. Although the amount of aromatics may be limited by new refinery
processes such as hydrocracking and isomerization, refiners may choose to add aromatics
to improve gasoline quality. For example, according to the California Air Resources Board
(CARB), benzene constituted about 1.4 volume percent of gasoline in 1984 and its conteni
3-4
-------
Component
TABLE 3-2
VARIATIONS IN COMPOSITION OF LIQUID GASOLINE
Volume (%)
Myer
Alkanes 56
Alkenes 7
Aromarics 18
Battelle Runion PS-6 0 to 145°F DuPont MVMA
62 62+9 63 86
11 7±1 9 13
27 33±7 28 1
Ave. Min. Max.
27 31 22 40
7 9 7 13
66 59 51 67
SOURCES:
Myer et al., 1975
Battelle, 1985
Runion, 1975 (mean of 9 US and European gasolines)
PS-6, from Domask, 1984
0 to 145°F fraction, from Haider et al., 1984
Unleaded commercial average, DuPont Road Octane Survey, 1976
Unleaded commercial fuels, Motor Vehicle Manufacturers Association Survey, 1980
3-5
-------
TABLE 3-3
MAJOR COMPOUNDS OF THE PRINCIPAL HYDROCARBON
GROUPS OF UNLEADED GASOLINE
Compounds
Volume (%)
Compounds
Volume (%)
n-Paraffins
n-Butane
n-Pentane
n-Hexane
Isoparaffins
Isobutane
Isopentane
10.19
1.14
10.26
Mono-olefins
Propylene
trans-2-Butene
cis-2-Butene
1-Pentene
trans-2-Pentene
cis-2-Pentene
0.03
0.75
1.22
2-Methylpentane
3-Methylpentane 8.81
2,3-Dimethylbutane
2-Methylhexane
3-Methylhexane 4.54
2,3-Dimethylpentane
2,4-Dimethylpentane
2,2,4-Trimethylpentane
2,3,4-Trimethylpentane 11.75
2,3,3-Trimethylpentane
2,2,3-Trimethylpentane
2-Methyloctane
3-Methyloctane 1.51
4-Methyloctane
Cvcloparaffins (Naphthenes)
Methylcyclohexane 0.15
l-cis-3-Dimethylcyclopentane 0.97
Cyclopentane 0.77
Methylcyclopentane
2-Methyl-l-pentene
2-Methyl-2-pentene
Aromarics
Benzene
Toluene
o-Xylene
m-Xylene
p-Xylene
Ethylbenzene
1-Methyl-1,3-ethylbenzene
l-Methyl-4-ethylbenzene
1,2,4-Trimethylbenzene
1.26
1.3
3.99
9.83
5.33
SOURCE: Domask, 1984.
3-6
-------
TABLE 3-4
REFINERY STREAMS AND COMPONENTS IN AN UNLEADED GASOLINE
BLENDED SPECIFICALLY FOR HAZARD EVALUATION
Stream Volume
Plus
Ethyl 733 Antioxidant 51bs/1000 bbl.
DuPont DMD-2 Metal Deactivator 51bs/1000 bbl.
Light Catalytic Cracked Naphtha 7.6
C4 to Ci i hydrocarbons, large proportion
of unsaturated hydrocarbons
Heavy Catalytic Cracked Naphtha 44.5
C6 to Ci2 hydrocarbons, large proportion
of unsaturated hydrocarbons
Light Catalytic Reformed Naphtha 21.3
Cs to Cn hydrocarbons, large proportion
of aromatic and branched hydrocarbons
Light Alkylate Naphtha 22.0
Cy to CIQ hydrocarbons, large proportion
of branched, saturated hydrocarbons
Benzene added to bring to 2% 0.8
Butane added to increase Reid Vapor Pressure 3 . 8
100.0
SOURCE: MacFarland et al., 1984.
3-7
-------
in gasoline is projected to increase 31 percent to 1.8 volume percent by 1990 as lead is
phased out (GARB, 1987).
3.1.2 Composition of Gasoline Vapors
Table 3-5 presents the composition of gasoline vapors reported in the literature from
several refueling studies. Comparison of these data indicate that there is a similarity in the
composition of the major chemical classes in gasoline vapor regardless of the product line.
Because gasoline components have different vapor pressures and solubilities, their
concentrations in the environment vary greatly from liquid gasoline. This is illustrated in
Table 3-6 in which the hydrocarbon composition in gasoline vapors is compared with that
in liquid gasoline.
Gasoline vapors are produced from aerosolization of liquid gasoline (used primarily
for research purposes) or volatilization of components at the air/liquid interface (Battelle,
1985). Aerosolization results in the vaporization of all components in the liquid gasoline
whereas the gasoline vapors in ambient air (e.g., while refueling cars) are comprised
primarily of the highly volatile components of gasoline. According to a refueling
monitoring study by McDermott and Killiany (1978), 21 components dominate the vapor
phase of gasoline. These components are listed in Table 3-7. The two dominant
compounds in the vapor phase were n-butane and isopentane, which represent
approximately 38 percent and 23 percent by volume of the vapor composition.
Differences in hydrocarbon composition also exist between summer and winter
gasoline vapor samples. In a study by Tironi et al. (1986), the primary difference between
the seasonal blends was a greater content of n-butane in the winter. This increase in n-
butane is consistent with the higher vapor pressure of winter-grade gasoline. N-butane,
isopentane, isobutane, n-pentane, and propane represent 82 percent of the vapor
composition in the winter and 73 percent in the summer. The dominant aromatics were
benzene and toluene, present in 0.5 percent in the winter and about 1 percent in the
summer.
The formation of gasoline vapors may also result from the domestic use of
gasoline-contaminated water. In this case, liquid gasoline has migrated through the soil to
the water table where it will spread out over the surface and float on top of the water table.
Components will then partition into environmental media according to their vapor pressure
and water solubility. The soluble components will dissolve into the groundwater and be
transported through the aquifer, and the highly volatile components will tend to volatilize
upward through the soil pore spaces. The 20 most soluble components of gasoline, based
on aqueous solubility data, are presented in Table 3-8. Coleman et al. (1983) reported that
3-8
-------
TABLE 3-5
VAPOR COMPOSITION OF GASOLINE
Volume %
Alkanes
Alkenes
Aromatics
McDermott
84
6
3
Haider
93
2
4
Battelle
92
5
5
R union
89
8
1
SOURCE: McDermott and Killiany, 1978; Haider etal., 1984; Battelle, 1985; Runion, 1975.
3-9
-------
TABLE 3-6
HYDROCARBON COMPOSITION (HC) IN GASOLINE VAPORS
AS COMPARED WITH LIQUID GASOLINE
Compound
n-Paraffins
Propane
n-Butane
n-Pentane
n-Hexane
n-Heptane
Isoparaffins
Isobutane
Isopentane
2-Methylpentane
3-Methylpentane
2,2,4-Trimethylpentane
2,3.4-Trimethylpentane
Cycloparaffins
Methylcyclopentane
Methylcyclohexane
Olefins
Trans-2-Butene
Cis-2-Butene
2-Methyl-l-butenc
t-2-Pentene
2-Methyl-2-butene
Aromatics
Benzene
Toluene
Xylenes
Ethyl benzene
Liquid
Phase (%)
0.1
6.2
4.0
2.7
1.3
0.7
7.4
3.6
2.6
1.8
1.1
1.7
0.9
0.4
0.4
0.9
1.2
1.7
2.1
10.4
4.9
1.2
Vapor*
Phase (%)
5.2
41.1
5.6
0.9
0.2
8.8
16.4
2.1
1.2
0.2
0.1
0.6
0.1
1.7
1.7
1.4
1.6
2.0
0.9
0.8
0.1
0.4
V/L Phase
Ratio
52
6.6
1.4
0.33
0.15
12.6
2.2
0.58
0.46
0.11
0.09
0.35
0.11
4.25
4.25
1.56
1.33
1.18
0.43
0.08
0.02
0.33
* Averaged for winter and summer concentrations.
SOURCE: Adapted from Tironi et aL, 1986.
3-10
-------
TABLE 3-7
PRINCIPAL COMPONENTS OF THE GASOLINE VAPOR PHASE
Vapor composition
airborne easoline
Compound
Alkanes
Propane
Normal Butane
Isobutane
Isopentane
Normal pentane
Cyclopentane
2,3-Dimethylbutane
2-Methylpentane
3-Methylpentane
Normal hexane
Methyl cyclopentane
2,4-Dimethylpentane
2,3-Dimethylpentane
2,2,4-Trimethylpentane
Alkenes
Isobutylene
2-Methyl-l-butene
cis-2-Pentene
2-Methyl-2-butene
Aromatics
Benzene
Toluene
Xylene (p-, m-, o)
Boiling
point, °C
-42.1
-0.5
-11.7
27.9
36.1
49.3
58.0
60.3
63.3
68.7
71.8
80.3
89.8
99.2
6.9
31.2
37.0
38.6
80.1
110.6
142.0
Mean volume
percent
0.8
38.1
- 5. -2
.22.9
7.0
0.7
0.7
2.1
1.6
1.5
1.3
0.4
0.7
0.5
1.1
1.6
1.2
1.7
0.7
1.8
0.5
Total percent 92.1
Standard
deviation
1.1
5.7
1.9
6.1
4.0
0.7
0.5
1.3
0.9
0.9
0.4
0.5
0.6
0.5
1.5
2.1
1.7
.8
0.4
1.3
0.6
SOURCE: Adapted from McDermott and Killiany, 1978.
3-11
-------
TABLE 3-8
TWENTY MOST SOLUBLE HYDROCARBON
COMPONENTS OF
1. Benzene 11. 1-Methylcyclopentene
2. 1,3-Butadiene 12. 3-MethyIcyclopentene
3. 2-Methyl-l,3-butadiene 13. 2-Pentene
4. 1,4-Pentadiene 14. Ethylbenzene
5. Cyclopentene 15. o-Xylene
6. Toluene 16. p-Xylene
7. 2-Butene 17. m-Xylene
8. Isobutene 18. Cyclopentane
9. 1-Butene 19. 2-Methyl-l-butene
10. Cyclohexene 20. 2-Methyl-2-butene
a Based on solubility data presented in Appendix A (API, 1986b).
b There are several additives of blending agents such as ethanol and methyl t-butyl ether that are
more soluble than the hydrocarbon components listed here.
3-12
-------
at least 95 percent by weight of the components in the water soluble fraction of gasoline are
aromatic compounds having 6 to 13 carbons.
3.2 PHYSICAL AND CHEMICAL PROPERTIES OF GASOLINE AND
BENZENE. TOLUENE. AND XYLENE
Gasoline exists in the environment in four states: as a free-moving liquid, adsorbed
to soil, as a solute in groundwater, and in the air as an aerosol and as vapor. The fate and
transport of gasoline in the environment are governed by the physical and chemical
properties of gasoline components and the properties of the environmental medium in
which they exist. Because of the wide array of gasoline types, only a general
characterization of the mixture of gasoline is presented. A detailed discussion of the
physical and chemical properties of the major components of gasoline (benzene, toluene,
and xylene) is presented. Selected physical properties of these components are presented in
Table 3-9.
3.2.1 Gasoline
The physical and chemical properties of gasoline vary widely. The hydrocarbons
contained in gasoline distill within a range of 100° to 400°F. Gasoline is saturated in
distilled water at 299 mg/L at 20°C. The properties and specifications for gasoline are
developed by The American Society for Testing and Materials (ASTM-D439). According
to ASTM specifications, each grade of gasoline should meet specifications for physical
properties such as boiling point, vapor pressure, and color, as well as octane ratings and
volatility limits. ASTM (1984) has determined that all categories of unleaded gasolines
should meet the following specifications: lead content - 0.013 g/L maximum; distillation
residue - 2 percent maximum; existent gum - 5 mg/100 ml maximum; sulfur - 0.10 weight
percent maximum; and phosphorus - 0.0013 g/L maximum. The Clean Air Act of 1970
(42 USC 1957 et seg.) requires that unleaded gasoline contain no more than 0.05 g of lead
and no more than 0.005 g of phosphorus per gallon. In addition, the gasoline must pass a
corrosion test to guarantee it will not corrode the metal in a fuel system and an oxidation
stability test to assure it will not form gum during short-term storage.
Five volatility classes are recognized by ASTM as a result of user performance
requirements under varying weather conditions. These classes are used in a schedule
developed by ASTM that indicates seasonal and geographical distribution of gasoline and
specifies the appropriate volatility class for each month for all areas of the United States,
based on altitude and probable air temperature.
3-13
-------
TABLE 3-9
SELECTED PHYSICAL PROPERTIES OF 0-, M-, AND P-XYLENES, TOLUENE, AND BENZENE
Propertv
Autoignition temperature, °C
Boiling point, °C
Melting (solidification) point, °C
Flashpoint (closed cup), °C
Density (g/cm^)
at25°C
at20°C
Log KQW
Vapor pressure, mmHg
at30°C
at25°C
at20°C
at!5°C
Water solubility, ppm
at25°C
at20°C
Seawater solubility, ppm
at25°C
Air conversion factors
1 mg/m3
1 ppm
Vapor density (air=l)
Flammability limits
(% by volume in air)
p-Xvlene
NA
139.12
-25.2
17
0.8802
NA
3.13
NA
6.62
5
NA
175
170.5
220.8
NA
129.6
0.23 ppm
4.34 mg/m3
NA
NA
o-Xvlene
NA
144.41
-47.9
25
0.8642
NA
3.20
NA
8.25
6
NA
146
196
160.3
NA
106.0
0.23 ppm
4.34 mg/m3
NA
NA
m-Xvlene
NA
138.37
13.3
25
0.8610
NA
3.18
NA
8.78
6.5
NA
156
198
214.5
NA
110.9
0.23 ppm
4.34 mg/m3
NA
NA
Toluene
552
110.6
-95
40
NA
0.8664
2.65
NA
28.7
NA
NA
534.8
NA
NA
NA
379.3
0.27 ppm
3.77 mg/m3
3.20
1.17 to 7.10
Benzene
560
80.1
5,5
-11
100
0.8765
2.13
118
100
76
60
1,800
NA
NA
1,780
NA
0.31 ppm
3.26 mg/m3
2.77
1.3 to 7.10
NA=Not available
SOURCE: McAuliffe, 1963; Ransley, 1984; Tewari el al., 1982; U.S. EPA, 1983a, b, 1986f; Verschueren, 1983; CRC, 1987.
-------
In the absence of a federal program to control gasoline volatility, NESCAUM began
evaluating a regional control approach in the fall of 1986. In February 1987, NESCAUM
first announced the Northeast Gasoline Volatility Reduction Strategy. NESCAUM
sponsored a public forum on April 23,1987, to receive comments from interested parties
on the Northeast states' proposal to reduce gasoline volatility. The Environmental
Commissioners in the Northeast states signed a Memorandum of Understanding on
November 17,1987, stating their intention "to propose a gasoline volatility control
program." Beginning in January 1988, states held public hearings and adopted regulations
according to their administrative procedures.
Seven NESCAUM states (Connecticut, Maine, Massachusetts, New Jersey, New
York, Rhode Island, and Vermont) have adopted final state gasoline volatility regulations.
In addition, the U.S. Environmental Protection Agency has passed national regulations
restricting the volatility of gasoline to different levels for different parts of the country (54
FR 11868). Because the Northeast states wanted to set lower limits on volatility, all of the
seven states submitted requests as part of their plans (State Implementation Plan or SIP) to
attain the ozone standard. Once the requests are approved by the U.S. EPA, the state
regulations take precedence over the less restrictive national regulation. SIPs have been
approved for Connecticut, Massachusetts, New Jersey, New York, and Rhode Island, and
are pending for Maine and Vermont. New Hampshire plans to have regulations in place for
the summer of 1990.
In addition to volatility classes, gasolines are graded into three anti-knock index
levels. The anti-knock index is determined by averaging the research octane number
(RON) and the motor octane number (MON). The higher the anti-knock index (ranging
from 85 to >90), the greater the anti-knock performance under severe operating conditions
(ASTM, 1984).
3.2.2 Ben/ene
Benzene usually comprises approximately 1 to 2 percent (by volume) of unleaded
gasoline (Runion, 1975). Benzene (formula C6H6) is a simple, single-ring organic
compound of the type known as an aromatic hydrocarbon (U.S. DHHS, 1983; U.S.EPA,
1983a; CRC, 1968). It has been assigned CAS No. 71-43-2, and has a molecular weight
of 78.11 daltons. Benzene exhibits a melting (freezing) point of 5.553°C. At standard
temperature and pressure (STP), benzene exists as a volatile, colorless, flammable liquid
having a density of 0.8787 (g/ml), measured at 16°C. At STP, it exhibits a boiling point of
80.1°C. The solubility of benzene is partial in water (1.8 g/L at 26°C) and infinite in
alcohols, ethers, acetone (CH3COCH3), and chloroform (CHC13). In the absence of
3-15
-------
turbulent mixing, benzene that is undissolved in water floats on the surface, as indicated by
the fact that its density is less than that of water (1.0 g/ml).
Benzene is highly flammable, exhibiting lower to upper limits of flammability in air
estimated at between 1.5 and 8.0 percent by volume (U.S.EPA, 1983a). Its flashpoint is
relatively low, at -11.1°C. The volatility of benzene is quantified by its vapor pressure
which, measured at 26°C, is 100 mmHg. Benzene vapors are nearly three times as dense
as air. In the absence of turbulent mixing, therefore, benzene may accumulate in the
breathing zone within confined spaces into which it has been released. One ppm of
benzene is equivalent to 3.19 mg/m3 (Patty's, 1981).
Benzene tends to form azeotropes, which may be important in such contexts as
designing systems to purify contaminated water via evaporation (U.S.EPA, 1983a).
Azeotropes are mixtures of liquids exhibiting constant properties, such as boiling points
and freezing points, that characterize the mixture rather than each component thereof.
Benzene forms a two-phase, or binary, azeotrope with water consisting of 91 percent
benzene by weight. The boiling point of this azeotrope is 69°C, which is distinct from the
boiling point of either pure benzene (80.1°C) or pure water (100°C). Benzene may also
form three-phase, or ternary, azeotropes with water and other organic liquids. The
formation of azeotropes with water and other organic liquids may significantly influence
exposure processes involving complex mixtures, such as gasoline.
3.2.3 Toluene
Toluene (methyl benzene, formula C6H6CH3) is a simple, single-ring organic
compound of the type known as an aromatic hydrocarbon (Anderson, 1987; CRC, 1987).
It has been assigned CAS No. 108-88-3, and has a molecular weight of 92.13 daltons.
Toluene exhibits a melting (freezing) point of-95°C. At STP, toluene exists as a volatile,
clear, colorless, flammable liquid having a density of 0.8669 (g/ml), measured at 20°C. At
STP, it exhibits a boiling point of 110.6°C. Toluene is slightly soluble in water ( Patty's,
1981; CRC, 1968). Its solubility is infinite in alcohols, ethers, and benzene, and slight in
acetone (CH3COCH3) (CRC, 1968). In the absence of turbulent mixing, toluene that is
undissolved in water floats on the surface, as indicated by the fact that its density is less
than that of water (1.0 g/ml).
Toluene is highly flammable, exhibiting lower to upper limits of flammability in air
estimated at between 1.4 and 6.7 percent by volume (Patty's, 1981). Its flashpoint is
relatively low, at 4.4°C. The volatility of toluene is quantified by its vapor pressure which
is 36.7 mmHg at 30°C (Patty's, 1981). Toluene vapors are 3.1 times as dense as air. In
the absence of turbulent mixing, therefore, toluene may accumulate in the breathing zone
3-16
-------
within confined spaces into which it has been released One ppm of toluene is equivalent to
3.77 mg/m3 (Patty's, 1981).
Chemically, toluene undergoes substitution reactions, either on the aliphatic group
(-CH3) or on the ortho or para positions of the benzene ring. Nitration, sulfonation,
halogenation, methylation, and chloromethylation of toluene are some examples of
substitution reactions, which occur between 2.1 and 467 times faster than with benzene
(U.S. EPA, 1983b). Toluene can be converted to benzene following dealkylation of the
methyl group in the presence of heat or a catalyst Toluene can also undergo
disproportionation, transalkylation, hydrogenation, hydrolysis, and oxidation reactions to
produce various chemicals, including methylcyclohexane, benzoic acid, and benzaldehyde
(U.S. EPA, 1983b).
3.2.4 Xvlene
Xylenes consist of the onho, meta, and para isomers of dimethyl benzene (formula
C6H5(CH3)2) (NTP, 1986a). They are simple, single-ring organic compounds of the type
known as aromatic hydrocarbons (NTP, 1986a; Patty's, 1981). The molecular weight of
xylenes is 106.17 daltons. The technical grade, which has been assigned CAS No. 1330-
20-7, also includes ethylbenzene (C6H5CH2CH3; molecular weight 106.17). The melting
(freezing) points of the xylenes are -25°C, -47.4°C, and 13.3°C for o-, m-, and p-xylene,
respectively, and -94°C for ethylbenzene (CRC, 1968). At STP, xylenes exist as a volatile,
clear, colorless, flammable liquid having a density of 0.864 (g/ml), measured at between
4°C and 20°C (NTP, 1986a; CRC, 1968). At STP, they exhibit boiling points in the range
of 136 to 144°C. Xylenes are slighdy soluble in water (CRC, 1968). Their solubility is
infinite in alcohols and ethers (CRC, 1968). In the absence of turbulent mixing, xylenes
not dissolved in water float on the surface, as indicated by the fact that their density is
0.864: less than that of water (1.0 g/ml).
Xylenes are highly flammable, exhibiting lower to upper limits of flammability in
air estimated at between 1.0 and 7.0 percent by volume (Patty's, 1981). Their flashpoint is
37.6°C. The volatility of xylenes is quantified by its vapor pressure, which is 6.16 mmHg
at 20°C. One ppm of xylenes is equivalent to 4.34 mg/m^.
Xylene is produced in various commercial grades. About 70 percent of all the
mixed xylene grades produced commercially in 1984 had a purity of 98 to 100 percent, and
the remainder had a purity of 90 to 97.9 percent (U.S. EPA, 19860- In addition to the
three xylene isomers, ethylbenzene and toluene are present in commercial mixed xylenes.
The purity required depends upon the intended use such as gasoline blending, solvents, or
isomer separation. The typical purities of p-, m-, and o-xylene (expressed in mole percent)
3-17
-------
when they are isolated as end-products are 99.1,98.5, and 96.0 percent, respectively
(U.S. EPA, 1986f).
Xylene can undergo chemical reactions involving the position of the methyl groups,
the methyl groups themselves, and the benzene ring (Ransley, 1984). Reactions involving
the position of methyl groups include isomerization, disproportionation, and dealkylation.
The methyl groups can be oxidized to phthalic acid (from o-xylene) or to terephthalic acid
or dimethyl terephthalate (from p-xylene).
3.3 SUMMARY
Gasoline is a complex mixture of hydrocarbons that distills within the range of 100
to 400°F. It is produced through the refining of crude oil. The composition of gasoline
varies in order to satisfy user performance requirements under varying weather conditions
or other measures of performance such as octane rating. Various additives and blending
agents are also incorporated into gasoline at the refinery stage to enhance performance and
stability.
Gasoline consists principally of paraffins (66 to 69 percent), aromatics (24 to 27
percent), and olefins (6 to 8 percent). Unleaded gasoline contains higher concentrations of
isoparaffins and aromatics than does leaded gasoline. Typical gasoline product lines may
contain more than ISO separate compounds.
Gasoline exists in the environment in four states: as a free moving liquid, adsorbed
into the soil, in groundwater, or in the air as an aerosol or vapor. Partitioning among these
media depends on specific properties of the individual compounds, such as vapor pressure,
water solubility, and soil/water or soil/gas partition coefficients. Gasoline vapors consist
mainly of short-chained paraffins and isoparaffins, while the aromatic compounds are the
most soluble in water.
3-18
-------
4. ECOSYSTEM CONSIDERATIONS
4.1 INTRODUCTION
Data are limited with regard to assessing the effects of a mixture as variable as
gasoline on a system as complex as an ecosystem. As described elsewhere in this
document, risk assessments of mixtures are complicated by several factors, including the
range of absorption and metabolic characteristics presented by the various components, and
interactive effects that may occur as the result of concurrent exposure to several toxic
substances. In an attempt to eliminate some of the variables present in the review of
mixtures, assessments of mixtures are usually based on study of a few components,
selected for their relative toxicity and concentration in the mixture. Therefore, the
following discussion of gasoline's toxicity is based primarily on studies of benzene,
toluene, and xylene. It should be noted that the resulting toxicity profile is not truly
representative of the entire mixture. However, benzene, toluene, and xylene are more
water soluble and have greater bioavailability than do possibly more toxic, but less soluble
gasoline constituents. This assessment should include the following steps.
1. description of the mixture, including identity and concentration of all components
and reaction products
2. description of the ecosystem, including identities of all potential environmental
receptors, and description of dynamic relationships among and between species
3. profile of each chemical's toxicology in each receptor species, including all
interactive effects associated with exposures to multiple toxic agents
4. identification of all potential changes in the ecosystem related to impacts of
contaminants
While the scheme outlined above indicates the need for a large volume of data, for
reasons of practicality, ecotoxicological risk assessments have most often been based on
single species laboratory tests and/or on uncontrolled field observation studies. In addition
to being relatively inexpensive, single-species laboratory studies present several
advantages. They have been standardized by various organizations, making both testing
protocols and interpretation of results fairly routine (Cairns, 1988). These studies are
based on single-species mammalian toxicity tests utilized in protecting human health, and
4-1
-------
they have relatively easily quantifiable endpoints (Cairns, 1988). Laboratory studies
normally focus on physiological or other effects observed in individual organisms.
Observed physiological responses include changes in respiration, immune reactions, nerve
function, membrane permeability, biosynthetic reactions (including synthesis of stress-
induced compounds), cell development and differentiation, gene expression and regulation,
and genetic information. Effects observed at the organism level include changes in growth,
development, orientation, mobility, behavior, reproduction, and survival.
These effects are of significance in describing the toxicity of a compound to
individuals or species, however, their pertinence to projections of impacts expected at
higher levels of biological organization is less clear. Important population characteristics
that are often inadequately accounted for in laboratory studies include abundance,
variability, distribution, biomass, and population dynamics. Ecosystem traits that are not
evident from laboratory studies include indices of diversity, succession, and aspects of
trophic structure, such as predator/prey relationships, parasitism, symbiosis, and
competition.
Field observation studies of environmental contamination, such as studies of
accidental spills, can fill some of the data gaps left by laboratory studies. However,
because pre-contamination baseline data are usually scant, interpretation of observation
studies and conclusions regarding changes in populations or communities drawn from this
type of data are highly subjective.
Pursuant to the general agreement that past approaches to ecotoxicological risk
assessments have been inadequate, several alternatives have been suggested. The simplest
of these operates on the premise that the lowest threshold effect observed in the most
sensitive species represents a level of contamination below which no adverse effects will be
expected in the ecosystem. While this tactic is probably protective of acute effects expected
in all species in the ecosystem, it cannot account for biocenotic effects. For example,
bioconcentration may not directly affect receptor organisms, but may have important
impacts on organisms at the higher trophic levels.
Researchers for the German Federal Ministry of Research and Technology
developed the concept of using early recognition signals to identify adverse environmental
effects. These signals, termed key indicators, are defined as "scientific measurement and
observation methods which respond to early stages of anthropogenic environmental
changes in either exposure or effects" (Schmidt-Bleek et al., 1988). Key indicators are
proposed for use as environmental monitoring tools. Only secondarily should these
indicators be used as a means of determining cause-effect relationships following the
observation of significant environmental changes. While the focus of this system is not
4-2
-------
ecotoxicological, risk assessment may suggest new approaches to this problem as well as
provide a substantial body of baseline data to assess environmental changes. Finally, it
may provide insights into the collection and cataloging of environmental data.
A third alternative proposed for the assessment of ecotoxicological risks is the use
of mesocosm testing. Mesocosm testing is suggested to be more environmentally realistic
than single species testing, because it depends on actual ecosystem dynamics rather than on
projected or modelled responses. For instance, as Cairns (1988) points out, ecosystem
components function differently when isolated from species interactions. Also, mesocosm
testing permits accurate accounting for chemical fate of contaminants, including
transformation, partitioning, and transport.
In the case of gasoline, the uptake and physiological responses of individual
organisms to some petroleum hydrocarbons are documented in laboratory studies. Data
reflecting long-term impacts of gasoline on communities are scarce (Jackson et al., 1989).
Available observation data suggests that impacts vary with such factors as the volume and
area of the spill, its chemical composition, its physical (as distinct from metabolic) effects
on organisms, the rate of weathering, and the nature of the location, habitat, and ecosystem
affected. In a study by Jackson et al. (1989) of the ecological impact of an oil spill, many
biological effects were observed that were unexpected and contradicted previous beliefs
regarding the toxicity and ecological impact of petroleum. These researchers suggest that
sublethal effects may have long term ecological implications, and that the damage caused by
petroleum hydrocarbons in the marine environment may lead to chain reactions that could
continue to occur long after the contaminants have dissipated.
The following ecotoxicity assessment is based on data collected from single species
laboratory studies of three components of gasoline, and on field observations from the
accidental gasoline spill of the Dona Marika.
4.2 AQUATIC VERTEBRATES
Information is somewhat limited on the effects of gasoline mixtures on plants and
aquatic wildlife. Studies of the results of accidental environmental releases, such as the
gasoline spill of the Dona Marika. provide limited on-site observations. However, because
gasoline is a mixture of components with various properties and partitioning affinities, the
effects of gasoline on the biota will be largely restricted to the bioavailable components.
Specifically, in aquatic systems, the water-soluble fraction (benzene and other aromatics) of
gasoline represents the highest concentration of bioavailable components. Cooper et al.
(1982) found that the water-soluble fraction of test jet fuels were less toxic to aquatic
species than were alkane components. However, Cooper et al. (1982) concluded that
4-3
-------
because the aromatic fraction, primarily benzene, toluene, and xylene, was more soluble in
water, aquatic animals will be exposed to greater dose levels resulting in comparable or
greater toxic effects.
Baker (1976) reported on observations from the gasoline spill of the Dona Marika.
No fish kills were verified, although mackerel were reported to have a gasoline-like flavor.
Observations of water birds revealed up to 100 dead gulls in a bay north of the spill; death
was possibly attributable to consumption of gasoline-contaminated limpets. However, no
dead gulls were found in the feeding area.
In static acute tests with marine fish, the 24- and 96-hour LCsos for benzene in this
fish were 6.9 and 5.8 ppm, respectively. The juvenile striped bass (Morone saxatilis)
revealed identical 24- and 96-hour LCsos. For toluene and meta-, ortho-, and para-
xylenes, LCsos were 7.3,9.2,11, and 2.0 ppm, respectively. Other acute tests are
reported for toluene, including a 96-hour LCso of 8 ppm at 12°C (Benville and Kom,
1977) in the pink salmon (Oncorhynchus kisutch), and a 96-hour LCso between 280 and
480 ppm in the sheepshead minnow (Cyprinodon variegatus) (Heitmiller et al., 1981).
Acute early life stage tests with eggs of Pacific herring (Clipea pallasi) showed severe
abnormalities and eventual death in hatched larvae exposed to 45 ppm benzene for 24 to 96
hours.
Sublethal effects of benzene at 30 to 35 ppm delayed development of larvae in 24-
to 48-hour exposures (Struhsaker et al., 1974). Tests with meta- and para-xylene resulted
in a 70 percent reduction in fertilization at 37 and 28 ppm, respectively, in eggs of the
Atlantic cod (Sadus morhua). At 20 to 30 ppm of the xylene isomers, 100 percent of the
eggs exposed exhibited abnormal cleavage; exposure to 16 to 35 ppm 6 hours after
fertilization resulted in 40 percent mortality of eggs, except with para-xylene, which caused
over 90 percent mortality during the first 17 days of development (Korseik et al., 1982). A
chronic value between 3.2 and 7.7 ppm is reported for the sheepshead minnow in tests
with toluene (U.S. EPA, 1978c).
For freshwater fish, static acute 24-hour LCsos of 18,16, and 13 ppm for meta-,
ortho-, and para-xylene, respectively, were reported by Bridie et al. (1979) in the goldfish
(Carrassius auratus). Brenniman et al. (1976) reports a 96-hour LCSO of 16.9 ppm xylene
in a flow-through test with C. auratus. The acute toxicity of toluene in a static 96-hour lest
revealed an LCso of 13 ppm in the bluegill sunfish (Buccafusco et al., 1981). In a 48-hour
flow-through test with the zebrafish (Brachydanio rerio), the LCso of toluene was 25 ppm
(Sloof, 1979). In flow-through tests with benzene, the acute LCso was 53 ppm in juvenile
rainbow trout (Salmo gairdneri); in a 96-hour exposure, 30 percent mortality was recorded
at 151 ppm benzene for fathead minnows (Pimephales promelas) (DeGraeve et al., 1980).
4-4
-------
Two amphibians, the Mexican axolote (Ambystoma mexicanum) and the clawed
toad (Xenopus laevis) were tested for sensitivity to benzene. The 48-hour LCsos,
respectively, were 370 and 190 ppm (Sloof and Baerselman, 1980).
4.3 AQUATIC INVERTEBRATES
Baker (1976) reported on studies performed immediately after the Dona Marika
accident and noted narcotized bivalves and large numbers of dead sea urchins
(Echinocardium). There was evidence of limpet (Patella sp.) detachment and retraction of
winkles (Littorina sp.), top shells (Gibbula sp.), and dog whelks (Nucella lapillus) into
their shells. Mussels (Mytilus edulis) were gaping. Within a day after washing, at least 50
percent of each species had apparently recovered. Five transects that formed part of the
Milford Haven monitoring program were surveyed before, immediately after, and a year
after the Dona Marika mishap. The two transects closest to the spill (within a kilometer)
showed a larger reduction in limpet numbers shortly after the spill. Limpet numbers
returned to normal the following year, although the limpets were very small, probably due
to recolonization from the larval stage. The numbers of gastropod mollusks, such as dog
whelks (N. lapillus), fell off after the spill but recovered in a year; recolonization was
largely by adult animals. Baker (1976) suspected that animals that retracted under the
influence of pollutants were washed into the sublittoral zone and eventually crawled back
into the littoral area. The population of barnacles (Balanus balanoides) seemed to be
unaffected by the spill.
Burnett and Kontogiannis (1975) studied the effect of gasoline (lead content
unspecified) on the survival of a tidepool copepod, Tigriopus calif ornicus. In the 8-day
exposure period, 85,45, and 25 percent of the T. californicus were killed within the first
24 hours at levels of 1.0,0.50, and 0.25 ml/L, respectively. After the first day, the effect
of the gasoline was lessened. The evaporation rate of gasoline was measured at the test
temperature of the study, demonstrating a half-life of 2 hours for a pure sample. Thus, the
rapid evaporation may explain why codepod deaths leveled out after the first day of
exposure.
Berry and Brammer (1977) investigated the effect of the water-soluble components
of gasoline (regular gasoline) on the fourth-instar larvae of the mosquito, Aedes aegypti.
The solutions were prepared by adding 1 to 100 ml gasoline to 1 L distilled water. The
mortality of the larvae exposed to the water-soluble fraction for 24 hours in a static
bioassay reached 9 percent from a 20 ml gasoline/liter water mixture. The same percentage
of mortality was observed at higher levels of the gasoline mixtures. The authors reported
that as the amount of gasoline per liter water was increased from 1 to 100 ml, the total
4-5
-------
water-soluble aromatic content only increased from 6.94 ppm to 8.39 ppm. In addition, no
detectable levels of any water-soluble component remained after 24 hours.
LeBlanc (1980) reports static acute 24- and 48-hour LCsos of 250 and 200 ppm for
freshwater zooplankton Daphnia magna in exposures to benzene. For toluene exposures,
the 24- and 48-hour LCsos were both 310 ppm, and the no observed effect level (NOEL)
was 28 ppm. The NOEL for benzene was less than 13 ppm. In a static 24-hour
immobilization study with xylene, 150 ppm of xylene caused a 50 percent reduction in
photosynthesis (ECso) in Daphnia magna (Bringmann and Kuhn, 1977). Hermens et al.
(1984) reports a 48-hour ECso for Daphnia magna of 14.3 ppm based on measured
concentrations of meta-xylene.
Marine invertebrates tested showed greater sensitivities to all three components.
Benville and Kom (1977) exposed the bay shrimp Cragofranciscorum to m-, p-, and o-
xylene in static tests at 16°C and 25 ppt salinity. The 24-hour LCsos were 4.8,5.3, and
2.0 ppm, and the 96-hour LCsos were 3.7, 1.3, and 2.0 ppm for the xylene isomers,
respectively. The 24-hour LCso for benzene was 22 ppm, and the 96-hour LCso was 20
ppm. Acute tests of the effects of toluene in bay shrimp revealed a 24-hour LCso of 12
ppm and a 96-hour LCso of 4.3 ppm. Exposure of brine shrimp nauplii (Anemia salina) to
toluene revealed a static 96-hour LCso of 33 ppm (Price et al., 1974). Neff et al. (1976)
reports a static 96-hour LCso of 7.4 ppm in exposures of the grass shrimp Palaemonetes
pugio to xylene. Freshwater mussels Dreissend polymorpha were able to detect 16.2 ppm
of xylene and responded by shutting their shells (Sloof et al., 1983). An upper subleihal
dose of xylene was 7.9 ppm for aquatic animals (Berry et al., 1978). Caldwell et al.
(1977) reports xylene sensitivity for the Dungeness crab (Cancer magister) to be similar to
that of bay shrimps. Ranges for the 96-hour LCso were 4 to *2,1 to 6, and 2 ppm for m-,
o-, and p-isomers, respectively.
4.4 AQUATIC FLORA AND MICROORGANISMS
The toxic effects of gasoline components on primary producers and
microorganisms is of considerable importance because of the fundamental role they play in
the food chain, decomposition processes, and nutrient recycling.
In cell multiplication inhibition tests with benzene, toxic thresholds of 92 ppm for
bacterium Pseudomonas putida, and thresholds of more than 1,400 and 700 ppm,
respectively, for green algae Scenedesmus quadricauda and protozoan Entosiphon sulcatum
were reported by Bringmann and Kuhn (1980). In earlier tests by Bringmann and Kuhn
(1978), the inhibition of cellular propagation in blue algae (Microcystis aeruginosa) and 5.
quadricauda resulted from exposure to more than 1,400 ppm benzene in 8-day exposure
4-6
-------
periods. Kauss and Hutchinson (1975) exposed the freshwater algae Chlorella vulgaris to
toluene, revealing an EC$Q of 245 ppm. The 96-hour ECso to toluene (based on in vivo
chlorophyll measurements) was 433 ppm for Selenastrum capricornunon, another
freshwater species (U.S. EPA, 1978c). The marine algae Skeletonema costatum showed
greater sensitivity to toluene, demonstrating inhibition of growth at 8 ppm (Dunstan et al.,
1975). In 96-hour tests, algal species S. costatum, Criscosphaera cartara, and
Amphidinium cartarae showed inhibition at 100 ppm (Dunstan et al., 1975). In xylene
exposures of C. vulgaris, Kauss and Hutchinson (1975) report a 24-hour EC50 of 55
ppm. In 96-hour exposures to para-xylene, a 50 percent reduction in photosynthesis was
reported at 104 ppm for C. vulgaris. Clamydomonas angulosa showed a 50 percent
reduction in photosynthesis at 45 ppm to para-xylene (Hutchinson et al., 1980).
4.5 BIOCONCENTRATIQN
The ability of a chemical to bioconcentrate is determined by the concentration of the
chemical in the organism relative to the chemical's concentration in the water
(bioconcentration factor, BCF). Thus, a chemical with a high BCF can potentially
accumulate in an organism at much greater concentrations than those in the surrounding
water. Lyman et al. (1982) cites numerous studies concluding that water is far more
significant than food as a source of organic compounds in fish bioconcentration.
Furthermore, bioconcentrated chemicals in the biota may result in further increases in
concentrations as a result of movement through the food chain via successive trophic levels
(i.e., biomagnification). From an environmental standpoint, these phenomena become
important when acute toxicity of an agent is low and the physiological effects remain
unnoticed until chronic effects become evident (Verschueren, 1983). The potential of a
chemical to bioconcentrate is related to its affinity to partition to organic matter, to its
solubility and environmental persistence, and to variability in target species affected.
No information was found in the available literature on the bioconcentration of
gasoline. This is not unusual, considering the broad variability in properties among the
numerous components of gasoline. Once gasoline is released into the environment, the
components will partition at various rates into different environmental compartments.
Although data are absent for gasoline, the potential BCFs for representative gasoline
components can be determined by using log octanol-water partition coefficients (log KQW)
in the equation of Veith et al. (1979, as cited in Lyman et al., 1982). The higher the log
KQW for a compound, the greater its potential to bioconcentrate. Table 4-1 lists the
representative components of gasoline, their log octanol-water partition coefficients, and
potential BCFs.
4-7
-------
TABLE 4-1
THE OCTANOL-WATER PARTITION COEFFICIENTS FOR MAJOR
COMPONENTS IN GASOLINE
Representative
gasoline component
Potential to
L°§ KOW bioconcentrate (BCF)
Source
Alkanes
n-pentane
n-hexane
2-methyl pentane
2-methyl hexane
2-methyl octane
3-methyl pentane
3-methyl hexane
3-methyl octane
Alkenes
proplyene
pentene
Aromatics
benzene
toluene
ethylbenzene
m-xylene
3.62
4.11
2.77
3.30
4.38
2.88
3.41
4.46
0.65
1.73
2.13
2.65
3.13
3.20
332
783
75
190
1,255
91
230
1,444
1.9
12
24
61
141
159
Miller etal., 1985.
Miller etal., 1985.
Coates et al., 1985.
Coates et al., 1985.
Coates et al., 1985.
Coates et al., 1985.
Coates et al., 1985.
Coates et al., 1985.
Coates et al., 1985.
Coates et al., 1985.
Miller etal., 1985.
Miller etal., 1985.
Miller et al., 1985.
Miller etal., 1985.
4-8
-------
With the exception of the methyl octanes and possibly hexane, these gasoline
components exhibited minimal potential to bioconcentrate. The only measured BCF found
for gasoline components was 150 for methylcyclohexane in a 42-day exposure with
rainbow trout (S. gairdneri) (Verschueren, 1983). A study on the accumulation of xylene
is also reported. Japanese eels (Anquillajaponicd) were exposed by 50 mg/L of crude oil
for 10 days. Equilibrium for xylene accumulation was reached in 5 days.
Bioconcentration levels in eels ranged from 21.4 to 23.6 times the concentration in the
water for the three xylene isomers. Return to clean water showed half-lives of 2 to 2.6
days (Ogata and Miyake, 1979). Compounds with BCFs over 1,000 may pose a
considerable hazard to the biota. Potential effects are dependent upon persistence, volumes
released, affected biota, and water quality parameters. Observed BCFs have often proven
to be much greater than estimated BCFs, but variables may affect BCFs in either direction
(Lyman et al., 1982). The log Kows and estimated BCFs for most gasoline components
indicate that bioconcentration in aquatic animals is not expected to be a significant
environmental factor.
4.6 BIODERRADATION
Biodegradation of gasoline and its various components has been studied to some
extent. Horowitz and Atlas (1977) found that after an accidental spill of 55,000 gallons of
leaded gasoline into an arctic freshwater lake, microorganisms could convert the
hydrocarbons to CO2, but nutrient additives and bacterial inoculation were needed to
enhance biodegradation. Gasoline hydrocarbons persisted in the lake for months after the
spill. Jamison et al. (1975) observed many hydrocarbon-utilizing microbial species after a
gasoline spill into groundwater in a temperate zone.
A study of refinery effluents conducted by the U.S. Department of Energy and the
U.S. EPA showed that components of gasoline (n-alkanes, isoalkanes, cycloalkanes,
alkenes, and alkylated benzenes) were degraded in activated sludge systems.
Concentrations of these compounds (upper limit was 544 ng/g; most values ranged from 20
to 100 ng/g) after dissolved air flotation treatment were 99 percent degraded after activated
sludge treatment (Snider and Manning, 1982).
The Food and Agriculture Organization of the United Nations (FAO, 1977)
reviewed literature on microorganisms that were believed to be capable of degrading
petroleum, including gasoline, in the marine environment and found over 90 species
including open ocean, coastal, and estuarine bacteria, fungi, and yeasts. However, FAO
noted that microbial action does not become important until a week or so after
environmental entry. After this time period, gasoline has probably undergone
4-9
-------
photooxidation or evaporation. FAO also mentioned that alkanes are attacked by a greater
number of microbial species at a more rapid rate than are aromatics.
4.7 SUMMARY
Due to limited data on gasoline as a mixture, ecosystem considerations are based
primarily on the components benzene, toluene, and xylene. The status of research on the
ecological import of these components is also limited. However, from the studies and
reports available, it appears that action on biota is restricted to bioavailable components.
Water soluble fractions such as benzene and other aromatics are most bioavailable. Some
bioaccumulation in the food chain are also suggested from studies of the Dona Marika
gasoline spill. Effects have been observed in vertebrates in both field studies and
laboratory investigations. This included effects on mortality rates, fertilization, and growth
development. Similar responses were seen in invertebrates and aquatic flora and
microorganisms. Limited data on bioconcentration indicates that there is not a strong
tendency of gasoline or its components to bioaccumulate. Biodegredation is dependent on
the presence of other nutrient sources and ecological conditions, and varies between
components. The overall set of data available for ecological evaluation is too limited for
conclusive decisions to be made.
4-10
-------
5. EXPOSURE ASSESSMENT
The major sources of exposure to gasoline and gasoline vapor are from service
station operations and as a result of gasoline leaking from underground storage tanks. The
principal exposure pathways from these sources are vapors transported in ambient air from
service station operations and indoor air during use of contaminated groundwater, gasoline
migration into basements of homes, and the ingestion of gasoline-contaminated
groundwater. Six specific exposure scenarios based on these major exposure pathways
will be assessed in this chapter.
5.1 SOURCES AND FATE OF GASOLINE AND GASOLINE VAPORS
Emissions of gasoline vapors to the atmosphere occur everywhere along the chain
of fuel handling and marketing, beginning at the refinery and continuing through bulk
loading, transport, and unloading operations down to local service stations, where
refueling of individual vehicles occurs. Gasoline vapors are also released into the air from
the fuel tank, carburation system, and tailpipe of vehicles. At each step in the fuel
handling, vapors released into the ambient air are subject to the processes of transport,
dilution, and dispersion. These processes spread the contaminants over wide areas while
reducing their concentration in the ambient air. Due to the differences in the partial pressure
of various hydrocarbons, gasoline vapors consist of relatively more of the lighter
compounds (e.g., alkanes) and less of the heavier ones (e.g., branched alkanes) than liquid
fuel.
The populations that receive the greatest exposure in this chain of fuel handling are
refinery workers, bulk fuel truck drivers, service station attendants, self-service customers,
and residents of neighborhoods close to refineries, bulk storage terminals, and service
stations. This assessment estimates exposure of service station attendants, self-service
customers, and nearby residents to gasoline vapor emissions associated with service station
operations.
Leakage of gasoline from USTs and associated piping is a major cause of
groundwater contamination in the United States. Domestic use of gasoline-contaminated
groundwater is a major exposure pathway. In addition, residents living near LUSTs can be
exposed to gasoline vapors through vapor migration below ground into their homes.
The migration of substances leaking from USTs depends on the quantity of
gasoline released, the physical properties of that particular fuel, and the structure of the
subsurface soils, rock, and water table. The fate and transport of gasoline underground is
a very complex process because each gasoline component has a unique set of chemical and
5-1
-------
physical properties that determines its phase fractionation and flow characteristics.
Gasoline exists in soil and groundwater in four states: as a free-moving liquid, adsorbed to
soil, as a solute in the groundwater, and as vapor. Gasoline vapor, as well as the liquid
phase, is subject to dispersal and migration as precipitation moves through the
contaminated soil. Gasoline leaked into the soil volatilizes because of its high vapor
pressure, filling soil grain pore spaces with vapors. The liquid gasoline forms a moving
mass in the soil where it settles by gravity to the water table and floats on top of it.
Concentrations in the groundwater will be affected by dilution and dispersion of the soluble
components. Aromatic hydrocarbons, such as benzene, toluene, and xylene, are highly
volatile and soluble and are therefore commonly found both in the vapor phase and in
gasoline- contaminated groundwater.
5.2 EXPOSURE ASSESSMENT SCENARIOS
Six scenarios have been chosen to represent the major sources of exposure to
gasoline vapors. Scenarios 1 through 3 estimate exposure from emissions at a service
station that occur during automobile refueling.
1. A self-service customer is exposed to gasoline vapors while in a service station
and is exposed at a higher level while pumping gas into his/her vehicle.
2. A service station attendant is exposed to gasoline vapors from a variety of
activities during a normal workday at the station.
3. A resident living near a service station is exposed to vapors whenever he/she is
downwind of the station.
By contrast, scenarios 4 through 6 estimate human exposure that can occur when an
underground storage tank at a service station is leaking gasoline.
4. A resident living near a service station with a leaking underground storage tank
is exposed to vapors that have diffused through the soil and entered the lowest
level of the house.
5. A resident whose domestic water supply is contaminated by gasoline ingests
gasoline-contaminated drinking water.
5-2
-------
6. A resident whose domestic water supply is contaminated by gasoline is exposed
to vapors in his/her home released during non-ingestion household use of water
(e.g., showering).
5.3 EXPOSURE ASSESSMENT METHODOLOGY
A number of philosophies guide the assessment of exposure to gasoline and
gasoline vapors. First, because gasoline is a complex mixture of hundreds of
hydrocarbons, benzene, toluene, xylene (the sum of p-, m-, and o-), and total
hydrocarbons (THC)1 have been selected as surrogate compounds for the gasoline
mixture. The basis for selecting these compounds is presented in Chapter 2. This chapter
will present the ambient concentrations to which humans are exposed under the six
exposure scenarios. The lifetime exposure dose, in mg/kg/day, of benzene, toluene,
xylenes and total hydrocarbons for each scenario are presented at the end of this chapter.
As U.S. EPA has recommended in their Guidelines for Exposure Assessment
(U.S. EPA, 1987a), assessments should ideally be based on measured data, or if not, then
on simple, generally-accepted models. This assessment follows that approach in using
measurement surveys as the primary source of data. By design, this assessment presents a
range of exposures from "average concentration" to "upper limit" (or maximum
concentrations measured). Exposure assessment prudently involves the use of upper limit,
health-conservative assumptions because ambient air emissions and the transport and
dispersion of these emissions in environmental media are subject to large uncertainties. A
recent study (Travis, 1987) of federal agency decisions-making on human exposure to
contaminants shows that such decisions are routinely made on the basis of upper-limit
estimates. This study provides, in addition to the upper bound exposure estimate, a value
for the average concentration or mean exposure. Although some references in the literature
on gasoline vapor exposure use measures, including a geometric mean, median, or 90th
percentile value, to represent the expected value for any one exposure event, a simple
arithmetic mean (referred to as the "average concentration") is used here instead, since it is
the measure of a cumulative, lifetime dose. Thus, calculating both the average
concentration and upper-limit exposures not only provides a range of possible health risks,
but also indicates the uncertainty of the analysis.
In selecting data that are representative, preference is given to monitoring data from
locations in the Northeast states, under meteorological conditions typical of those found in
1 Monitoring data indicate that propane, isobutane, n-butane, isopentane, and
n-pentane comprise a consistent set of the five most volatile hydrocarbon
components of gasoline.
5-3
-------
the Northeast, and for unleaded gasoline only. Since the results of this study will be used
to guide public policy in future years when fully leaded gasoline will not be available, only
tests involving unleaded fuel are considered in this analysis.
The goal of this assessment is to estimate typical and maximum gasoline vapor
exposure levels. In scenarios 1 through 3, the situations are relatively well defined and
many researchers have conducted both measurement surveys and modeling studies of
gasoline vapor exposure.
Scenarios 4 through 6 are different since the range of possible exposures is very
large and varies on a case-by-case basis. Thus, the assessment for these scenarios only
gives a "case-study" illustration of the potential human exposures that have occurred in
certain homes.
An extensive body of literature on ambient air monitoring at or near service stations
was reviewed to identify data applicable to scenarios 1 through 3. The following
discussion presents the criteria for selecting the monitoring studies used to estimate
exposures of self-service customers, service station attendants, and residents living near a
typical service station. The literature reviewed for service station exposures is discussed in
detail in Appendix B and summarized in this chapter.
5.3.1 Criteria for Selecting Monitoring Studies
Numerous factors affect the quality of reported data and authors often do not give a
full recitation of the facts surrounding a monitoring program. By examining a large
number of studies and presenting a range of exposures, the effects of these errors can be
minimized. However, it is not possible to completely screen out faulty or unrepresentative
data for this study.
The criteria which were considered in evaluating the literature are:
1. Applicability. Is a given measurement representative of the exposure a person
would encounter in one of the scenarios, or did atypical conditions prevail during
the monitoring?
2. Type of Sampling. Was a fixed monitor or a personal sampler used? Were
sample collection and analytical techniques appropriate?
3. Quality Assurance. Was an audit of the sampling method and laboratory
analysis performed using a sample blank and a known calibration gas? Were
5-4
-------
estimates made of the accuracy (bias) of the monitoring systems? Were
interferences checked and corrections made?
4. Distance from Source. Air concentrations near a source typically vary by
several orders of magnitude within a relatively short radial distance. Is the
distance from the source clearly stated for each measurement?
5. Meteorology. Conditions at the time of monitoring, including wind speed,
direction, temperature, and turbulence, affect measured values. Are these
conditions described?
Over 15 studies obtained from the literature review were used in constructing
scenarios 1 through 3, though in most cases only a subset of the data in each study, judged
to be representative of the scenarios in the Northeast states, were used.
5.3.2 Monitoring Methods Used in the Studies
Organic vapors in the air are generally collected using either whole air sampling or
solid adsorbent methods. Whole air samples are simply captured in a containment vessel
such as a stainless steel canister or Tedlar bag. Whole air sampling is well suited to short-
term exposure studies but, due to the bulkiness of the equipment, sampling must be
performed at fixed locations.
Solid adsorbent sampling uses media, such as TenaxR or charcoal, that have a
strong affinity for organic vapors to collect and concentrate these compounds. Elution of
the sample is achieved by either heat desorption or liquid extraction (e.g., carbon
disulfide). Collection and desorption efficiencies vary depending on the ambient
concentration levels, types of adsorbent used, adsorbent mass, air flow rate, and handling.
Also, chemical desorbants can sometimes be contaminated with the various compounds
being collected. Thus, strict quality assurance and audit procedures are required. Without
such controls, solid adsorbent sampling can substantially underestimate actual
concentrations.
The number of different sampling techniques for organic vapors and the many
stages in the data collection process make it impossible to ascribe a typical error to
hydrocarbon measurements in general. One source of variability is the estimate of air flow
rates. Ambient concentrations are determined by dividing an estimate of the detected
pollutant mass by an estimate of the air flow rate. Due to non-steady state operation of air
pumps and changes in back-pressure from filter media over a sampling period, estimates of
5-5
-------
flow rates are often no better than ±10 percent (Intersociety Committee, 1989) and possibly
worse.
5.4 ESTIMATING VAPOR EXPOSURE AT AND NEAR SERVICE
STATIONS (Scenarios 1-3)
At service stations, the sources of gasoline vapors include breathing and working
losses from underground storage tanks that are vented to the atmosphere, displacement
-------
following: a single refueling operation involves a mean volume of 10 gallons and an upper
limit volume of 20 gallons; annual refueling for a single vehicle involves a mean volume of
700 gallons and an upper limit volume of twice the mean or 1,400 gallons; a typical self-
service pumping rate is 8 to 10 gallons/minute.
Table 5-1 presents each monitoring study reviewed to estimate exposure of self-
service customers and service station attendants, the citations, the type of sampling
conducted, the number of measurement samples collected that are judged to be
representative of the listed scenarios in the Northeast states, and the arithmetic mean and
upper limit hydrocarbon concentrations for the selected samples. In many of the studies,
the measured exposure levels are thought to be underestimated for a variety of reasons;
these are listed in the notes section of the table. Note that the data in Table 5-1 are a small,
select subset of all the samples found in the literature.
5.4.1 Scenario 1 • Exposure of Self-Service Customers
The inhalation exposure to gasoline vapors for a self-service customer or service
station attendant is composed of two factors: the refueling operation and other time spent at
the service station. (Please note that in Table 5-1, the refueling studies are the first four
studies listed followed, by five occupational exposure studies.) To calculate the ambient air
concentrations from refueling operations for a self-service customer, the measurement data
presented in Table 5-1 for self-service refueling exposure (Scenario 1) are averaged
together to form an ensemble arithmetic mean exposure for THC, benzene, toluene, and
xylene (i.e., the study means are weighted by the number of samples).
For refueling exposure of the self-service customer, data from two of the studies,
Ellis and Obendorfer (1984) and Clayton (1983), are considered to be of lower quality and
more likely to be underestimated. For that reason, these studies were only weighted half as
much as the other two refueling exposure studies.
Based on the foregoing studies, the exposure for scenario 1 assumes the following:
• Air concentrations during the refueling operation as given in Table 5-2.
• Air concentrations during the other time spent in the station equal to an average
of 2 percent and an upper bound of 10 percent of the refueling operation levels
(refer to study by Bond et al. [1986a] in Appendix B).
• Annual refueling times of 77 to 155 minutes per year.
• Average time of a station visit of 5 minutes and an average of 70 visits per year.
• Average adult body weight of 70 kg (Diem and Lentner, 1973) and a light-
activity total respiration rate of 1.2 m3/hr (Snyder et al., 1975), and an alveolar
ventilation rate of 0.8
5-7
-------
TABLE 5-1
GASOLINE VAPOR MEASUREMENT STUDIES CONSIDERED IN SCENARIOS 1 AND 2
(All concentrations given in ppmv)
Sponsor
Philadelphia Air
Mans sment Services
General Motors
U.S. EPA
American Petroleum
Institute
Ol
oo American Petroleum
Institute
UofWashlr Ion
Mobil Oil
Shell Oil
Amoco
Citations
Ellis and Obendorfer, 1984.
Tironi et al.. 1986.
Bond et al. 1986.
Clayton, 1983.
Rappaport et al., 1987.
Hartle. 1980
Kearne and Dunham. 1986.
McDermott and Vos 1979.
Haider et al.. 1986.
Sample Number ol
Type samples
THC Benzene Toluene X lene Notes
Mean UL Mean UL Mean
VWR
W/R 1
W/R
4
9
5
SA/R 35
SA/O 33
SA/O 2
SA/O 1
SA/O 2
SA/O 2
1
4
a
1
0.164 0.612 - -
207.8 682.8 1.1 2.9 0.7
23.4 32.1 0.23 0.33 0.19
86 260 0.92 4.2 0.59
33 - - 0.273 - - 0.269
- - - - 0.122 0.237 - -
1.5 14.3
13.5 - - 0.128 0.840 - -
0.3 1.3 - -
UL Mean UL
1, 2, 3
2 0.1 0.2
0.26 0.13 0.19
2.4 0.32 1.1 4. 5. 6
7
2. 3, 6.
- - 0.132 - - 78
7
2. 9
2,6, 7,
10
2, 3. 6,
7811
-• = no data
Sample Types: W=Whole Air. SA=Solid Adsorbent, R=Relueling Exposure, O=Occupalional Exposure
1=Sample loss 7=lnadequale quality assurance
2=Meteorological data missing 8=No information on geography/site factors
3=No data on gasoline grades in use 9=Very high detection limit
4=No information on distance Irom vehicle filler pipe 10= Includes the use of leaded gasoline
5=An independent analysis shows THC underestimated 11=Assumed zero exposure during part of workday underestimates results.
6=lnsufficient information on sampling and analysis
-------
• Average adult refuels his car for 55 years out of 70 year lifetime (API, 1988).
5.4.2 Scenario 2 • Exposure of Service Station Attendants
To calculate ambient air concentrations for occupational exposure, the measurement
data from occupational exposure studies listed in Table 5-1 are averaged together to form an
ensemble arithmetic mean exposure for THC, benzene, toluene, and xylene.
The inhalation exposure to gasoline vapors for a service station attendant employs
these assumptions:
• Air concentrations during attendant's working hours as given in Table 5-2.
• An average workday of 8 hours, 5 days/week, 50 weeks/year, 50 years/lifetime
(API, 1988).
• An average adult has a body weight of 70 kg, a total respiration rate of 1.2
m^/hr, and an alveolar respiration rate of 0.8 m^/hr.
5.4.3 Scenario 3 - Exposure of Residents in Neighborhoods Near
Service Stations
For scenario 3, exposure of residents near a service station, a number of studies
sponsored by the U.S. EPA (Braddock et al.. 1986; Braddock, 1988; U.S. EPA, 1986a),
General Motors (Furey and Nagel, 1986), and Shell Oil (McDermott and Killiany, 1978)
provide estimates of mass emissions rates from refueling operations and underground
storage tank losses. A full discussion of this literature is given in Appendix B.
Inhalation exposure to gasoline vapors in a residential neighborhood near a service
station is determined by many factors, one of which is the distance downwind a person
lives from the station. Obviously, those living closer have higher exposure; however, the
size of the exposed population would be expected to increase at distances farther from the
station. Therefore, a range of distances needs to be carefully examined. In an urban area,
the closest a house could reasonably be to the emission release points at an adjoining
service station (i.e., gas pumps) is 30 meters (100 feet). Therefore, exposure values are
calculated for distances of 30, 50,100, and 200 meters downwind.
The U.S. EPA Industrial Source Complex Long-Term (ISCLT) dispersion model is
the recommended refined model for annual average air pollutant concentrations in both rural
and urban areas. ISCLT is a steady-state Gaussian plume model that uses a statistical
frequency distribution of 576 combinations of wind speed, wind direction, and
atmospheric stability to calculate long-term average air concentrations at an array of ground-
level receptors. Since service stations are usually found in urban areas, the urban
5-9
-------
TABLE 5-2
ESTIMATED EXPOSURE LEVELS FOR SCENARIOS 1 AND 2
(ppmv)
Hydrocarbon
Exposure
THC
Mean
Upper Limit 1
Benzene
Mean
Upper Limit
Toluene
Mean
Upper Limit
Xylene
Mean
Upper Limit
Scenario 1
Refueling
134.0
682.8
0.9
4.2
0.6
2.4
0.2
1.1
Scenario 2
Occupational
19.8
NO*
0.2
1.3
0.3
ND
0.1
ND
Estimated weighted average values reported in the refueling studies listed for Scenario 1
in Table 5-1.
Upper limit values are not reported in cases where ranges were not reported in the study.
5-10
-------
dispersion rates are used in this analysis. ISCLT is capable of representing a wide array of
source types, including the non-buoyant gasoline vapor emissions from service stations.
Long-term average exposures at these distances are determined in part by
meteorological conditions. To accurately represent this factor, the ISCLT dispersion model
is run using a 5-year climatology of wind speed, wind direction, and stability class as
measured in Boston, Massachusetts. Stability-wind conditions for Boston are considered
representative of the Northeast states for the purpose of calculating annual average pollutant
concentrations. The stability of long-term model predictions is such that use of
climatological data from a different station in the Northeast would not substantially alter the
results.
The bulk of the emissions released by a service station are from refueling operations
and these occur at the side of the motor vehicle being refueled (see Table C-l, Appendix
C). As the wind flows over and around a parked vehicle, a mixing cell is set up with
characteristic dimensions equal to those of the vehicle (about 1.5 meters on a side). The
refueling emissions are mixed into this volume of air before being transported off site to
downwind receptors. Thus, for modeling at distances of 30 meters or more, the service
station emissions were represented as a volume source in the ISCLT model with a
characteristic dimension of 1.5 meters.
The following assumptions were employed in scenario 3 and the predicted exposure
values are summarized in Table 5-3.
• Air concentrations based on mass emission rates (g/gal) for service station
activities (see Appendix C).
• An average service station pumps 1,000,000 gallons/year (U.S. EPA, 1984d;
Kearney and Dunham, 1986).
• Pollutant transport, dilution, and dispersion is described by the ISCLT model.
Mean exposure is the arithmetic mean of 144 receptors located 30 and 200
meters downwind. Upper limit exposure is the maximum modeled average
concentration corresponding to a distance of 30 meters.
• People are home 24 hours/day.
• Indoor concentrations are equal to outdoor concentrations.
• Average adult body weight of 70 kg, a total respiration rate of 1.2 m^/hr, and
an alveolar respiration rate of 0.8 m^/hr.
The results of the ISCLT dispersion model are given in Appendix C.
5-11
-------
TABLE 5-3
ESTIMATED EXPOSURE LEVELS FOR SCENARIO 31
(ppbv)
Hydrocarbon Scenario 3
Exposure Residential Neighborhood
THC
Mean 5.4
Upper Limit 28.0
Benzene
Mean 0.04
Upper Limit 0.16
Toluene
Mean 0.08
Upper Limit 0.37
Xylene
Mean 0.03
Upper Limit 0.15
Refer to Appendix C.
5-12
-------
5.5 ESTIMATING EXPOSURE FROM LEAKING UNDERGROUND
TANKS (Scenarios 4-6)
Several factors have to be taken into account when estimating vapor exposures from
gasoline leaking from underground storage tanks. These include the fuel composition,
volume of fuel spilled, vertical distance to groundwater, horizontal distance and slope from
the underground spill to the house, soil structure and porosity, house foundation
construction, hydrology and meteorology, and air exchange rate in the house.
5.5.1 Scenario 4 - Exposure of Residents Near Leaking Underground Tanks
The air inside a house is typically under negative pressure relative to the outside, a
condition which pulls soil gases into the house through cracks and holes in the foundation.
Because of the variability of each case and uncertainty of modeling such a case for this
scenario, the exposure estimate in this assessment is based on actual measured
concentrations of contaminants in the indoor air of homes and offices where gasoline
vapors have migrated from a LUST.
Two sets of case-study measurements have been obtained. In the first, benzene,
toluene, and xylene concentrations inside six places of business (still in operation) near a
leaking underground gasoline storage tank in New England were reported in confidence to
NESCAUM by Groundwater Technology (1988). The data were collected using solid
adsorbent samples, NIOSH methods and proper quality assurance procedures were
followed.
The mean and upper limit values (in ppmv) for 20 samples, collected by
Groundwater Technology (1988) each collected over a 200 minute period, are:
Upper Limit
No Data
2.9
7.6
4.0
The second study involved air concentrations inside two homes downgradient of a
large UST leak of gasoline in North Babylon, New York (Suffolk County DHS, 1989).
The data were collected by the New York State Department of Transportation using solid
adsorbent samples and NIOSH methods. Sampling and quality assurance procedures were
based on protocols established by the Department of Transportation. Although several
5-13
THC
Benzene
Toluene
Xylene
Mean
No Data
0.55
0.80
0.47
-------
aromatic compounds were reportedly measured, only results for benzene are given. The
mean and upper limit values (in ppmv) for twenty 24-hour average samples are:
Mean Upper Limit
Benzene 0.02 0.20
The inhalation exposure to gasoline vapors that enter a home from the soil
contacting its lowest level is based on the case study with the highest measured levels. The
following assumptions were employed in scenario 4:
• Air concentrations in the home equal to the mean and upper limit values from
Groundwater Technology (1988).
• People are home 24 hours/day.
• Average body weight of 70 kg, a respiration rate of 1.2 m^/hr, and an alveolar
ventilation rate of 0.8 m^/hr.
5.5.2 Scenario 5 - Exposure to Gasoline-Contaminated Water
Benzene, toluene, and xylene are the predominant water-soluble components of
gasoline (ICF, 1987). Methyl butyl ten-ether is also a major component detected in
gasoline contaminated groundwater. However, insufficient data are available to include
this compound in the assessment. Exposure estimates to gasoline contaminated water used
as a source of drinking water (scenario 5) or as a domestic water supply for non-ingestion
household uses (scenario 6), such as showering or bathing, also rely on monitoring data
from case studies in five New England states in which a leaking underground storage tank
has contaminated a domestic water supply.
Hydrocarbon concentrations in groundwater downgradient of a leaking
underground gasoline storage tank were reported in confidence to NESCAUM (IEP, 1989)
for six such sites in Massachusetts. Monitoring wells located along the plume centerline
were at distances from 20 to 280 feet downgradient in a variety of soil and geologic
conditions. Proper quality assurance procedures were reportedly followed. The mean and
range of values (in ppb or ug/L) for 34 samples are:
5-14
-------
Detection Upper
Mean Limit Limit
Benzene 4,700 7 84,000
Toluene 8,400 4 78,000
Xylene 3,500 2 24,000
THC No Data No Data No Data
A LUST at a service station in Hardwick, Vermont, prompted a study of
groundwater contamination by the Vermont Department of Water Supply (1987). Two
monitoring wells sunk within 50 feet of the leaking tank recorded high levels of
hydrocarbon contamination. The mean and range of values (in ppb or ug/L) for five
samples are:
Benzene
Toluene
Xylene
THC
Mean
14,200
30,400
15,300
No Data
Detection
Limit
18
18
31
No Data
Upper
Limit
22,000
42,000
22,700
No Data
At the sites of underground gasoline spills in Plainfield, Madison and Bethlehem,
Connecticut, test measurements at nine nearby locations where gasoline components are
clearly identifiable in well water were compiled (CT DEP, 1989). The mean and range of
values (in ppb or ug/L) are:
Benzene
Toluene
Xylene
No. of
Samples
9
4
4
Mean
263
143
265
Detection Upper
Limit Limit
4 2,200
<1 490
<1 1,000
THC No Data No Data No Data No Data
The State of Maine relies heavily on groundwater wells for its public water supply.
A report by the Maine DHHS (1986) reports on THC concentrations in 109 gasoline and
oil contaminated wells, as measured in 1985. The values (in ppb or ug/L) are:
5-15
-------
No. of Detection Upper
Samples Mean Limit Limit
THC 109 5,800 10 100,000
In Northwood, New Hampshire, drinking water wells at five homes were
contaminated by gasoline (NH DHHS, 1988). Concentrations of gasoline components
drawn in eight samples (in ppb or ug/L) are:
Detection Upper
Mean Limit Limit
Benzene 766 90 2,346
Toluene 347 5 2,426
Xylene 325 <1 1,737
The ingestion exposure to gasoline components present in contaminated drinking
water is based on the ensemble average of measurements from actual wells in Connecticut,
Maine, and New Hampshire. The other two case studies present much higher
concentrations, which could only be obtained from a well along the gasoline-plume center-
line and very close to the spill. Such conditions are not deemed representative of drinking
water contamination. The assumptions used in scenario 5 are:
• Water concentrations equal to the mean and upper limit values in Table 5-4.
• Average daily water ingestion of 2 liters/day and a body weight of 70 kg.
5.5.3 Scenario 6 • Non-ingestion Exposure to Contaminated Water
The non-ingestion pathways of volatile water contaminants through inhalation and
dermal absorption are receiving increased attention. These contaminants are transferred to
the indoor air by showers, baths, toilets, dishwashers, and washing machines. Dermal
absorption of contaminants may occur while people are bathing or washing dishes. Indoor
residential environments are not static, and the variables of air exchange rates and
temperature vary from house to house and over time. Therefore, estimating non-ingestion
intake of pollutants from household use of contaminated water is a complex process that
requires the quantification of several exposure parameters. These include source
contribution of various contaminants from different exposure media, contaminant exchange
rates from indoor and outdoor air, transfer efficiency of contaminants in water to air, and
the frequency and duration of various activities in the home. Consequently, a wide
variability in the non-ingestion exposure estimates may result from estimated contaminant
5-16
-------
TABLE 5-4
ESTIMATED EXPOSURE LEVELS FOR SCENARIOS 5 AND 6
Hydrocarbon
Exposure
THC
Mean
Upper Limit
Benzene
Mean
Upper Limit
Toluene
Mean
Upper Limit
Xylene
Mean
Upper Limit
Scenario 5
Drinking Water
(ppbv or ug/L)
5,800
100,000
500
2,300
280
1,800
300
1,500
Scenario 6
Peak Shower Air* Average Shower Air^
(ppbv) (ppbv)
34,800
600,000
3,000
13,800
1,680
10,800
1,800
9,000
17,400
270,000
1,500
6,210
840
4,860
900
4,050
1 Peak concentration is 6 times the drinking water concentration (Shehata, 1985).
2 Average concentration is 2.7 times the drinking water concentration (Shehata, 1985).
5-17
-------
levels and the model assumptions employed to determine exposure dose. Rather than
attempting to derive each of these exposure parameters, this exposure scenario focuses on
the exposure parameters that portray a reasonable upper estimate of exposure to the total
non-ingestion intake of gasoline from contaminated water supplies.
This upper estimate or "worst-case" scenario considers the following.
1. bathing and showering represent significant exposure routes that may be as large
or larger than exposure from ingestion alone (McKone, 1987; Andelman, 1985).
2. the inhaled dose during bathing and showering represent most of the total daily
inhaled dose (McKone, 1987).
3. exposure during showering or bathing are likely to be to the maximum air
concentration of contaminants because they involve the maximum release rate of
volatile chemicals from water (due to high temperature and usage rates) into a
small volume of air (e.g., shower stall).
4. showering and bathing provide the maximum opportunity for dermal absorption
due to the amount of exposed body surface area.
5. because limited data exist on most non-aromatic components of gasoline,
pulmonary and dermal absorption estimates are based on data available for
benzene, toluene, and xylene.
It is recognized that other household activities (e.g., dishwashing, laundry) may
also contribute significantly to the non-ingestion exposures. These concerns can be
addressed, however, by maximizing the water use component of the bathing/showering
scenario. For example, Andelman (1985) estimated that the upper limit of the typical per
capita water consumption was 95 liters for bathing, 60 liters for dishwashing (for a family
of four), and 34 liters for laundry. Thus, if it is assumed that the total daily water use
(approximately 200 L/day) is associated totally with bathing and showering (the activity
associated with the highest non-ingestion exposure), the resulting estimate should provide a
conservative prediction of non-ingestion exposure from all daily water-use activities.
5-18
-------
5.5.3.1 Estimation of Reasonable Upper Bound of Non-Ingestion Exposure
Pulmonary Absorption
Based on the upper limit total water consumption, estimated by Andelman (1985),
this scenario estimates a 200 liters/day water use for showering. This estimate corresponds
to a 20 minute shower with a flow rate of 10 liters/minute. If it is assumed that a bath also
takes 20 minutes, and that dermal absorption is constant throughout the exposure period,
the upper limit estimate of the dermal dose should be equal for both bathing and showering.
In a qualitative sense, however, the relative contribution of dermal exposure to the total
exposure should be greater during bathing, as showering would enhance volatilization and
hence, exposure through the inhalation route.
For the purposes of this assessment, gasoline components are assumed to exist in a
steady state for all systems for the entire duration of the shower or bath. This assumption
is not entirely accurate, as initial uptake through the pulmonary route is likely to be
relatively rapid, and initial uptake through the skin is likely to be relatively slow, when
compared to equilibrium conditions. Data are very limited, however, to estimate uptake of
the hydrocarbons during non-steady state conditions. More refined analysis in this area is
beyond the scope of this assessment.
For risk assessment purposes, it is important to estimate the doses to the lung or the
skin when evaluating toxicity to these organs. If systemic effects are evaluated, the actual
contributions via dermal absorption or inhalation are not as important as the total absorbed
dose. In either case, however, the absorbed gasoline components are assumed to enter the
bloodstream directly, thus alleviating the need to apply pharmacokinetic adjustments to
account for route-of-exposure differences.
Estimates of potential inhalation exposures of volatile compounds released during
showering have been made using both modeling and monitoring approaches. McKone
(1987) modeled a maximum shower concentration of 20 to 30 mg/m^ resulting from a 10
minute shower using 300 liters of water, containing 1 mg/L of various chlorinated organic
solvents. This corresponds to an air/water concentration ratio of 5.
Foster and Chrostowski (1987) estimated the benzene dose from a 15 minute
shower of water containing 75 ug/L of benzene. They assumed a ventilation rate of 15
liters/minute and 100 percent absorption. This resulted in an absorbed dose from inhalation
of contaminated shower air of approximately 2.3 ug/kg/day. Scaling this dose to reflect a 1
mg/L water concentration would result in an absorbed dose of 1000 ug/75 ug x 2.3
ug/kg/day = 30 ug/kg/day. For a 70 kg adult breathing 15 liters for 15 minutes, the
5-19
-------
equivalent air concentration for this dose is 9 mg/m3 (3 ppm). This air concentration is 3
times higher than the water concentration.
Work cited in Shehata (1985) presented ratios (expressed as ppb) of benzene in the
air (breathing zone) to water concentrations during 15 minute showers in eight households.
These ratios ranged from 0.7 to 5.6 with a mean of 2.7. These ratios are consistent with
the ratios associated with the modeled data presented above. Considering the preference
for monitored values over modeled values, the Shehata study (1985) are used to predict
inhalation exposure during showing in Scenario 6. Assuming that 1 ppb of benzene
concentration equals approximately 3 ug/m3, a water benzene concentration of 1 mg/L (1
ppm) would yield a mean air concentration of about 8 mg/m^ (1 ppm x 3 mg/m^ x 2.7) and
a maximum of about 16 mg/m3 (1 ppm x 3 mg/m3 x 5.6). Thus, a plausible estimate of
the inhaled dose of gasoline vapors may be between 2 and 4 mg/day. Assuming a dose of
2 mg/day from ingestion (assuming 2 liters/day water consumption), these data indicate a
reasonable upper limit on the inhaled dose from showering may be equal to or twice as
much as oral dose of the contaminants through drinking water.
Dermal Absorption
Estimation of dermal absorption is based on Pick's Law. According to this law, the
permeation rate (or flux, Js) is equal to the permeability constant (Kp) times the
concentration difference of a solute in dilute solution across a specified tissue (A Cs).
Js = Kp A Cs
For short exposure durations, such as this one, it is assumed that the concentration
differential is equal to the aqueous solute concentration. To maximize the estimate of the
dermal absorption pathway, it is also assumed that the solute concentration remains
constant (i.e., no evaporation) and that the entire body surface is covered with
contaminated water. This second assumption is not intended to be predictive of the actual
dermal dose. Rather, it is used to provide a theoretical upper limit of what that dose might
be.
The permeation constants for several alkyl benzenes have been estimated to be
about 0.001 liters/cm x hr (Brown et al., 1984). These constants represent the permeation
of the solvents through the hand and forearm. No data were found concerning the constant
for benzene. For the purposes of this assessment, therefore, it is assumed to be similar to
the constants calculated for the alkyl benzenes. Data on other gasoline components were
5-20
-------
also unavailable; however, it is assumed that the permeation constant for these compounds
is no higher than for the alkyl benzenes.
If it is assumed that the entire body of an adult is covered with the contaminated
water, the surface area available for absorption is approximately 18,000 cm2. While it is
unlikely that the total body surface is exposed throughout the shower or bath, penetration
may be greater in other areas of the body (e.g., the scalp or the genital area) than it is in the
hand or forearm. Also, absorption through the scalp may occur for a significant period of
time following the shower due to retention of contaminants by hair.
Given these assumptions, an upper limit of dermally absorbed gasoline or gasoline
components can be calculated:
Js = (0.001 liters/cm2 x hr) x (X mg/L)
= 0.001 (X)mg/cm2-hr
Uptake = (0.001 X mg/cm2 x hr) x (18,000 cm2/person) x (0.33 hr)
= 6(X) mg
Where:
0.001 liters/cm2 - hr = permeation constant
Xmg/liter = water concentration of contaminants
18,000 cm2/person = available adult surface area for absorption
0.33 hr = 20 minutes
Thus, for a concentration of 1 mg/L, the maximum absorbed dermal dose is
estimated to be 6 mg. Assuming consumption of 2 liters of water per day, the dermal
absorbed dose is 3 times the oral dose (2 mg). It also represents 3 percent of the total
exposure solvent available during the shower (in this case, 200 mg).
Available data indicate that considerable volatilization of gasoline hydrocarbons
(about 50 percent) may occur (McKone, 1987; Foster and Chrostowski, 1987). The
remaining contamination is then available for dermal contact where it can either evaporate
from the skin surface or be dermally absorbed. The potential incremental increase to the
inhaled dose contributed by skin evaporation is unknown. Maximum dermal absorption
(assuming no volatilization from the skin) would be 3 mg (half of the original 6 mg
estimate, assuming 50 percent evaporation).
5-21
-------
Combined Pulmonary and Dermal Uptake
The derivations just presented indicate a maximum combined pulmonary and dermal
uptake estimate from contaminated water of 5 to 7 mg/day (2 to 4 mg/day from pulmonary
uptake and 3 mg/day from dermal uptake). If it is assumed that oral water consumption of
2 liters/day, 2 mg/day (water concentration of 1 mg/L) represents the oral ingestion
contribution to the body burden. Given these estimates, it can be assumed that a plausible
upper bound on non-ingestion intake of contaminants from residential water use is
approximately 2.5 to 3.5 times the oral intake. Alternatively, oral intake represents
approximately one-quarter of the total body burden from contaminated domestic water.
Use of less conservative assumptions (e.g., smaller surface area available for dermal
uptake, less evaporation, less permeation through the skin) could significantly decrease
these estimates of non-ingestion exposure. For example, if average air to water
contaminant concentration ratios are assumed for benzene (see Shehata, 1985), and
benzene evaporates from the skin rather than being dermally absorbed, then non-ingestion
exposure would be approximately equal to the ingestion exposure (2 mg/day for a water
concentration of 1 mg/1).
In summary, therefore, modeling and empirical data indicate that, in general,
inhaled doses from bathing and showering are equal to the ingested dose from drinking
water. Considerations of plausible upper bounds (achieved through maximizing the
inhalation exposure) results in an estimated non-ingestion to ingestion ratio of
approximately 2:1.
5.6 CALCULATION OF BODY BURDENS IN HUMANS FROM
EXPOSURE TO GASOLINE. BENZENE. TOLUENE. AND XYLENE
Risk characterization compares the exposure values presented in this chapter with
appropriate health criteria developed from the information presented in Chapters 7 through
10. Because these lexicological criteria are expressed in terms of mg/kg/day doses,
adjustments need to be made to the exposure data (expressed in units such as ppm, mg/m^,
or mg/L) so that they appear in a form which allows for comparisons with the health data.
These adjustments are made using information on exposure levels, exposure duration, and
injection or inhalation rates. Estimates of exposure levels have been derived in previous
sections of this chapter. In the following calculations, estimates of exposure durations are
made for scenarios 1 through 4; scenario 5 assumes a daily intake of 2 liters of water, and
the exposure duration assumptions are presented above with regard to dose calculations for
scenario 6. Inhalation rates used in these calculations are 15 L/min (for light activity).
Alveolar ventilation is two-thirds of the total ventilation. Doses are calculated using both
5-22
-------
the total and alveolar ventilation rates. Doses based on total ventilation are used when the
health effects from inhalation occur directly on the respiratory tract Alveolar ventilation
rates are used to compare exposures with systemic effects, as it is assumed that only the
fraction of the inhaled dose that reaches the alveoli is absorbed into the systemic circulation.
Further discussion of the pulmonary absorption assumptions occurs in Chapter 6
(Pharmacokinetics). A summary of the doses for the six scenarios is presented in Table 5-
5.
Air concentrations are converted to a mass basis as follows:
(mg/m3) = (ppm) x (Molecular Weight)/ 24.45
UL = upper limit exposure
mean = mean exposure
Benzene (mg/m3) = (ppm) x 3.19
Toluene (mg/m3) = (ppm) x 3.77
Xylene (mg/m3) = (ppm) x 4.34
THC (mg/m3) = (ppm) x 2.76 (based on Tironi et al., 1986)
5.6.1 Scenario 1
A self-service customer is exposed via inhalation to gasoline vapors while in a
service station and, at a higher level, while pumping gas into his/her vehicle.
Self-Service Exposure = Exposure during pumping + Other exposure while in station*
The exposure for Scenario 1 assumes the following:
• Air concentrations during the refueling operation as given in Table 5-2.
• Air concentrations during the other time spent in the station equal to an average
of 2 percent and an upper bound of 10 percent of the refueling operation levels
(refer to study by Bond et al. [1986a] in Appendix B).
• Annual refueling times of 77 to 155 minutes per year.
• Average time of a station visit of 5 minutes and an average of 70 visits per year.
• Average adult body weight of 70 kg (Diem and Lentner, 1973) and a light-
activity total respiration rate of 1.2 m3/hr (0.02 m3/min) (Snyder et al., 1975),
and an alveolar ventilation rate of 0.8 m3/hr (0.01 m3/min).
• Average adult refuels his car for 55 years out of 70 year lifetime (API, 1988).
5-23
-------
TABLE 5-5
SUMMARY OP AMBIENT CONCENTRATIONS1 AND EXPOSURE DOSES
ASSOCIATED WITH EXPOSURE TO GASOLINE AND SELECTED INDICATOR
CONSTITUENTS
Scenario
Estimated exposure doses
mean maximum
(concentration)
Ambient concentrations
based on alveolar ventilation
mean maximum
Crns/ke/dav)
V
fe
scenario 1: self-service customer at gas station exposed via inhalation^
(ppm)
gasoline1 134.0 682.0
benzene 0.9 4.2
toluene 0.6 2.4
xylenes 0.2 1.1
scenario 2: gas station attendant exposed via inhalation^
(ppm)
gasoline1 19.8
benzene
toluene
xylenes
0.2
0.3
0.1
1.3
9.4 x ID'3
7.3 x lO-5
5.7 x 10'5
2.2 x lO'5
1.8
2.1 x 1C-2
3.8 x lO-2
1.5 x ID'2
scenario 3: resident living downwind of gas station exposed via inhalation^
(ppb)
gasoline1 5.4 28.0 3.1 x 10'3
benzene 0.04 0.16 2.6 xlO'5
toluene 0.08 0.37 6.2 x 10'5
xylenes 0.03 0.15 2.7 x 10'5
scenario 4: resident inhaling vapors from nearby leaking underground storage tank**
(ppm)
gasoline1
benzene 0.55 2.9 3.6X10'1
toluene 0.80 7.6 6.2 x 10'1
xylenes 0.47 4.0 4.2X10'1
1.0 x 10-•
7.2 x lO'4
4.9 x 10'4
2.6 x lO'4
1.4 x 10'1
1.6x ID'2
1.1 x 10'4
2.9 x lO'4
1.3 x lO'4
1.9
5.9
3.6
-------
to
in
TABLE 5-5
(continued)
Estimated exposure doses Ambient concentrations
based on alveolar ventilation
Scenario mean maximum mean maximum
(concentration) fmg/kg/day)
scenario 5: resident exposed to gasoline via ingestion of contaminated well water2*4
(ppb)
gasoline 5.800 100.000 1.0 xlO'1 2.9
benzene 500 2300 1.4 x 10'2 7.0 xlO'2
toluene 280 1800 8.0 x 10'3 5.0 x 10'2
xylenes 300 1500 8.6 x 10'3 4.0 x 10'2
scenario 6: resident exposed via inhalation and dermal contact during showering1*^
gasoline
benzene
toluene
xylenes
NA
NA
NA
NA
NA
NA
NA
NA
1.7 x 10'1
1.4 x 1C'2
8.0 x 10'3
8.6 x 10'3
3.4 x lO'1
2.8 x ID'2
1.6 x ID'2
1.7 x ID'2
a assumes inhalation of 14.4 cuM/day
b assumes inhalation of 21.6 cuM/day
1 refer to Chapter 5 and Appendix B for detailed derivation of exposure doses
2 assumes ingestion of 2 L water/day
3 based upon arithmetic means of monitoring studies described in "Exposure Assessment"
4 based upon limited case-study information. Estimated risks for any given site need to be determined on a site-specific basis.
5 assumes mean values equal mean drinking water exposures, and upper limits equal twice drinking water maxima
Equations:
Benzene (mg/m3) = (ppm)(3.19)
Toluene (mg/m3) = (ppm)(3.77)
Xylene (mg/m3) = (ppm)(4.34)
THC (mg/m3) = (ppm)(2.76)
-------
Exposure during pumping = (Air concentration during pumping) x (breathing rate) x
(time per avg. lifetime day spent pumping gas)/ (body weight)
'"Other exposure while in station = (2 to 10 percent of air concentration during pumping) x
(breathing rate) x (time per avg. lifetime day spent in station)/ (body weight)
Based on Total Ventilation Rate (0.02 m3/min)
THC, mean
THC, UL
Benzene, mean
Benzene, UL
Toluene, mean
Toluene, UL
Xylene, mean
Xylene UL
= (134 ppmv) x (2.76) x (0.02 m3/min) x (55/70) x (77 min/yr + (273
min/yr x 0.02)/((365 day/yr) x (70 kg))
= 1.9 x 10'2 mg/kg/day
= 682.8 x 2.76 x 0.02 x (55/70) x (155 + (195 x 0.10))/(365 x 70)
= 2.0 x 10-] mg/kg/day
= 0.9 x 3.19 x 0.02 x (55/70) x (77 + (273 x 0.02))/(365 x 70)
= 1.5 x 10'4 mg/kg/day
= 4.2 x 3.19 x 0.02 x (55/70) x (155 + (195 x 0.10))/(365 x 70)
= 1.4 x 1Q-3 mg/kg/day
= 0.6 x 3.77 x 0.02 x (55/70) x (77 + (273 x 0.02))/(365 x 70)
= 1.1 x lO'4 mg/kg/day
= 2.4 x 3.77 x 0.02 x (55/70) x (155 + (195 x 0.10))/ (365 x 70)
= 9.7 x 10'4 mg/kg/day
= 0.2 x 4.34 x 0.02 x (55/70) x (77 + (273 x 0.02))/(365 x 70)
= 4.4 x 10'5 mg/kg/day
= 1.1 x 4.34 x 0.02 x (55/70) x (155 + (195 x 0.01))/(365 x 70)
= 5.1 x 10'4 mg/kg/day
Based on Alveolar Ventilation Rate (0.01 m2/min)
THC, mean
THC, UL
= (134 ppmv) x (2.76) x (0.01 m3/min) x (55/70) x (77 min/yr + (273
min/yr x 0.02)/((365 day/yr) x (70 kg)
= 9.4 x 10-3 mg/kg/day
= 682.8 x 2.76 x 0.01 x (55/70) x (155 + (195 x 0.10))/(365 x 70)
= 1.0 x 10-! mg/kg/day
Benzene, mean = 0.9 x 3.19 x 0.01 x (55/70) x (77 + (273 x 0.02))/(365 x 70)
= 7.3 x 10'5 mg/kg/day
Benzene, UL = 4.2 x 3.19 x 0.01 x (55/70) x (155 + (195 x 0.10))/(365 x 70)
= 7.2 x lO'4 mg/kg/day
5-26
-------
Toluene, mean
Toluene, UL
Xylene, mean
Xylene UL
= 0.6 x 3.77 x 0.01 x (55/70) x (77 + (273 x 0.02))/(365 x 70)
= 5.7 x 10'5 mg/kg/day
- 2.4 x 3.77 x 0.01 x (55/70) x (155 + (195 x 0.10))/ (365 x 70)
= 4.9 x 10'4 mg/kg/day
= 0.2 x 4.34 x 0.01 x (55/70) x (77 + (273 x 0.02))/(365 x 70)
= 2.2 x 10-5 mg/kg/day
= 1.1 x 4.34 x 0.01 x (55/70) x (155 + (195 x 0.01))/(365 x 70)
= 2.6 x 10'4 mg/kg/day
5.6.2 Scenario 2
A service station attendant is exposed to gasoline vapors from a variety of activities
during a normal work-day at the station.
The inhalation exposure to gasoline vapors for a service station attendant employs
these assumptions:
• Air concentrations during attendant working hours as given in Table 5-2.
• An average workday of 8 hours, 5 days/week, 50 weeks/year, 50 years/lifetime
(API, 1988).
• An average adult has a body weight of 70 kg, a total respiration rate of 1.2
nv/hr, and an alveolar respiration rate of 0.8 m^/hr.
Occupational exposure = (air concentration during work) x (breathing rate) x (time per avg.
lifetime day spent working)/ (body weight)
Based on Total Ventilation Rate
THC, mean = (19.8 ppmv) x (2.76) x (0.02 m^/min) x (60 min/hr) x (8 hr/day) x
(250/365) x(50/70)/(70 kg)
= 3.7 mg/kg/day
Benzene, mean = 0.2 x 3.19 x 0.02 x 60 x 8 x (250/365) x (50/70)/(70)
= 4.2 x ID'2 mg/kg/day
Benzene, UL
Toluene, mean
Xylene, mean
= 1.3 x 3.19 x 0.02 x 60 x 8 x (250/365) x (50/70)/(70)
= 2.8x 10'1 mg/kg/day
= 0.3 x 3.77 x 0.02 x 60 x 8 x (250/365) x (50/70)/(70)
= 7.6 x ID'2 mg/kg/day
= 0.1 x 4.34 x 0.02 x 60 x 8 x (250/365) x (50/70)/(70)
= 2.9 x 10-2 mg/kg/day
5-27
-------
Based on Alveolar Ventilation Rate
THC, mean = (19.8 ppmv) x (2.76) x (0.01 m3/min) x (60 min/hr) x (8 hr/day) x
(250/365) x (50/70)7(70 kg)
= 1.8 mg/kg/day
Benzene, mean = 0.2 x 3.19 x 0.01 x 60 x 8 x (250/365) x (50/70)/(70)
= 2.1 x lO'2 mg/kg/day
Benzene, UL = 1.3 x 3.19 x 0.01 x 60 x 8 x (250/365) x (50/70)/(70)
= 1.4 x 10'! mg/kg/day
Toluene, mean = 0.3 x 3.77 x 0.01 x 60 x 8 x (250/365) x (50/70)/(70)
= 3.8 x 10-2 mg/kg/day
Xylene, mean = 0.1 x 4.34 x 0.01 x 60 x 8 x (250/365) x (50/70)/(70)
= 1.5x10-2 mg/kg/day
5.6.3 Scenario 3
A resident living near a service station is exposed to vapors whenever the wind
direction is such that they are downwind of the station. The U.S. EPA Industrial Source
Complex Long Term model was used to calculate ambient air concentrations in residential
neighborhoods near a station. The upper limit exposure calculation uses the maximum
modeled annual average concentration, corresponding to a distance of 30 meters, the
closest distance a house could reasonably be to a service station. The mean exposure
calculation uses the average modeled annual concentration in an area defined by rings of 30
meters and 200 meters downwind.
ISCLT modeling is based on an emission rate of 1 g/s. Using the mass emission
rates in Table 3-3 and an annual gasoline volume of 1,000,000 gallons, annual average
emissions in g/s are:
Mean Upper Limit
THC 1.8 xlO'1 2.2 xlO'1
Benzene 1.3 x 10'3 1.5 x 10'3
Toluene 3.5 x 10'3 4.1 x 10'3
Xylene 1.6 x 10'3 1.9 x 10'3
For a 1.0 g/s emission rate, the ISCLT output in Appendix C shows a maximum annual
concentration of 346.6 ug/m3 and an ensemble average for all 144 receptors of 82.2
5-28
-------
ug/m3. Average concentrations at the four distances over all 36 directions are: 220.4
ug/m3 (30 m); 79.1 ug/m3 (50 m); 23.4 ug/m3 (100 m); 6.1 ug/m3 (200 m).
1 <>
THC, mean = (1.8 x 10'1) x (82.2 ug/m3) x (10'3 mg/ug) x (0.9 m3/hr) x
(24hr/day)/(70kg)
= 4.6 x 10'3 mg/kg/day
Neighborhood exposure = (annual average air concentration) x (breathing rate)/ (body
weight)
The following annual average air concentrations are assumed (in ug/m3).
Mean Upper Limit
THC 15 76
Benzene 0.11 0.52
Toluene 0.29 1.4
Xylene 0.13 0.66
The same concentrations in ppb are as follows.
THC
Benzene
Toluene
Xylene
Based on
Mean
5.4
0.04
0.08
0.03
Total Ventilation Rate
Upper Limit
28
0.16
0.37
0.15
THC, mean = (5.4) x (O.OOl).x (2.76) x (21.6 m3/d)/ 70 kg
= 4.6 x 10~3 mg/kg/day
THC, UL = (28.0) x (0.00l).x (2.76) x (21.6 m3/d)/ 70 kg
= 2.4 x 10'2 mg/kg/day
Benzene, mean = (0.04) x (O.OOl).x (3.19) x (21.6 m3/d)/ 70 kg
= 3.9 x 10'5 mg/kg/day
Benzene, UL = (0.16) x (0.001) x (3.19) x (21.6 m3/d)/ 70 kg
= 1.6 x 10~4 mg/kg/day
5-29
-------
Toluene, mean
Toluene, UL
Xylene, mean
Xylene, UL
= (0.08) x (0.00 l).x (3.77) x (21.6 m^/d)/ 70 kg
= 9.3 x 10-5 mg/kg/day
= (0.37) x (0.00l).x (3.77) x (21.6 m3/d)/ 70 kg
= 4.3 x 10~4 mg/kg/day
= (0.03) x (0.00l).x (4.34) x (21.6 m3/d)/ 70 kg
= 4.0 x 10'5 mg/kg/day
= (0.15) x (0.001).x (4.34) x (21.6 m3/d)/ 70 kg
= 2.0 x 10'4 mg/kg/day
Based on Alveolar Ventilation Rate
THC, mean = (5.4) x (O.OOl).x (2.76) x (14.4 m3/d)/ 70 kg
= 3.1 x ID'3 mg/kg/day
THC, UL = (28.0) x (O.OOl).x (2.76) x (14.4 m3/d)/ 70 kg
= 1.6 x 10'2 mg/kg/day
Benzene, mean = (0.04) x (O.OOl).x (3.19) x (14.4 m3/d)/ 70 kg
= 2.6 x 10'5 mg/kg/day
Benzene, UL = (0.16) x (O.OOl).x (3.19) x (14.4 m3/d)/ 70 kg
= 1.1 x 10'4 mg/kg/day
Toluene, mean = (0.08) x (O.OOl).x (3.77) x (14.4 m3/d)/ 70 kg
= 6.2 x 10'5 mg/kg/day
Toluene, UL = (0.37) x (O.OOl).x (3.77) x (14.4 m3/d)/ 70 kg
= 2.9 x 10'4 mg/kg/day
Xylene, mean = (0.03) x (O.OOl).x (4.34) x (14.4 m3/d)/ 70 kg
= 2.7 x 10'5 mg/kg/day
Xylene, UL = (0.15) x (O.OOl).x (4.34) x (14.4 m3/d)/ 70 kg
= 1.3x 10'4 mg/kg/day
5.6.4 Scenario 4
The inhalation exposure to gasoline vapors that enter a home from the soil
contacting its lowest level is based on the case study with the highest measured levels. The
following assumptions were employed in scenario 4.
• Air concentrations in the home equal to the mean and upper limit values from
Groundwater Technology (1988).
5-30
-------
• People are home 24 hours/day.
• Average body weight of 70 kg, a respiration rate of 1.2 m3/hr, and an alveolar
ventilation rate of 0.8 m3/hr.
The mean and upper limit values for 20 samples, collected by Groundwater
Technology (1988), each collected over a 200 minute period, (in ppmv) are:
THC
Benzene
Toluene
Xylene
Based on Total Ventilation Rate
Mean
No Data
0.55
0.80
0.47
Upper Limit
No Data
2.9
7.6
4.0
Benzene, mean = (0.55) x (3.19) x (21.6 m3/d)/ 70 kg
= 5.4 x lO'l mg/kg/day
Benzene, UL = (2.9) x (3.19) x (21.6 m3/d)/ 70 kg
= 2.9 mg/kg/day
Toluene, mean = (0.8) x (3.77) x (21.6 m3/d)/ 70 kg
= 9.3x 10'1 mg/kg/day
Toluene, UL = (7.6) x (3.77) x (21.6 m3/d)/ 70 kg
= 8.8 mg/kg/day
Xylene, mean = (0.47) x (4.34) x (21.6 m3/d)/ 70 kg
= 6.3xlO"1 mg/kg/day
Xylene, UL = (4.0) x (4.34) x (21.6 m3/d)/ 70 kg
= 5.4 mg/kg/day
Based on Alveolar Ventilation Rate
Benzene, mean = (0.55) x (3.19) x (14.4 m3/d)/ 70 kg
= 3.6x 10'1 mg/kg/day
Benzene, UL = (2.9) x (3.19) x (14.4 m3/d)/ 70 kg
= 1.9 mg/kg/day
Toluene, mean = (0.8) x (3.77) x (14.4 m3/d)/ 70 kg
= 6.2x 10'1 mg/kg/day
Toluene, UL = (7.6) x (3.77) x (14.4 m3/d)/ 70 kg
= 5.9 mg/kg/day
5-31
-------
Xylene, mean
Xylene, UL
: (0.47) X (4.34) x (14.4 m3/d)/ 70 kg
4.2 x 10'1 mg/kg/day
•• (4.0) x (4.34) x (14.4 m3/d)/ 70 kg
• 3.6 mg/kg/day
5.6.5 Scenario 5
The ingestion exposure to gasoline components present in contaminated drinking
water is based on the ensemble average of measurements from actual wells in Connecticut,
Maine, and New Hampshire. The assumptions used in scenario 5 are as follows.
• Water concentrations equal to the mean and upper limit values in Table 5-5.
• Average daily water ingestion of 2 liters/day and a body weight of 70 kg.
THC, mean
THC, UL
Benzene, mean
Benzene, UL
Toluene, mean
Toluene, UL
Xylene, mean
Xylene, UL
= (5800) x (0.001) x (2 L/day))/70 kg
= 0.17 mg/kg/day
= (100,000) x (0.001) x (2 L/day))/ 70 kg
= 2.9 mg/kg/day
- (500) x (0.001) x (2 L/day))/ 70 kg
= 1.4x 10'2 mg/kg/day
= (2300) x (0.001) x (2 L/day))/70 kg
= 6.6 x 10'2 mg/kg/day
= (280) x (0.001) x (2 L/day))/ 70 kg
= 8.0 x ID'3 mg/kg/day
= (1800) x (0.001) x (2 L/day))/70 kg
= 5.1 x 10'2 mg/kg/day
= (300) x (0.001) x (2 L/day))/ 70 kg
= 8.6 x ID'3 mg/kg/day
= (1500) x (0.001) x (2 L/day))/ 70 kg
= 4.3x 10'! mg/kg/day
5.6.6 Scenario 6
Modelling and empirical data indicate that, in general, inhaled doses from bathing
and showering are equal to the ingested dose from drinking water. Considerations of
plausible upper bounds (achieved through maximizing the inhalation exposure) result in an
estimated non-ingestion to ingestion ratio of approximately 2:1. Non-ingestion doses
resulting from residential exposure to gasoline-contaminated drinking water can be derived,
therefore, by using the ingested dose estimates for scenario 5.
5-32
-------
THC, mean = 0.17 mg/kg/day
THC, UL = 0.17 mg/kg/day x 2 = 0.34
Benzene, mean = 1.4 x 10*2 mg/kg/day
Benzene, UL = 1.4 x 10~2 mg/kg/day x 2 = 0.028
Toluene, mean = 8.0 x 10'3 mg/kg/day
Toluene, UL = 8.0 x 10'3 mg/kg/day x 2 = 0.016
Xylene, mean = 8.6 x 10'3 mg/kg/day
Xylene, UL = 8.6 x 10'3 mg/kg/day x 2 = 0.017
5.7 SUMMARY
In this assessment, exposure to gasoline is based on contaminant levels in
environmental media and estimated absorption from different exposure routes. Two
exposure pathways were considered: ambient air transport of vapors and groundwater
transport of fuel leaking from an underground tank. Six possible scenarios were
calculated: exposure of the self service customer, service station attendant, and nearby
resident from service station emissions and exposure to contaminated indoor air, drinking
water, and home water use (e.g., showering) from gasoline contaminated groundwater.
Measurement survey data was used for the ambient air pathway and reports of gasoline-
contaminated water was used for the groundwater pathway. A summary of the exposure
doses is presented in Table 5-6. This analysis showed that significant exposures occur for
the self-service customer, the fuel attendant, and the nearby resident with an average of 134
ppm, 19.8 ppm, 5.4 ppb THC. For benzene, the averages are 0.9 ppm, 0.2 ppm, and
0.04 ppb. Upper limits are about 5- to 9-fold higher. For the groundwater pathway
scenarios, data from case studies reported in the NESCAUM states were evaluated and
showed that benzene is a predominant contaminant Calculation of estimated absorption
from lungs and through skin from non-ingestion use of gasoline contaminated water
suggest about a 2:1 non-ingestion to ingestion ratio.
5-33
-------
6. PHARMACOKINETICS
6.1 INTRODUCTION
Phannacokinetics is a description of the effects of an organism on the substances to
which it is exposed. These effects consist of absorption, distribution, metabolism, and
excretion (or elimination). The substance may exert its toxic effect(s) during any of these
phases. Phannacokinetics depends upon both the organism and the characteristics of the
particular substance(s) to which it is exposed. Indeed, the toxic effects of a substance may
determine an organism's capabilities to absorb, distribute, metabolize, and excrete the
substance. This possibility is particularly relevant to gasoline because gasoline is a mixture
of substances, but little if any information about such synergistic effects is available.
This chapter will discuss the absorption, distribution, metabolism, and excretion of
gasoline and its major components: benzene, toluene, and xylene.
6.2 PHARMACOKINETICS OF GASOLINE
Human exposure to gasoline and gasoline components may occur via inhalation,
oral, dermal, or perinatal administration. The pattern of absorption, distribution, and
elimination via inhalation varies for the individual components of the gasoline mixture.
One factor that determines the uptake of gasoline components is the blood/air partition
coefficient. This coefficient is the ratio of the free gas concentrations in the blood and air
under equilibrium concentrations. Compounds with higher blood/air partition coefficients
are absorbed to a greater extent than are those compounds with lower partition coefficients.
Determinations of blood/gas partition coefficients for selected gasoline components are
presented in Table 6-1.
Other factors which influence pulmonary absorption of gasoline are exposure
duration and exposure concentration. Initially, pulmonary absorption proceeds quite
rapidly, until the blood concentrations of the gasoline components equilibrate with the air
concentrations. Subsequently, the absorption should equal the clearance unless changes
occur in the exposure concentration (Andersen, 1983).
Investigations concerning oral ingestion of gasoline compounds have found that
absorption via this route is essentially complete. This observation is attributable to the high
lipophilicity of the hydrocarbon molecules in gasoline, the large surface area of the
gastrointestinal tract, and the long residence time of the gasoline components in the
gastrointestinal tract relative to the respiratory tract
Information concerning the dermal absorption of gasoline and gasoline components
is very limited. Most data on skin absorption are based on experiments in which thumbs,
6-1
-------
TABLE 6-1
HUMAN BLOOD/GAS PARTITION COEFFICIENTS AT 37°C FOR
SELECTED COMPOUNDS IN GASOLINE*
Substance Blood/gas Partition Coefficient
Benzene 9.0
Cyclohexane 0.6
Ethane 0.1
Ethyl Benzene 28.4
Ethylene 0.15
Toluene 15.6
Xylene (mixed) 42.1
*When multiple coefficients were given, the maximum value was selected.
SOURCE: Fiserova-Bergerova, 1983.
6-2
-------
hands, or arms were immersed in pure liquid with subsequent measurement of blood or
exhaled air levels. These experiments have indicated that dermal absorption of
hydrocarbon solvents is very low relative to oral absorption. Tables 6-2 and 6-3 and
Figure 6-1 describe the penetration rates of selected aromatic and aliphatic hydrocarbons.
These data indicate that skin penetration is greater for the aromatic components, particularly
for benzene. Skin penetration rates for aromatic compounds appear to increase with
increasing application times.
There are several factors that may contribute to enhanced absorption of volatile
hydrocarbons through the skin (Brown et al., 1984). Among these are the degree of skin
hydration, the lack of a homeostatic response of the individual to dilute solvent exposure
(e.g., the compacting of the stratum corneum), skin temperature, skin condition, regional
differences in absorption (e.g., penetration of organics through the scalp, underarm, and
scrotum are much higher than through the epidermis of the hand), and individual
variability. These factors are rarely investigated in dermal absorption studies. Such
limitations in the data base may serve to significantly underestimate the degree of skin
absorption associated with aqueous solutions of gasoline and other volatile organic
substances.
Dowty et al. (1976), in a study of 11 pregnant women, actually found higher
concentrations of aromatic compounds (benzene and styrene) in the umbilical cord blood
relative to the maternal blood. This finding is significant in that it indicates that perinatal
exposure to toxic gasoline constituents may be at least as great as maternal exposure.
Once absorbed into the blood, the gasoline components partition themselves
between the plasma and the serum. Binding to serum proteins reduces the amount of free
compounds available for exchange with air and tissues. Bound and unbound gasoline
components are transported through the blood to the various tissues. Free compound is
then partitioned into these tissues at a rate determined by the tissue/blood partition
coefficients, the degree of intracellular binding, and the rate of clearance by the tissue.
Various tissue/blood partition coefficients for selected gasoline components are presented in
Table 6-4.
Tissue clearance is accomplished by the metabolism of the parent hydrocarbons into
more polar compounds, which can be excreted more efficiently. Many hydrocarbons are
metabolized by high capacity mixed function oxidase systems in the liver, although
metabolism may occur in other organs as well. Metabolic efficiency is dependent on
several variables, including the tissue concentration, the affinity of the specific compound
for the enzyme system, the enzyme interactions with other gasoline components (the nature
of the interaction is dependent on dose of the chemicals), the nutritional status of the
6-3
-------
TABLE 6-2
AMOUNTS OF PENETRATION OF FIVE AROMATIC HYDROCARBONS
THROUGH RAT SKIN
Solvents S
Benzene
Toluene
Styrene
Ethylbenzene
o-Xylene
i kin area
(cm2)
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
Application time
(hr)
0.5
1.0
1.5
2.0
2.5
1
2
3
4
5
2.5
3
4
5
6
3
3.5
4
5
6
3
3.5
4
5
6
Number of
experiments
9
10
8
12
7
2
7
7
10
8
4
9
7
5
3
9
6
4
8
4
7
7
6
8
9
Amount of solvent
penetrated (jig)
41.3 + 6.9
157 + 19
437 + 66
581+62
917±91
9.8 + 1.6
58.5 + 8.7
185 + 25
270 + 34
431+22
42.8 + 8.9
136+16
161 + 10
257 + 41
332 ±59
18.2 + 2.5
22.3 + 3.1
31.8 + 3.1
41.1 +5.2
68.0 + 12.0
10.4 + 1.3
18.8 + 2.1
34.3 + 4.2
41.1 +5.5
59.4 + 9.9
NOTE: One ml of test solvent was applied to 2.55 cm2 of excised rat skin fixed to the
diffusion cell. After an adequate application time, the amount of penetration through
excised rat skin was determined by the multiple-phase method. The results are the
experimental average and standard error.
6-4
-------
TABLE 6-3
AMOUNTS OF PENETRATION OF FIVE ALIPHATIC HYDROCARBONS
THROUGH RAT SKIN
Solvents Skin area
n-Pentane
2-Methylpentane
n-Hexane
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
Application time
(hr)
3
3.5
2.55
4.5
5
5.5
17.5
18
18.5
4
5
5.5
17.5
19
20
22
2
3
4
5
5.5
6
7
16.5
17
17.5
18
22
Number of
experiments
6
4
4
3
8
8
2
2
2
2
2
2
2
3
2
2
5
4
6
10
9
3
3
2
3
2
2
2
Amount of solvent
penetrated (jig)
5.19 + 0.90
6.89 + 1.58
8.49 + 1.43
10.4 + 3.58
8.99 + 0.81
12.6 + 2.09
85.1 + 22.6
88.9 + 12.0
82.4 + 19.8
0.940 + 0.058
1.26 + 0.19
1.29 + 0.01
4.27 + 0.56
5.65 + 0.52
5.45 + 0.77
6.31 ± 1.51
0.353 + 0.072
0.407 + 0.165
0.688 + 0.090
0.661+ 0.082
0.742 + 0.070
0.993 + 0.306
1.25+0.14
2.43 + 0.58
3.16 + 0.27
2.58 + 0.35
2.80 + 0.55
2.99 + 0.03
6-5
-------
TABLE 6-3
(continued)
Solvents
n-Heptane
n-Octane
Skin area
(cm2)
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
2.55
Application time
(hr) exj
4
4.5
5
5.5
6
10
13
16
20
23
29
51
Number of
jeriments
2
2
2
2
2
4
3
4
9
9
3
2
Amount of solvent
penetrated (fig)
1.30 + 0.35
1.82 + 0.09
2.28 + 0.45
2.33 ± 0.37
2.29 + 0.71
3.53 ± 0.34
4.60 + 0.62
5.88 + 0.34
7.37 + 0.51
not detectable
not detectable
0.073 ± 0.019
NOTE: Experimental conditions were the same as described in Table 6-2.
6-6
-------
Figure 6-1
Skin penetration for five aromatic hydrocarbons
9—O BENZEN
O—O TOLUENE
STYRENE
ETHYL BENZENE
O—OO-XYLENE
Y=-0 152X-0.310
/ Y=0 USX-0.326-.
2 4
APPLICATION TIME(hr)
NOTE: The equation for the steady-state linear region of each curve was determined from a
plot of the cumulative amount of penetration of solvent versus application time by using the
method of least-squares. Each point is presented as the mean and standard error.
SOURCE: Tsuruta, 1982.
6-7
-------
TABLE 6-4
HUMAN TISSUE-GAS PARTITION COEFFICIENTS AT 37°C FOR
SELECTED COMPOUNDS IN GASOLINE*
Substance
Benzene
Octane
Toluene
Brain
18
16.5
36
Eat
425
233
962
Heart
17
18
30
Kidnev
12
8
18
Lung
12
5
21
Liver
23
26
48
Muscles
16
9
35
*When multiple coefficients were given, the maximum value was selected.
SOURCE: Fiserova-Bergerova, 1983.
6-8
-------
individual, exposure to other chemicals that influence metabolism (e.g., ethanol), and the
presence or absence of tissue injury. Except for benzene, toluene, and xylene, the
metabolism of most gasoline components has not been well characterized. The metabolism
of these specific chemicals is discussed in the following sections of this chapter.
Interactive effects among hydrocarbons in the gasoline mixture may occur as a
result of modifications in metabolizing enzymes. These modifications may also alter the
body's response to other substances. While interactions may be significant at occupational
levels of exposure, their importance at lower, ambient concentrations has not been
characterized.
When administered in doses capable of saturating biotransforming enzymes in the
liver, toluene has been shown to reduce the toxicity of benzene (Andrews et al., 1977).
This competition reduced the conversion of benzene to its toxic metabolite (Gut, 1981,
1982). Andrews et al. (1977) found that, in rats, co-administration of toluene and benzene
produced a smaller adverse effect on the incorporation of iron into erythrocytes than did
benzene alone. Depression of bone marrow cellularity in mice was less when toluene and
benzene were administered together than where benzene was administered alone over a six-
day period (Tunek et al., 1982). According to Russian studies cited in Dean (1985b),
inhalation of benzene and toluene produced additive clastogenic effects on mouse
lymphocytes. The purity of the toluene used in these studies was not stated. Gad El-
Karim et al. (1984) found that pure toluene ameliorated the clastogenic effect of benzene on
bone marrow cells in mice. These and other findings of toluene inhibition of benzene
toxicity are of questionable importance at low exposure concentrations when enzyme
systems are not saturated (Goldstein, 1985).
Indeed, it is possible that toluene may enhance the toxicity of other solvents through
the induction of the cytochrome P-450 enzyme system (Ungvary et al., 1981; Pyykko,
1980,1984). Toftgard et al. (1982), after exposing rats for three days to 1,875 mg/m3,
5,625 mg/m3, and 11,250 mg/m3 of toluene, found dose-dependent increases in the liver
microsomal metabolism of biphenyl and benzo(a)pyrene.
Depending on the toxic endpoint of concern, the fate of the gasoline metabolites
also require investigation. The biological half-lives of the metabolites are important
considerations when determining whether the toxicity is local (e.g., free radical formation
at the site of metabolism) or whether it could be systemic.
There are three important points to consider in regard to the pharmacokinetics of
gasoline:
6-9
-------
1. Pharmacokinetics parameters are inter-related. For example, exercise increases
uptake (by increasing the inhaled dose rate) as well as the distribution of inhaled
gases (by diverting the blood flow to the muscles). Also, increased metabolism,
by lowering the blood level, would increase the pulmonary uptake of gasoline
vapors.
2. Metabolism varies with underlying physiological status (age, diet, pre-existing
toxicity) and exposure to other chemicals (e.g., medications). Gasoline
exposure may also modify the metabolism (and hence toxicity) of otherwise non-
harmful exposure to medications or chemical contaminants.
3. Chemical mixture exposures may generate a metabolic and lexicological profile
that could be quite different than what would be predicted on the basis of
individual chemical exposures. These profiles may also vary depending on
dose, route of administration, timing of administration, and interactions with
other xenobiotics.
6.3 PHARMACOKTNETICS OF BENZENE. TOLUENE. AND XYLENE
6.3.1 Absorption
6.3.1.1 Benzene
Inhalation
The efficiency with which benzene is absorbed following inhalation was determined
in two studies conducted by Nomiyama and Nomiyama (1974a, b, as cited in U.S. EPA,
1985b). Men and women exposed to air levels of 52 to 62 ppmv for 4 hours absorbed a
mean of 46.9 percent.
Ingesrion
The efficiency with which benzene is absorbed following ingestion has been
determined in assays involving Fischer 344 rats and B6C3F1 mice (NTP, 1986b). Based
upon use of 14C-benzene as a tracer, gastrointestinal absorption has been shown to be
nearly 100 percent in all three rodent strains. The dose(s) to which the animals were
exposed were not enumerated, although it appears that two dose levels administered were
0.5 mg/kg and 300 mg/kg.
6-10
-------
Dermal contact
The efficiency with which benzene is absorbed following dermal contact was
determined in two studies conducted by Cesaro (1946, as cited in U.S. EPA, 1985b) and
Conca and Maltagliati (195S, as cited in U.S. EPA, 198Sb). Neither study revealed
significant cutaneous absorption in humans (U.S. DHHS, 1983), although the details of
exposure levels, durations, and skin locations were not provided and, according to the
U.S. EPA, small amounts of benzene absorbed by this route may not have been detected.
More recent investigations, however, have revealed significant dermal absorption (NTP,
1986b). In one study, hairless mice exhibited significant dermal absorption of benzene
(Susten et al., 1985, as cited in NTP, 1986b). Based upon the animal data, Susten et al.
calculated potential dermal absorption of benzene by workers engaged in the manufacture
of automobile tires at 20 to 40 percent of their exposure level. Similar efficiencies were
calculated for absorption by hairless areas of human skin contacting gasoline consisting of
five percent benzene (Blank and McAuliffe, 1985, as cited in NTP, 1986b).
6.3.1.2 Toluene
Inhalation
According to Anderson (1987), "[t]oluene is readily absorbed through both the
upper and lower respiratory tract (U.S. EPA, 1983b). Astrand (1975) found that alveolar
and arterial concentrations reach a relatively constant value after 20 to 30 minutes in human
subjects exposed to 375 to 750 mg/m3 of toluene. At the end of a 30 minute exposure to
375 mg/m3 of toluene, the concentration of toluene in the expired air was 18 percent of the
inspired air concentration, while the arterial concentration was 270 percent of the inspired
air value. This corresponds to a blood/air partition coefficient of 15. Pulmonary uptake of
toluene by human subjects under steady-state conditions was determined to range from 40
to 60 percent (U.S. EPA, 1983b; WHO, 1981). Initial uptake may be closer to 75 percent
(Cohr and Stockholm, 1979)."
Ingestion
According to Anderson (1987), "[d]ata are unavailable on human gastrointestinal
absorption of toluene (U.S. EPA, 1983b). Experiments have been conducted on
laboratory animals. The results of these experiments indicate that while gastrointestinal
absorption of toluene may be slow relative to pulmonary absorption, it is, nonetheless,
complete (U.S. EPA, 1983b)."
6-11
-------
Dermal Contact
According to Anderson (1987), "[i]n experiments with human subjects, Dutkiewicz
and Tyras (U.S. EPA, 19835) measured a flux of 14 to 20 mg/cm2/hr for liquid toluene
across the skin of the human forearm. Absorption of toluene from aqueous solutions (180
to 600 mg/L) ranged proportionally from 160 to 600 mg/cm2/hr when the subjects' hands
were immersed in the solutions. The results of these experiments suggest that the relative
absorption rate of toluene in aqueous solutions is higher than the absorption rate of the pure
solvent. This interpretation, however, is inconsistent with other data which show a dose-
related enhancement of skin permeability in response to concentrated toluene solutions
(Barry et al., 1985). Absorption of toluene vapor through the skin appears to be negligible
(U.S. EPA, 1983b; WHO, 1981; Barry et al., 1985)."
6.3.1.3 Xvlene
Inhalaiion
According to NTP (1986a) "[absorption [of xylenes] has been well studied in
humans. Six men exposed to an industrial xylene mixture at concentrations of 435 mg/m3
(100 ppm) or 870 mg/m3 (200 ppm) absorbed 60 percent of the amount of xylenes
supplied to the lungs (Astrand et al., 1978). The concentration in alveolar air was relatively
low throughout the entire exposure. The ratio between the concentration in arterial blood
(milligrams per kilogram) and alveolar air (milligrams per liter) was 30-40:1 at rest or
during exercise.
"In humans exposed at 100 or 200 ppm during rest or exercise, the amount of
solvent taken up was closely related to the amount of body fat (Engstrom and Bjurstrom,
1978)." In a study of humans voluntarily exposed to a commercial xylene mixture at levels
of 200 mg/m3 (46 ppm) or 400 mg/m3 (92 ppm) for 8 hours, 64 percent of xylene isomers
was absorbed (Toftgard and Gustafsson, 1980, as cited in NTP, 1986a).
Ingestion
Data pertaining to the efficiency with which xylenes are absorbed following
ingestion by animals or humans were not found in the available literature.
Dermal Contact
According to NTP (1986a), percutaneous absorption of o-xylene was estimated to
be 0.058 mmol/hour per cm2 for SD-JCL rats (Tsuruta, 1982), 1.82 mmol/hour per cm2 in
6-12
-------
mouse skin (unknown strain), and 1.13 mmol/hour per cm^ for human skin (Engstrom et
al., 1977). Neat (stock) xylenes applied to the clipped skin of guinea pigs reportedly
caused increased vascular permeability and produced erythema after 1 minute of exposure;
the effects diminished after about 5 minutes (Steele and Wilhelm, 1966).
6.3.2 Distribution
6.3.2.1 Benzene
The principal tissues and organs to which benzene is distributed are the bone
marrow, the hematopoietic system, and the male and female reproductive systems
(Klaassen et al., 1986). In a study of eleven women, for example, benzene concentrations
in blood from the umbilical cord exceeded concentrations in maternal blood (Dowty et al.,
1976). Additional information about the distribution of benzene following exposure was
not found in the available literature.
6.3.2.2 Toluene
Following absorption, toluene, which is insoluble in water, binds to plasma
lipoproteins in the blood, and is then rapidly distributed throughout the body with the
general circulation (Anderson, 1987; Cohr and Stockholm, 1979). Its pattern of
distribution is similar following exposure via inhalation or ingestion (U.S. EPA, 1983b).
The highest concentrations occur in adipose (fat) tissue, bone marrow, kidneys, liver, and
nervous tissue (U.S. EPA, 1983b). Toluene exhibits relatively high tissue/blood partition
coefficients in bone marrow and adipose tissue (U.S. EPA, 1983b) and, accordingly,
autoradiographical studies have revealed relatively long toluene retention in these tissues
(Anderson, 1987). Repeated occupational exposure to toluene at levels ranging from 130
to 1,325 mg/m3 over a 5-day work-week resulted in no accumulation of toluene in the
blood of exposed employees (Ovrum et al., 1978). However, a weak trend of generally
increasing toluene levels in the blood of individuals occupationally exposed to air levels of
toluene ranging from 750 to 1,125 mg/m^ was discemable despite great variability
(Konietzko et al., 1980).
6.3.2.3 Xvlene
The principal tissue to which xylenes are distributed following absorption is fat
(adipose tissue) (NTP, 1986a). In one study, male Sprague-Dawley rats inhaling 14C-
labeled xylenes at a level of 45 ppm for 1 to 8 hours exhibited the highest concentration of
xylenes in subcutaneous fat (Carlsson, 1981). However, when exposure levels were
6-13
-------
increased to 250 ppm, significant concentrations of xylene metabolites were found in
nervous and muscle tissue. Indeed, metabolite concentrations in the rat cerebrum,
cerebellum, and muscles attained about 40 percent of arterial blood concentrations.
6.3.3 Metabolism
6.3.3.1 Benzene
Benzene is metabolized to a variety of products, including phenol, catechol,
hydroquinone, 1,2,4-thhydroxybenzene, /-phenylmercapturic acid, and trans-trans-
muconic acid (Klaassen et at, 1986; Konietzko et al., 1980; Lauwerys, 1975; Nomiyama
and Nomiyama, 1974a, b; NRC, 198la; NTP, 1986b). Initial metabolism in the liver
apparently consists of hydroxylation by the hepatic microsomal mixed-function oxidase to a
benzene oxide intermediate, followed by non-enzymatic production of phenol (via
spontaneous molecular rearrangement). Further hydroxylation of phenol yields
hydroquinone (pora-dihydroxybenzene). Alternatively, benzene oxide may react with
glutathione, forming premercapturic acid and then /-phenylmercapturic acid. Finally,
benzene oxide may react with the enzyme epoxide hydrolase, yielding benzene dihydrodiol,
which is in turn enzymatically oxidized by a cytosolic dehydrogenase to o-catechol (onho-
dihydroxybenzene). Pathways of benzene metabolism are depicted in Figure 6-2 (NTP,
1986b).
6.3.3.2 Toluene
Toluene metabolism is primarily mediated by microsomal enzymes in the liver
(Anderson, 1987; NRC, 198la; U.S. EPA, 1980e, 1983b). Toluene is metabolized via
three principal pathways, of which one yields hippuric acid and two yield cresol. The
cresol pathways both involve formation of unstable intermediate arene oxides, which may
be lexicologically significant insofar as these intermediates may bind with cellular
macromolecules and thereby cause mutations and cancers. These pathways, however,
otherwise account for a only relatively minor fraction of total metabolism of toluene
(Woiwode and Drysch, 1981).
The major hippuric acid pathway of toluene metabolism begins with oxidation of
toluene to benzole acid, and is followed by conjugation of the benzole acid with the amino
acid glycine, forming hippuric acid (Anderson, 1987; NRC, 1981a; U.S. EPA, 1980e,
1983b). The rate-limiting step appears normally to be the oxidation to benzoic acid,
although conjugation may be rate-limiting under conditions of glycine deficiency
(Riihimaki, 1982, as cited in Anderson, 1987). Saturation of the glycine conjugation
6-14
-------
Figure 6-2
Pathways of benzene metabolism
•copied from: NTP. Toxicology and Carcinogenesis Studies of Benzene (CAS No.
71-43-2) in F344(N Rats and B6C3Fi Mice (Gavage Studies). Research Triangle
Park, NC. National Toxicology Program, National Institutes of Health, Public.
Health Service, U. S. Department of Health and Human Services, Technical
Report Series No. 289, NIH Publication No. 86-2545,277 pp., February 1986
6-15
-------
pathway may activate an alternative conjugation pathway, namely glucuronide conjugation
with benzoic acid (WHO, 1981), although this pathway is otherwise minor owing to the
lower affinity of its enzymes (presumably for benzoic acid) (Amsel and Levy, 1969).
Identification of the rate-limiting step in toluene metabolism has attained a
(surprising) degree of importance, inasmuch as the efficiency of metabolic clearance of
toluene under chronic environmental exposure conditions appears to depend upon it
(Anderson, 1987). Anderson (1987) has drawn upon theoretical arguments evinced by
Andersen (1981,1983) and Astrand (1983), as well as empirical data evinced by Andersen
(1981), Ogata (1970), Pyykko (1984), and Clayson et al. (1985), to show that the amount
of toluene processed via the hippuric acid pathway is proportional to the concentration of
toluene in ambient air. This is equivalent to flow limitation of metabolic clearance of
toluene, as distinguished from capacity limitation. Under the flow limitation assumption,
some 70 percent of absorbed toluene is metabolically cleared, and 30 percent is retained
(Anderson, 1987). Details of the pathways described above are depicted in Figure 6-3,
which was copied from Anderson (1987, citing NRC, 1981a).
6.3.3.3 Xvlene
In animals, metabolism of xylenes appears to depend upon the route by which it
was initially absorbed (Elovaara et al., 1984; Engstrom et al., 1984; Heinonen et al., 1983;
Pyykko, 1980; Savolainen et al., 1978; Toftgard and Nilsen, 1982; Toftgard et al., 1983,
as cited in NTP, 1986a). The predominant metabolic pathways in rats (Ogata et al., 1970)
and rabbits (Bray et al., 1949) following xylene ingestion yields methylhippuric acid as an
end product, although small amounts of dimethylphenol and the methylhippuric acid
precursor, methylbenzyl alcohol, may also be produced (Bakke and Schelilne, 1970). The
importance of the minor metabolites may be increased in tissues poor or deficient in alcohol
dehydrogenase, an enzyme apparently essential to completing metabolism all the way to
methylhippuric acid.
The highest alcohol dehydrogenase activity levels have been found in the liver,
whereas the lung apparently exhibits less than 5 percent of the hepatic level, according to
one review (Bosron and Li, 1980). Thus, in one study, rabbit lung tissue exhibited both
alcohol dehydrogenase deficiency and the predominance of p-methylbenzyl alcohol as the
in vitro p-xylene metabolite, with some 2,5-dimethylphenol as well (Smith et al., 1982). It
is reasonable to extrapolate this in vitro finding to the situation of toluene inhalation in vivo,
which might favor incomplete organismal metabolism of xylenes to methylbenzyl alcohol
and dimethylphenol prior to excretion (NTP, 1986a). Details of proposed pathways of
6-16
-------
Figure 6-3
Toluene metabolism in humans and animals
UllMM
*lcBhol
C-KM-CT COOM
hlppvrlc «etd
i
tlucuroBld*
2-CTtMl
•copied from: Anderson, N. T. Risk Assessment Document for Toluene. Final Report.
Augusta, State of Maine Bureau of Health, 57 pp., January 1987; original source:
NRC. The Alkyl Benzenes. Washington, DC, National Research Council,
National Academy of Sciences Press, 1981.
6-17
-------
xylene and ethylbenzene metabolism are depicted in Figure 6-4, which was copied from
NTP (1986a).
In humans, unlike animals, metabolism of xylenes appears not to depend upon the
route by which it was initially absorbed (NTP, 1986a). This is based upon the observation
that inhalation of xylenes by humans has resulted in little if any methylbenzyl alcohol and
dimethylphenol production (Toftgard and Gustafsson, 1980). Over 95 percent of the
metabolites produced by humans voluntarily inhaling a commercial xylene mixture at levels
of 200 mg/m3 (46 ppm) or 400 mg/m3 (92 ppm) in air for 8 hours consisted of
methylbenzoic acid, which was further biotransformed to methylhippuric acid prior to
excretion. In other studies, individuals voluntarily or occupationally exposed to xylenes by
inhalation metabolized it to methylhippuric acid (Dworzanski and Debowski, 1981;
Engstrom et al., 1979). Humans inhaling air containing 150 mg/m^ of m-xylene excreted
m-methylhippuric acid (Engstrom et al., 1984; duration unspecified as cited in NTP,
1986a). Ethylbenzene (same concentration) was metabolized to mandelic and
phenylglyoxylic acids. Combined exposure to both compounds caused mutual inhibition
of the metabolism of each (Engstrom et al., 1984, as cited in NTP, 1986a).
Another synergistic effect involving xylene was noted wherein human ingestion of
ethanol at a dose of 0.8 g/kg 4 hours prior to inhalation of 6.0 or 11.5 mmol/m^ (147 or
282 ppm) of m-xylene altered the pharmacokinetics of xylene (Riihimaki et al., 1982, as
cited in NTP, 1986a). Specifically, ethanol appeared to diminish metabolic clearance of
xylenes by about 50 percent, as indicated by reduced production of methylhippuric acid.
Levels of xylene in blood increased to 150 to 200 percent. Riihimaki et al. (1982, as cited
in NTP, 1986a) suggested that this synergism was attributable to inhibition by ethanol of
microsomal enzyme mediated xylene metabolism, presumably primarily in the liver.
•
Interaction between xylenes and ethanol consistent with the above was also observed in at
least one study involving animals (Savolainen et al., 1978).
6.3.4 Excretion
6.3.4.1 Benzene
Benzene, whether inhaled or ingested, may be exhaled via the lungs in significant
quantities (Klaassen et al., 1986; CRC, 1968). In addition, benzene that has been
metabolized to mono-, di-, and tri-hydroxybenzenes may then be conjugated with
glucuronide and sulfate, and excreted in a relatively water-soluble conjugated form in the
urine. At a benzene dose of 300 mg/kg, approximately 60 percent of (14C-labeled)
benzene administered to rats is exhaled (NTP, 1986b). However, at a lower dose of 0.5
6-18
-------
Figure 6-4
Proposed metabolism pathways of xylenes in animals and humans
p-XYLENE
ETHY18ENZENE
to- and /n-«yl«n«
would b*
expected to give
analogous
retulu)
CH,CHj
p-Methylbcnzyl
alcohol
T-Phenyltthanol
CHCHCH,
p-Tolualdehyde
Acetophenan*
COCH7
p-Toluic acid
p-Tolunc acid
(p-Meinyl-
hippunc acid)
Benzoic acid
Hippune acid
COMHCHjCOOH
Hydroiy
acttophtnone
COCHjOH
Mandtlic acid
CONHCHjCOOH CHOMCOOH
•copied from: NTP. Technical Report on the Toxicology and Carcinogenesis Studies of Xylenes
(Mixed) In F344fN Hats and S6C3Fj Mice (Gamp Studies). Board Draft. Research
Triangle Park. NC. National Toxicology Program. NTP TR 327, 92 pp. plus appendices,
March 1986 f
6-19
-------
mg/kg, almost all of the benzene is excreted in the urine, indicating the failure of the low
dose to saturate metabolic processes.
6.3.4.2 Toluene
The pattern of excretion of toluene and its metabolites by animals and humans is
apparently not significantly affected by the exposure route through which toluene was
initially absorbed (Anderson, 1987; U.S. EPA, 1983b). Twenty percent of absorbed
toluene is exhaled unchanged in expired air, whereas 60 to 75 percent is eliminated as
hippuric acid in the urine (U.S. EPA, 1983b), and only a trace amount (0.06 percent) is
excreted unchanged in the urine (WHO, 1981). Biliary excretion amounts to less than two
percent, but such 'excretion' is not equivalent to elimination, since a significant fraction is
reabsorbed from the intestine (Fishbein, 1985). The time course of toluene excretion
suggests a three-compartment model, and most toluene inhaled or ingested by humans and
absorbed is excreted within 12 hours following termination of exposure (Ogata et ah,
1970).
6.3.4.3 Xvlene
The excretion of xylenes is dominated by urinary excretion of its metabolites. In
rats and rabbits exposed to xylenes orally, the major excretory product in urine appears to
be methylhippuric acid, with some excretion of methylbenzyl alcohol and dimethylphenol
(Bakke and Schelilne, 1970; Bray et al., 1949; Ogata et al., 1970, as cited in NTP, 1986a).
Following inhalation, however, it is possible that the major excretory products would be
methylbenzyl alcohol and dimethylphenol, since rabbit lung tissue produced these
metabolites in vitro, apparently owing to a deficiency of alcohol dehydrogenase in lung
tissue (see above discussion of xylene metabolism).
In humans, unlike animals, the major xylene excretory product in urine appears to
be methylhippuric acid regardless of the route of initial exposure to xylenes (NTP, 1986a).
Humans voluntarily inhaling a commercial mixture of xylenes at air levels of 200 mg/m^
(46 ppm) or 400 mg/m^ (92 ppm) for 8 hours excreted only 5 percent of absorbed xylene
unchanged in expired air (Toftgard and Gustafsson, 1980). Urinary excretion of xylenes
was negligible, whereas the predominant excretory product (apparently accounting for
more than 95 percent of total metabolites) consisted of methylhippuric acid conjugated with
the amino acid glycine. In other studies, individuals voluntarily or occupationally exposed
to xylenes by inhalation excreted methylhippuric acid in their urine (Dworzanski and
Debowski, 1981; Engstrom et al., 1979). Humans inhaling air containing 150 ppm of m-
xylene excreted m-methylhippuric acid (Engstrom et al., 1984, duration unspecified as
6-20
-------
cited in NTP, 1986a). Ethylbenzene (same concentration) was excreted in the urine as
mandelic and phenylglyoxylic acids (Engstrom et al., 1984, as cited in NTP, 1986a).
6.4 SUMMARY
Pharmacokinetics is a quantitative description of the manner in which a chemical is
handled by the body. Absorption, distribution, metabolism, and excretion are the principle
components of this description. For gasoline per se, quite rapid absorption will occur from
all exposure routes including perinatal. The dermal route appears to be slower than the
other routes. As expected, some components of the mixture are absorbed more rapidly
than others. Limited data is available on distribution, metabolism, and excretion of
gasoline. The aromatic components benzene, toluene, and xylene are relatively freely
absorbed by all exposure routes. They are distributed to several sites such as bone for
benzene, lipid for xylene and toluene, and other rapidly purfused tissues. Metabolic
pathways and rates have been defined for the three aromatics of interest, but the toxic
metabolites are not well understood. Excretion proceeds by the urinary and exhalation
routes, but excretion is relatively slow, taking hours or days, which may lead to
bioaccumulation.
6-21
-------
7. GENERAL TOXICITY OF GASOLINE AND SPECIFIC
GASOLINE COMPONENTS
This chapter describes toxic effects other than carcinogenicity, genetic toxicity,
teratogenicity, and reproductive effects. These other effects are presented separately
because they have received much greater research and public health interest than have the
aforementioned effects. This chapter on general toxicity is subdivided into three sections:
acute toxicity, subacute and subchronic toxicity, and chronic toxicity. Effects are identified
both for gasoline and for the surrogate components: benzene, toluene, and xylene. Some
overlap in effects occurs, especially for nephrotoxicity, which is evident following acute,
subchronic, or chronic exposure to gasoline.
Toxic effect categories are based on exposure duration. The studies in this chapter
are presented according to these categories. When sufficient information is available for
specific health endpoints (e.g., neurotoxicity or hematotoxicity), this information is
organized into specific subsections. This is particularly the case for the subacute,
subchronic, and chronic exposure effects of gasoline. Effects of acute exposure occur as a
result of short-term exposures (minutes to hours). Effects of subacute exposure result
from exposures over several days, and effects of subchronic exposure occur as a result of
exposure over several months. Effects of chronic exposure result from exposures of years
to lifetime. Protocols for animal toxicity studies generally specify more precise exposure
periods for categorizing toxic effect categories. These include periods of 1 day or less for
acute exposure effects, up to two weeks for subacute exposure effects, 3 months for
subchronic exposure effects, and 2 or more years for chronic exposure effects. No precise
corollary exists regarding human health effects, largely because of the wide variability
associated with human exposure situations.
Many toxic effects of gasoline are related to the basic physical and chemical
properties of the mixture that is inhaled, ingested, or contacts the skin. The saturated
hydrocarbons from C4 to C8 have strong narcotic properties and cause nausea, ataxia, loss
of consciousness, and death. The unsaturated hydrocarbons in the mixture, such as
butylene and isoprene, are weak anesthetics.
The pathologic effects of gasoline toxicity are due to its irritant action and its
lipolytic activity (Machle, 1941). Damage to the lung includes hypermia and petechial
hemorrhage, subpleural extravasation, and occasionally gross hemorrhage. Bronchitis and
pneumonia's usually result if the exposures are sublethal. In addition to the lungs, the liver,
kidney, and other tissues show parenchymal damage. In the kidney, edema and lipid
degenerative changes are seen in the proximal convoluted tubules and glomerula with
7-1
-------
albuminous deposits within glomerular spaces. The liver may show enlargement and
centrilobular cloudy swelling with some fatty changes.
Hemotoxic effects have also been associated with exposure to gasoline and benzene
(a component of gasoline). Hematotoxicity refers to the harmful effects of toxic agents on
the blood and blood forming organs (bone marrow, spleen, thymus). Common measures
of hematotoxicity include counts of various cells and cell categories within the blood,
weight changes in the blood forming organs, blood chemistry, and histopathology of blood
cells and tissues.
The approximate concentrations of the cells in normal blood are described in Table
7-1. Clinical hematological effects are defined relative to these values. Fancytopenia refers
to reductions in all blood cell types. It is a common manifestation of aplastic anemia, a
disease associated with benzene exposure. When blood cell values fall below the normal
range, the individual has an increased risk of developing infections or of hemorrhage due to
diminished blood clotting efficiency. Leucopenia refers to a decrease in white blood cell
level below the normal range. Granulocytopenia, neutropenia, lymphocytopenia (or
lymphopenia), and thrombocytopenia refer to abnormally low levels of granulocytes,
neutrophils, lymphocytes, and platelets, respectively. Conversely, leucocytosis,
lymphocytosis, and neutrophilia refer to abnormal elevations in white blood cells,
lymphocytes, and neutrophils, respectively. Leucocytosis, primarily in the form of
neutrophilia, is a characteristic response to many infectious diseases.
All blood cells are formed by the same pluripotential stem cells. At various stages
in their differentiation, some stem cells become committed to forming myeloid stem cells
(which further differentiate into erythrocytes, granulocytes, monocytes, eosinophils,
basophils, granulocytes and platelets) or lymphocytes. Thus, in addition to measuring the
specific blood cell types, hematological tests also measure effects on these stem cells.
Because of their significant proliferative capacity, changes in stem cell populations are
generally regarded as more serious than changes in blood cell levels. Thus, any damage to
these precursor cells would result in a greater and magnified loss or functional deficiency of
their progeny. By the same reasoning, changes in less differentiated stem cells are of
greater lexicological significance than changes in more differentiated stem cells (i.e., cells
more committed to the development of specific blood cells). Tests for stem cell changes
seem particularly indicated when decreases in many different white blood cells lines occur.
7-2
-------
TABLE 7-1
NORMAL VALUES FOR THE CELLULAR ELEMENTS IN
HUMAN BLOOD
Total WBC
Granulocytes
Neutrophils
Eosinophils
Basophils
Lymphocytes
Monocytes
Platelets
Erythrocytes
Females
Males
Cells/cu mm
(average)
9000
5400
270
60
2730
540
4.8 x 106
5.4 x 106
Reticulocytes
Hematocrit
Mean Corpuscular Volume
Mean Corpuscular Hemoglobin (Hb)
Mean Corpuscular (Hb)
Hemoglobin
Approximate
Normal Range
4000-11,000
3000-6000
150 - 300
0-100
1500 - 4000
300-600
200,000 - 500,000
—
0.5 - 1.5%
47% (men)
42% (women)
27 - 32 picograms
30%
16gm/dl(men)
14 grn/dl (women)
Percentage
of Total
White Cells
~
50-70
1-4
0.1
20-40
2-8
—
—
--
—
—
~
~
SOURCE: Ganong, 1967.
7-3
-------
7.1 GASOLINE
7.1.1 Effects of Acute Exposure
7.1.1.1 Human Studies
Accidental exposures to gasoline vapors may be fatal. An autopsy of a three year
old boy trapped in an overturned automobile with his head in a pool of gasoline showed
congestion, edema, and hemorrhage (Ainsworth, 1960). The mucosa of both of the
trachea and bronchi were hyperemic and filled with hemorrhagic fluid. The author reported
that the boy was exposed to more that 10,000 ppm of gasoline for 5 to 10 minutes. Wang
and Irons (1961) attributed the death of an aircraft mechanic to exposure for approximately
5 minutes to an estimated 5,000 to 16,000 ppm of gasoline vapor.
Aspiration of the hydrocarbons may produce cyanosis, tachycardia, and tachypnea,
with subsequent development of chemical pneumonia's (Rumack and Lovejoy, 1986;
Berkow, 1982). If the pneumonitis is not fatal, it may require several weeks for complete
resolution (Rumack and Lovejoy, 1986). Other acute pulmonary effects caused by
gasoline exposure are bronchoconstriction and edema (Berkow, 1982). The effects are not
peculiar to gasoline, but are generally associated with a variety of irritating anesthetic gases.
Respiratory infections may be a secondary effect, occurring as a consequence of the
pulmonary edema.
Every year, ingestion of petroleum distillates and halogenated hydrocarbon solvents
is responsible for more than 25,000 poisonings in infants and small children (Berkow,
1982). One quarter of all accidental poisoning deaths fall into this group (Berkow, 1982).
Initial signs associated with acute ingestion of gasoline and other petroleum distillates
include mucous membrane irritation, vomiting, and central nervous system depression
(Rumack and Lovejoy, 1986). Machle (1941) reviewed acute toxicity from accidental
ingestion of gasoline and reported a single fatal oral dose of 7.5 g/kg, although death has
been reported to occur from ingestion of as little as 10 g (0.14 g/kg), and recovery has been
reported following ingesrion of as much as 250 g (3.6 g/kg). Death may occur as a result
of severe chemical pneumonitis, from progressive medullary paralysis in the brain and
subsequent respiratory arrest, or through anythmias due to cardiac sensitization to
endogenous sympathetic amines (U.S. EPA, 1988). The cardiotoxic effects of lower level
gasoline vapor exposure have not been adequately investigated and may represent a
significant data gap.
The variability in the responses to gasoline could be due to differences in the
chemical composition of gasoline. For example, the absorption of gasoline will be
7-4
-------
enhanced if the mixture contains a high concentration of benzene and other aromatic
hydrocarbons (which are more soluble in blood). Other sources of variability could include
whether the substance was aspirated, time between ingestion and emesis (thereby affecting
the effective dose absorbed), individual susceptibility (e.g., age, size, pre-existing illness),
and presence in the stomach of food and fats that could alter absorption. Viscosity has
been shown to be the most important physical property of petroleum derivatives, as it
determines the degree of hazard of their aspiration (Rumack and Lovejoy, 1986; Berkow,
1982). This finding is supported by animal studies which indicate that hydrocarbons in the
respiratory tract are 140 times more toxic than they are in the gastro-intestinal tract
(Berkow, 1982).
Accidental poisonings may also result from gasoline intrusion into homes as a result
of gasoline spills or leaks from storage equipment. Skin and sensory irritation, as well as
central nervous system depression, have been reported by residents in these affected
buildings; however, few attempts have been made to correlate the frequency or severity of
symptoms with exposure concentrations (Shehata, 1985).
In addition to accidental poisonings, intentional inhalation of volatile organic
solvents (such as those contained in the aromatic fraction of gasolines) is common
throughout the world. It has been estimated that between 7 and 12 percent of high school
students in the United States have tried solvent-sniffing at least once, and that nearly 4
percent sniff regularly (Salamanca-Gomez et ah, 1989). A study of 15 adolescent patients
in Mexico City identified significant increases in sister chromatid exchanges and
chromosomal alterations associated with acute intoxication of solvents (paint thinners, nail
polish removers, toluene, benzene, and commonly available cements) (Salamanca-Gomez
et ah, 1989). These increases were significant relative to both controls and chronically
habituated users. These investigators also found electroencephalogram abnormalities in
nine patients, low hemoglobin values in six patients, increased creatinine in five patients,
high concentrations of alkaline phosphatase in four patients, and high concentrations of
transaminases in two patients.
Individuals who intentionally inhale gasoline have reported that 15 to 20 breaths of
the vapors are sufficient to produce intoxication for 5 to 6 hours (Bartlett and Tapia, 1966;
Black, 1967). Due to the lipid solubility of the components of gasoline, vapors are rapidly
absorbed from lungs and the onset of symptoms occurs within 3 to 5 minutes
(Anonymous, 1967, as cited in Poklis and Burkett, 1977). The mild intoxication may be
accompanied by nausea and vomiting. After prolonged inhalation or rapid inhalation of
highly concentrated vapors, the person may experience a phase of violent excitement
followed by unconsciousness and coma (Rudd, 1944, as cited in Poklis and Burkett,
7-5
-------
1977). Ocular signs observed may include wide dilation of the pupils, unequal pupils that
may be fixed, nystagmus, and conjunctiva! deviation (Poison and Tattersall, 1969).
Quantitative dose-response information concerning human exposure to gasoline
vapors are limited. Drinker et al. (1943) exposed 13 male human volunteers, aged 23 to 45
years, to vapors of unleaded gasoline and to the volatile fraction distilled below 111°C.
Slight eye irritation and some gastrointestinal disturbances were reported upon exposure to
140 to 270 ppm of either straight gasoline or the gasoline distillate for 8 hours. In another
study, 10 volunteers were exposed to 200, 500, and 1,000 ppm of each of three types of
commercial unleaded gasoline for 30 minutes and reported only eye irritation; the number
of subjects reporting irritation increased with increasing concentration (Davis et al., 1960).
In controlled studies, exposure to 500 to 1,000 ppm of gasoline vapor from 0.5 to 1 hour
produced varying degrees of irritation to eyes, nose, and throat, and slight dizziness
(Drinker et al., 1943; Machle, 1941; Poklisand Burkett, 1977). Exposure to 1,000 to
3,000 ppm for 0.5 to 1 hour produced varying degrees of nausea, headache, dizziness,
numbness, and anesthesia (Drinker et al., 1943; Machle, 1941; Poklis and Burkett, 1977).
Exposure to 10,000 ppm produced nose and throat irritation in 2 minutes, dizziness in 4
minutes, and deep anesthesia in 4 to 10 minutes (Poklis and Burkett, 1977).
Studies have shown that exposure to volatile hydrocarbon mixtures may cause
sensory irritation and central nervous system effects in humans at levels well below current
occupational standards. Molhave (1984) studied healthy human subjects exposed to
organic gases and vapors that are common in building materials. He found that the subjects
showed significant mucous membrane irritation in response to total hydrocarbon
concentrations of 5 and 25 mg/nv compared to individuals exposed to clean air. Swedish
investigators (Hanninen el al., 1976; Seppalainen et al., 1978; Struwe and Wennburg,
1983) studied industrial workers exposed to solvent mixtures. Solvent mixtures that
contained toluene and xylene had an average hygienic effect of 0.3 (i.e., the concentration
of the components was three tenths of the threshold limit value for the mixture). Central
nervous system effects such as fatigue, impaired performance on tests of intelligence,
memory, and visual and verbal ability, nervousness, and lack of manual dexterity were
observed. These studies raise concerns regarding gasoline exposure in that the physical
and chemical properties of these different hydrocarbon mixtures may be similar to those of
gasoline vapors.
Skin exposure to solvents is one of the common causes of allergic contact
dermititis. Skin exposure to solvents may also result in pulmonary sensitivity (Karol,
1986). The defatting capacity of solvents may also impair the barrier function of the skin,
thus facilitating the penetration of other irritants, allergins, or chemicals with inherent
7-6
-------
systemic toxicity (Wahlberg, 1984). An inverse correlation has been found between the
boiling point of a solvent and its primary irritating effect (Wahlberg, 1984).
Very little information is available on the acute dermal effects of gasoline exposure
in humans, aside from anecdotal accounts reporting irritation upon washing the hands in
gasoline. More research is needed in this area.
7.1.1.2 Animal Studies
The findings from animal studies support the human findings of narcosis, irritation,
and hemorrhage associated with acute gasoline exposure. Table 7-2 provides a summary
of the acute effects of gasoline in laboratory animals.
Pulmonary Toxicity
Specific pulmonary cell types may be affected as a result of acute gasoline
exposure. Mahvi et al. (1977) exposed C57BL/65 mice intraperitoneally to naphthalene (a
minor gasoline component) in doses ranging from 0.05 to 2.0 millimoles per kilogram (6.4
to 256 mg/kg). The mice were sacrificed at 10 minutes, 1 hour, 6 hours, 12 hours, 24
hours, 48 hours, and 7 days following administration. Minor changes in the bronchiolar
epithelium were observed following the 0.05 millimoles per kilogram (6.4 mg/kg).
Naphthalene administration at doses of 1.0 mmol/kg (128 mg/kg) or greater caused Clara
cells to expand and exfoliate within 6 to 12 hours of administration. Rapidly following
Clara cell loss, abnormalities began to appear on the surface of neighboring ciliated cells.
According to the authors, this secondary effect may have occurred as a result of the
diminished capability of the Clara cells to produce the secretions necessary for normal
ciliary function.
Neurotoxicitv
A single oral dose of 15 to 25 g gasoline produced narcosis in rabbits (Lewin,
1888, as cited in Browning, 1953). Legludic and Turlais (1914, as cited in Browning,
1953) reported a lethal oral dose of 20 ml/kg body weight for rabbits. In a histological
study of the brain and the spinal cord of rabbits exposed to gasoline at an unspecified
concentration via gastric or intravenous administration, Matsushita (1935, as cited in
Browning, 1953) observed destruction of nerve cells of the brain cortex, the medulla
oblongata, and the spinal cord. He also reported dilation of the blood vessels of the brain
with hemorrhages, especially in the area of the ventricles.
The earliest report of acute effects of gasoline vapor inhalation is probably that of
Poincare (1885, as cited in Browning, 1953). In those experiments, gasoline produced
7-7
-------
TABLE 7-2
SUMMARY OF THE ACUTE EFFECTS OF GASOLINE IN LABORATORY ANIMALS
Type of test
Primary skin
irritation
Species/ Number
and sex of animal
New Zealand
White rabbits;
3/sex
Procedure
Dermal application of 0.5 ml UGa
to each of four abraded skin sites of
the animal. Test material was kept
in contact with skin for 24 hr.
Effects
Primary skin irritation score
of 0.98 indicating a slight
irritation effect
Reference
Elars Bioresearch
Lab., 1979.
New Zealand Dermal application of 0.5 ml UG to one
White rabbits; nonabraded site on each rabbit. Test
,j 3/sex material was kept in contact with
oo skin for 24 hr.
Primary eye New Zealand 0.1 ml UG placed in the right eye of each
irritation While rabbits; rabbit Test eyes of two females and one
4/sex male flushed for 1 min. with distilled
water 30 sec. after the application.
Untreated eye of each rabbit served as
control.
Skin Albino guinea 0.5 ml UG (diluted to 50% sol. after
sensitizalion pigs; 10 males first insult patch) applied to shaved back
in occluded contact for 6 hr., 3 limes/wk
for 3 wk. for total of 10 treatments. After
2-wk rest period, a challenge dose given
in the same manner. Positive controls
were treated with 0.05%
chlorodinitrobenzene in ethanol.
No signs of erythema, edema, or
other dermal effects. Primary
skin irritation index 0.0.
Test material was nonirritating
with a 24-hour average Draize
score of 0.0. Similar results were
reported by Phillips (1984) in rabbits.
Animals scored for erythema and
edema 24 hr after treatment showed
that UG was nonsensitizing
to guinea pig skin. Lack of
effect in positive controls weakens
the conclusion.
Phillips, 1984.
Elars Bioresearch
Lab., 1979.
Elars Bioresearch
Lab., 1979.
-------
TABLE 7-2
(continued)
Type of test
Species/Number
and sex of animal
Procedure
Effects
Reference
Acute oral
toxicity
Respiratory
tract irrilancy
Acute dermal
toxicity
Sprague-Dawley
rats; 4 groups of
5/sex/group; one
group had 7 males
and 6 females.
OutbredCD-1
mice; 4 groups of
4 males
New Zealand
White rabbits;
4/sex
10, 15. 17.5, 20. or 25 ml/kg UG in
single dose administered by gavage.
Exposed head only to vaporized UG
at nominal concentration of 104.22 mg/L
for 1 min. Then allowed to recover for
10 min. by exposing to room air only.
Animals were again exposed for 1 min.
and then to room air for 5 min. Respiratory
patterns were monitored continuously.
Single application of 0.5 ml UG/kg
body weight to abraded and non-abraded
skin. Patches were in occluded contact
with the skin for 24 hr.
Elars Bioresearch
Lab., 1979.
j: 18.75 ml/kg with a 95%
confidence interval of 16.3 to 21.6
ml/kg; mortality rales in 5 dose
groups ranged from 0 to 90%, respectively.
Same toxic signs were seen in all dose
groups increasing with severity with
increased dose. Lungs showed effects
ranging from mild irritation and
congestion to fluid-filled abscesses.
Animals suffered from diarrhea.
Intestines and often the stomach were
hemmorhagic among animals that died
before the end of the 14-day observation
period. Animals had blood around eyes,
nose, and mouth. In some, heart was
irregularly shaped and enlarged.
Moderate respiratory pauses noted in Phillips, 1984.
all exposed animals; 20 to 50% decreases
noted in the respiratory rate in 3/4
animals during second exposure. UG
appeared to produce slight to moderate
upper airway irritation in male mice.
LD5Q 0.5 ml/ test material produced
skin irritation in animals. One
death occurred during 14-day observation
period. It was unclear whether the
death was treatment-related.
Elars Bioresearch
Lab., 1979.
-------
TABLE 7-2
(continued)
Type of test
Species/ Number
and sex of animal
Procedure
Effects
Reference
Subacute deimal New Zealand
toxicity White rabbits;
4/sex
Daily application of UG at doses of 2.5.
5, or 8 ml/kg to shaved area for 10 days
with a 2-day nondosing rest period after
first 5 days' application. Each
application was kept under occluded
contact with the skin for 24 nr.
The calculated dermal LDso >8 ml/kg; Elars Bioresearch
animals in 8-ml/kg group showed Lab., 1980.
decreased appetites and emaciation
resulting in weight loss. No mortality
observed in any test group. Post-mortem
examination showed treatment-related
skin lesions and congested kidneys in
all test groups. Histopathology
revealed acanthosis, acute inflammation,
chronic inflammation, crusting dermal
congestion, dermal edema, hyperkeratosis,
and epidermal necrolysis in all test groups.
The severity of cutaneous lesions varied
from very slight to severe at all lest sites
and at all dose levels.
a Undiluted unleaded gasoline.
-------
agitation, somnolence, loss of appetite, vomiting, and occasional intestinal hemorrhage, but
no mortality. Details regarding the species of animal, concentration of gasoline vapors, and
the duration of exposure were not specified by Browning (1953). Haggard (1921)
reported acute anesthetic and toxic effects in dogs resulting from inhalation of gasoline
vapors. Central nervous system effects were observed at 10,000 ppm (1 percent) and
death occurred at 25,000 ppm (2.5 percent). In a study by Gerarde (1963), two male
albino Wistar rats receiving 0.2 ml gasoline intratracheally died almost immediately. The
low viscosity of gasoline was considered to be the property causing the highly toxic
reaction.
Hematotoxicitv
A single intraperitoneal injection of approximately 1 mg/kg of gasoline in female
Swiss Wistar albino rats significantly decreased delta - aminolevulinic acid synthetase and
dehydratase levels after 20 hours.
Dermal Toxicitv
Elars Bioresearch Laboratories (1979,1980) tested unleaded gasoline (PS-6) for
potential acute toxicity using the following battery of tests: primary skin irritation, primary
eye irritation, skin sensitization, acute oral toxicity, acute dermal toxicity, and subacute
dermal toxicity. Table 7-2 presents a summary of the range of doses and responses in the
species tested. The test material exhibited slight irritating primary skin effects; however, no
adverse effects were noted in the primary eye irritation and skin sensitization tests. An oral
LD50 of 18.75 mlAg was found in rats. Acute dermal tests in rabbits showed that the test
material produced skin irritation and may have caused one death. The compound caused
acute dermal corrosion and slight systemic toxicity in rabbits.
Dermal toxicity was observed in New Zealand white rabbits exposed to unleaded
gasoline (PS-6) (Elars Bioresearch Laboratories, 1980). Six tested animals (three males
and three females) were shaved in four areas; two sites were abraded and two sites were
left intact. Plastic-backed gauze patches containing about 0.5 grams of the gasoline were
applied to each of these areas for 24 hours. Edema was observed in 4 of the 6 animals at
24 hours and was still present on the seventh day following exposure. These four animals
also developed erythema 72 hours after exposure. All erythema and edema were gone by
the fourteenth day post-exposure, although no hair had grown back in the sites. No
differences were observed between the abraded and unabraded sites.
Skin sensitization test were performed in male albino guinea pigs (Elars Bioresearch
Laboratories, 1979). Gauze patches containing 0.5 ml of PS-6 unleaded gasoline were
7-11
-------
applied to the animals for 6 hours, and the procedure was repeated three times per week for
3 weeks. After a 2-week rest, a challenge dose was given in the same manner. A positive
control group, exposed to dinitrochlorobenzene in the same manner as the test group, was
also assayed. No sensitization effects on either the test group or the positive control group
were observed. The lack of response in the positive control group limits the conclusions
that can be drawn from this study.
Liver Toxicity
Animal studies provide evidence that gasoline alters hepatic metabolism. Adult rats
given a single intraperitoneal injection of approximately 2 g/kg gasoline exhibited increased
liver lipid peroxidation (28 percent over controls) and increased liver alkaline phosphatase
activity 24 hours after injection (Rao and Pandya, 1978). The authors postulated that the
highly reactive epoxide metabolites of the hydrocarbons oxidized polyunsaturated fatty
acids in the cell membrane. The lipid peroxides generated from this reaction in turn altered
microsomal enzyme activity.
7.1.2 Effects of Subacute and Subchronic Exposure to Gasoline
7.1.2.1. Human Studies
No data was found in the available literature on the protracted exposure of humans
to gasoline vapors or gasoline-contaminated water.
7.1.2.2 Animal Studies
Several studies are available on the effects of subchronic exposure to gasoline
vapors via inhalation. Only one study was found on the subchronic exposure of animals to
liquid gasoline by the oral route. No subchronic dermal studies were reported. As
summarized in Table 7-3, these studies indicate the following.
• Oral dosing of liquid gasoline or inhalation exposure to wholly vaporized
gasoline induced nephropathy in male rats. Male rats appear to be particularly
sensitive to this effect
• Inhalation exposure to wholly vaporized leaded gasoline induces histological
changes in the lung parenchyma and decreased secretion of pulmonary
surfactant.
7-12
-------
TABLE 7-3
SUBCHRONIC TOXICITY OF LEADED AND UNLEADED GASOLINE
Animal
species
Dose concentration
Route of
administration
Exposure variables
Effects
Reference
o
Rat(Sprague- 1, 1.57. or 6.35 mg/L Inhalation
Dawley) (0,384. or 1.552 ppm);
total gasoline vapors,
unleaded EPA reference
fuel
Rat (Fischer- 0. 0.5. or 2.0 g/kg; Oral
344) API PS-6. unleaded
fuel gasoline
Unleaded Gasoline
20 males (M). 20 females
(F). 6 hours/day. 5 days/
week for 13 weeks at each
dose level.
73 M and 72 F; oral
gavage. once daily
7 days/week for a
28-day period.
At the high-dose level, kidney
hislopaihology revealed regenerative
epithelium and dilated tubules in the
kidney of males only. Other changes
included elevation of thrombocyte
counts in males and reliculocyte
counts in females. At the low dose,
male rats had increased liver weights.
Kuna and Ulrich,
1984; API. 1976.
Histopathological examination of
kidneys revealed a treatment-related
nephropaihy in males only, occurring as
early as 8 days at both dose levels. Renal
lesions included hyaline droplets in the
proximal tubules, increased foci of regenerative
epithelium, and dilated tubules at the
corticomedullary junction. Electron microscopy
revealed degenerative changes as early as day 4.
It was suggested that the observed serum protein
changes in treated males correlated with the
observed nephropathy. Likewise, increased
serum crcatinine and decreased creaiinine
clearance in high-dose males correlated
with the findings for severe nephropathy.
Female rats did not show nephropaihy.
Borriston. 1985.
-------
TABLE 7-3
(continued)
Animal
species
Dose concentration
Route of
administration
Exposure variables
Effects
Reference
Rat (Sprague-
Dawley)
Rat (Sprague-
Dawley)
0,0.11, 1,58, or
12.61 mg/L(29,416,
or 3,316 ppm by volume)
wholly vaporized unleaded
gasoline API PS-6
Inhalation
0,0.15, 1.44, or 14.70
mg/L (40, 379, or
3,866 ppm by volume);
totally vaporized
unleaded gasoline
API PS-6
Inhalation
Unleaded Gasoline (continued!
10 M and 10 F; 6
hours/day, 5 days/
week for 21 days (total
15 exposures) at each
dose level. Two types
of control: one group
exposed to filtered air
only (FAC) and another
group, the environmental
control (EC), was simply
maintained in the animal
room.
20 M and 20 F; 6 hours/
day. 5 days/week, for 90
days (total 65 exposures)
at each dose level. Half
of the rats of each sex/
group were sacrificed
following the last
exposure; the remaining
were retained for a 4-week
recovery period and then
sacrificed.
Renal lesions included increased levels
of hyaline droplet formation in cells
of the proximal tubules and mild
tubular degenerative and regenerative
changes in the treated male rats;
corticomedullary tubular dilation
and necrosis were observed in one
male rat of the high exposure group.
Haider etal.. 1984.
Renal tubular dilation and necrosis
at the corticomedullary junction was
observed in rats only. The effect was
dose related. Tubular necrosis was
observed in male rats both at the
terminal sacrifice and following a
4-week recovery period. The similarity
in the incidence and severity of the
lesions suggested that the nephrotoxic
effects may be irreversible.
Haider etal., 1984.
-------
TABLE 7-3
(continued)
Animal
species
Dose concentration
Route of
administration
Exposure variables
Effects
Reference
Unleaded Gasoline (continued)
Squirrel 0,0.157 or 6.35 mg/L
monkey (0, 384, or 1,552 ppm)
total gasoline vapor,
unleaded EPA reference
fuel
Inhalation
4 M, 4 F, 6 hours/day,
5 days/week for 13
weeks at each dose
level.
At the high-dose level, pulmonary
function data indicated significant
elevation in minute volume for males.
At both dose levels, no significant
differences in hematological or
urinalysis parameters were noted.
There was no evidence of treatment-
related histopalhology.
Kuna and Ulrich,
1984; API. 1976.
Rat (Sprague-
Dawley)
0.0.42. or 1.53 m/L
(0.103. or 374 ppm)
total gasoline vapor.
containing 1.94 g
lead/gasoline
Leaded Gasoline
Inhalation 20 M. 20 F. 6 hours/day.
5 days/week for 13 weeks
at each dose level.
Renal lesions in male rats, dose
level unspecified. Hematological
data indicated decreased mean
corpuscular hemaglobin concentration
in males and increased hemaiocrit and
mean corpuscular volume and decreased WBC
count in females at the high-dose level.
Tissue samples showed, in descending
order, increased levels of lead in liver,
kidney, brain, and blood. Urine lead
levels were similar in all groups.
Kuna and Ulrich,
1984; API, 1976.
-------
TABLE 7-3
(continued)
Animal
species
Dose concentration
Route of
administration
Exposure variables
Effects
Reference
Leaded Gasoline (continued)
Squirrel
monkey
-vl
.L. Rat (Wistar)
o\
0,0.42, or 1.53 mg/L Inhalation
(0, 103. or 374 ppm)
total gasoline vapor,
containing 1.94 g
lead/gallon
100 ppm "Super" Inhalation
grade (octane rating
98%), tetraethyl
lead concentration
0.45g/L(w/v)
4 M, 4 F, 6 hours/day
5 days/week for 13
weeks at each dose
level.
40 M, 8 hours/day
5 day/week, for up
to 12 weeks.
At the high-dose level, pulmonary
function data indicated increased
minute volume in male and decreased
tidal volume for females. No effect
on visual evoked response (a test of
neuroloxicily). No effect on hcmaiology
or urinalysis parameters. No evidence
of treatment-related histopalhology.
Kuna and Ulrich,
1984; API, 1976.
None of the animals exposed for less
than 6 weeks showed unequivocal
evidence of structural damage. Of
28 rats exposed for 6 to 12 weeks, 22
exhibited lung pathology ranging from
minor scattered foci to widespread scleroris.
Hypertrophy and hyperplasia of type 2
pheumocytes (cells that produce lung
surfactant) was seen in the early periods
of exposure, but their numbers decreased
in lungs with more marked focal sclerosis.
Lykke and Stewart,
1978.
-------
TABLE 7-3
(continued)
Animal
species
Dose concentration
Route of
administration Exposure variables
Effects
Reference
Leaded Gasoline (continued^
Rat (strain
unspecified)
I
^J
Rat (Wistar)
100 ppm "Super"
grade (octane rating
98%), tetraelhyl lead
concentration
0.45 g/L (w/v)
Inhalation
100 ppm "Super"
grade (octane rating
98%). tetraelhyl lead
concentration 0.45 g/L
(w/v)
Inhalation
An unspecified number
of "batches" of 20F were
exposed for 8 hours/
day, 5 days/week for
up to 12 weeks.
An unspecified number
of female rats were
exposed for 8 hours/
day. 5 days/week for
up to 45 days.
High incidence (frequency unspecified) Lykke el al., 1979.
of ultrastructural damage to lung
parenchyma. Initial changes appeared
more frequently after the 6lh week of
exposure and included degeneration of
the endothelium and interstitial fibroblasts.
Between 6 to 10 weeks, there was
hypertrophy of type 2 pneumocyles.
Finally, at 9 to 12 weeks of exposure.
irregular foci of fibrosis became more
frequent and were associated with
alveolar distortion and collapse.
Five days after initiation of exposure,
lung surfactant was markedly decreased
and reached lowest level (3- to 4-fold
reduction from controls) at IS days
of exposure. After transient increase
(correlating with hypertrophy and/or
hyperplasia of type 2 pneumocyles), it
decreased to constant level, about half
the mean value of controls.
LeMesurier et al.,
1979.
-------
Kidnev Toxicitv
The temporal and morphologic aspects of hydrocarbon-induced nephropathy in
male rats have been studied by Thomas et al. (1984a) using unleaded gasoline administered
daily, by gavage, to 9-week-old male and female Fischer 344 rats. The gasoline was
administered at dose levels of 0.5 or 2.0 g/kg with concurrent controls receiving isotonic
saline at the highest volume (2.0 g/kg). Study observations were performed at days 0,1,
3,7,14,21, and 28 after dosing. While clinical laboratory indices of kidney function were
not affected, gross kidney lesions were found in one or two of the high-dose male rats
sacrificed at days 3,7, and 28. Light microscopic observations revealed lesions in males
only, with no significant differences between dose levels. The following sequence of
histologic findings was observed in kidney secretions from the male rats.
• Substantial accumulation of hyaline droplets in the epithelial cells of the proximal
kidney tubules was observed after only 1 day of exposure and was seen after all
other exposure periods.
• Basophilic foci of regenerative tubular epithelial cells were seen with increased
frequency at 14 days of exposure and thereafter.
• Dilation of the proximal tubules at the corticomedullary junction and occlusion of
the lumen with granular-appearing cellular debris were seen at day 14 and
thereafter.
In a series of subchronic studies, Haider et al. (1984) exposed Sprague-Dawley rats
(135 grams) via inhalation for 21 and 90 days to seven petroleum naphtha streams and an
unleaded gasoline blend (see Table 7-4). Inhalation exposures were 6 hours/day, 5
days/week. The petroleum naphtha streams varied from light straight-run naphtha to heavy
catalytic-reformed naphtha. The test materials were vaporized by flash evaporation. Thus,
virtually all components were vaporized.
The dose levels used in the 21-day unleaded gasoline study were 29,416, and
3,316 ppm; for the 90-day study, the dose levels were 40, 379, and 3,866 ppm.
Nephrotoxic effects (hyaline droplets and cortical tubular degeneration) were seen at all
dose levels in male rats in the 21-day study (see Table 7-5). Conicomedullary tubular
dilation and necrosis were also observed in one male rat in the high exposure group. Thus,
29 ppm represents a lowest observed adverse effect level (LOAEL) for nephrotoxic effects
7-18
-------
TABLE 7-4
SUMMARY OF THE COMPOSITION AND BOILING RANGES OF THE
TEST MATERIALS USED IN HALDER ET AL. (1984)
Composition (%)
Material
Light Straight-Run
Naphthab
Light Catalytic-Cracked
Naphtha0
Light Catalytic -Reformed
Naphthad
Heavy Catalytic-Reformed
Naphtha6
Full-Range Alkylate
Naphtha
Polymerization Naphtha
Thermal-Cracked Naphtha
Unleaded Gasoline Blend
Alkanesa
96
39
67
7
98
8
58
45*
Alkenes
0
32
2
0
2
92
30
12*
Aromatics
4
29
31
93
0
1
12
43*
Boiling Range
°C
10% bp
71
174
137
290
124
205
112
112
90% bp
222
346
230
364
315
353
354
326
a Includes normal, branched, and cycloalkanes
b Predominantly branched chain saturated hydrocarbons from Oj to ClQ and boiling range
90 to 160°C
c Predominantly branched chain saturated hydrocarbons from C4 through Ci 1 and boiling
range -20 to 190°C, contains large proportion of unsaturated HCs
d Predominantly branched chain saturated hydrocarbons from Cg to C\2 and boiling range
65 to 230°C, contains large proportion of unsaturated HC
e Predominantly branched chain saturated hydrocarbons from Cf to Ci i and boiling range
35 to 190°C, stream may contain 10 volume percent or more benzene
f Estimated
7-19
-------
TABLE 7-5
NEPHROTOXIC EFFECTS IN MALE RATS FOLLOWING A 21-DAY
INHALATION EXPOSURE TO AN UNLEADED GASOLINE BLEND
Incidence
Renal Lesions
Hyalin Droplets
Cortical Tubular
Degeneration
Cortical Tubular
Regeneration
EC
NAb
NA
NA
FAC
0/10
0/10
0/10
O.lia
(29)
5/10
4/10
1/10
1.58a
(416)
8/10
6/10
0/10
12.61a
(3,316)
10/10
10/10
6/10
Corticomedullary
Tubular Necrosis
and Dilatation NA 0/10 0/10 0/10 1/10
a Analytical time weighted average in mg/L (ppm by volume)
b NA - Not available. These tissues were not examined.
EC - Environmental control group
FAC - Filtered air control group
SOURCE: Haider et al., 1984.
7-20
-------
in male rats in a 21-day subchronic study. No nephrotoxic changes were observed in
female rats.
The authors performed experiments on different fractions of the unleaded gasoline
in an effort to determine which fractions of the mixture were most toxic to the kidney
(Haider et al., 1984). They found that the full-range alkylate naphtha (consisting mostly of
paraffins) was the most toxic (see Table 7-6), followed by the thermal-cracked naphtha
(which consisted of paraffins, olefins, and aromatics in an approximately 6:3:1 ratio, Table
7-7). By contrast, the heavy catalytic-reformed naphtha (which consists mostly of
aromatics) showed the least toxicity (see Table 7-8). For the full range alkylate naphtha,
the authors determined that the 34 ppm exposure represented the LOAEL and that the 3
ppm exposure represented the no observed adverse effect level (NOAEL). This finding is
consistent with the finding that the 29 ppm exposure represented the LOAEL for unleaded
gasoline vapor exposure. The exposure levels used in the thermal-cracked naphtha
experiment were too high to determine a NOAEL. The results with the various types of
naphtha streams and unleaded gasoline strongly suggest that the streams having normal or
branched alkanes are more nephrotoxic than streams containing primarily alkenes or
aromatics.
In the 90-day Haider et al. study of unleaded gasoline, nephrotoxicity was also
observed, again only in male rats (see Table 7-9). A dose-related increase in tubular
dilation and necrosis at the corticomedullary junction was observed in male rats both at the
terminal sacrifice and following a 4-week recovery period, suggesting that the nephrotoxic
effects may be irreversible. No nephrotoxic effects were seen in either the environmental
controls or the filtered air chamber controls. The incidences of nephrotoxic effects at
terminal sacrifice were 1/10,7/10, and 5/10 for 40,379, and 3,866 ppm exposures,
respectively.
Because the single case of nephrotoxicity at 40 ppm was not statistically significant,
it is possible to consider the 40 ppm as a NOAEL and the 379 ppm exposure as the
LOAEL. There are several qualitative factors to consider, however, before making this
conclusion. First, under the conditions of the experiment, no nephropathic lesions were
found in the female rats or in either the filtered air control or the experimental control male
rats. Second, the finding of nephropathy in a single male rat in the 40 ppm exposure group
is consistent with the dose-response pattern observed at higher doses. This finding raises
the possibility that larger sample sizes could produce a statistically significant response in
this exposure group. Third, the single incidence of nephropathy (as well as all other
incidences) remained significant after a one-month recovery. This finding is important in
that it is a potentially irreversible lesion produced by a subchronic exposure.
7-21
-------
TABLE 7-6
NEPHROTOXIC EFFECTS IN MALE RATS FOLLOWING A
REPEAT 21-DAY INHALATION EXPOSURE TO FULL-RANGE
ALKYLATE NAPHTHA
Incidence
Renal Lesions
Hyalin Droplets
Conical Tubular Degradation
Cortical Tubular Regeneration
Corticomedullary Tubular
Necrosis and Dilatation
FAC
0/40
0/40
2/40
0/40
0.015*
(3)
1/20
0/20
3/20
0/20
0.1 52a
(34)
12/20
9/20
5/20
4/20
1.538*
(345)
20/20
20/20
17/20
11/20
a Analytical time weighted average in mg/L (ppm by volume)
FAC - Filtered air control group
SOURCE: Haider et al., 1984.
7-22
-------
TABLE 7-7
NEPHROTOXIC EFFECTS IN MALE RATS FOLLOWING A 21-DAY
INHALATION EXPOSURE TO THERMAL-CRACKED NAPHTHA
Renal Lesions
Hyalin Droplets
Conical Tubular
Degeneration
Cortical Tubular
Regeneration
EC
0/10
0/10
2/10
FAC
0/10
1/10
2/10
Incidence
1.13*
(230)
10/10
9/10
6/10
3.48*
(709)
10/10
10/10
5/10
9.88*
(2,014)
9/10
10/10
8/10
Corticomedullary
Tubular Necrosis
and Dilatation
0/10
0/10
3/10
4/10
7/10
a Analytical time weighted average in mg/L (ppm by volume)
EC - Environmental control group
FAC - Filtered air control group
SOURCE: Haider et ah, 1984.
7-23
-------
TABLE 7-8
NEPHROTOXIC EFFECTS IN MALE RATS FOLLOWING
A 21-DAY INHALATION EXPOSURE TO HEAVY CATALYTIC-
REFORMED NAPHTHA
Incidence
Renal Lesions
Hyalin Droplets
Cortical Tubular
Degeneration
Cortical Tubular
Regeneration
EC
0/10
1/10
2/10
1.03*
FAC (215)
NAb NA
NA NA
NA NA
2.8 la
(587)
0/10
0/10
2/10
11.20a
(2,132)
0/10
1/10
0/10
Corticomedullaiy
Tubular Necrosis
and Dilatation
0/10
NA
NA
0/10
0/10
a Analytical time weighted average in mg/L (pprn by volume)
b NA - Not available. Pathology was not done due to lack of adverse effects.
EC - Environmental control group
FAC - Filtered air control group
SOURCE: Haider et al., 1984.
7-24
-------
TABLE 7-9
NEPHROTOXIC EFFECTS IN RATS FOLLOWING A 90-DAY
INHALATION EXPOSURE TO AN UNLEADED GASOLINE BLEND
Incidence^
Group
and Concentration2
Environmental Control
Filtered Air Control
0.15 mg/liter (40 ppm)
1.44mg/liter(379ppm)
14/70 mg/liter (3866 ppm)
Terminal Sacrifice
M F
0/10
0/10
1/10
7/10
5/10
0/10
0/10
0/10
0/10
0/10
One-Month Recovery
M F
0/10
0/10
1/10
5/10
4/10
0/10
0/10
0/10
0/10
0/10
a Analytical time weighted average in mg/L (ppm by volume)
b Incidence of tubular dilatation and necrosis at corticomedullary junction.
SOURCE: Haider et al., 1984.
7-25
-------
In efforts to further identify the components in gasoline that are primarily
responsible for inducing nephropathy in male rats, Haider et al. (1985) and Borriston
Laboratories (1984) performed a series of 28-day gavage screening studies on gasoline
components. The following materials were tested: pure hydrocarbon components typically
found in gasoline, various distillation fractions of a typical gasoline blend, and
representative naphtha streams commonly used to blend gasolines. As controls, isotonic
saline and a reference unleaded gasoline blend (PS-6) were used. Groups of 10 young
adult male Fischer 344 rats were administered the test materials, by gavage, at doses of 0.5
or 2.0 g/kg, once daily, 5 days/week for 4 weeks (a total of 20 doses). The animals were
evaluated for mortality, clinical signs, terminal body weight, gross pathology, kidney
histopathology, and weight Due to the large number of materials being tested, the trials
were conducted in three phases. In each phase, a specific group of materials was tested,
together with the saline and unleaded gasoline controls.
Gross pathologic findings were confined mostly to the kidneys, stomach, and liver.
Stomach lesions were attributed to the irritant effect of the hydrocarbons. Gross kidney
and liver lesions generally consisted of discoloration and mottling and were attributed to
postmortem changes. Kidney weights were not correlated with nephrotoxicity. Significant
elevations in kidney weights were observed only for 2,3-dimethylbutane and for 2,2,5-
trimethylhexane. Significant decreases in kidney weights were observed only for unleaded
gasoline (only in phase 2) and for n-pentane and trans-2-pentene. No significant
differences from saline controls were observed in the other test groups.
Histopathologic analysis of the kidneys indicated that unleaded gasoline and a
number of its components were nephrotoxic. The lesions were scored according to the
scheme summarized in Table 7-10. These nephrotoxic compounds and mixtures are listed
in Table 7-11 and are ranked in decreasing order of their capacity to produce nephropathy,
reflected by their nephropathy scores. The materials producing a nephrotoxic response
were primarily branched alkanes. Within an alkane class, nephrotoxicity increased with the
degree of branching. The results obtained with the normal alkanes (plus 2-methyl butane),
the alkanes, and the aromatic compounds were not significantly different from those
obtained with saline controls. The distillation fractions and naphtha streams were all
nephrotoxic in varying degrees. Except for the light and heavy catalytic-cracked naphthas,
their nephrotoxicity correlated well with their content of branched alkanes. In the case of
the cracked naphthas, unidentified alkanes may have also contributed to their nephrotoxic
activity, in conjunction with the branched alkanes.
7-26
-------
TABLE 7-10
GRADING SYSTEM USED TO EVALUATE THE SEVERITY
OF NEPHROTOXIC RESPONSES
Characteristic lesion Grade (score) criteria
Hyaline droplet (phagolysosome) 1 ~ Minimal
accumulation 2 - Slight
3 - Moderate
4 ~ Severe
Regenerative epithelium 1 — 1 to 4 foci
2 - 5 to 8 foci
3 - 12 to 12 foci
4 -- >12 foci
Intratubular cast formation 1 — 1 to 3 foci
2 - 4 to 6 foci
3 - 7 to 12 foci
4-->12foci
a The changes were graded on a scale of 1 to 4 based on severity or number of pathologic
foci, as applicable. The scores for each lesion type were combined and then multiplied
by 10.
SOURCE: Adapted from Haider etal., 1985.
7-27
-------
TABLE 7-11
AVERAGE NEPHROTOXICITY SCORES OF TESTED GASOLINE
COMPONENTS
(Low-Dose Score/ High-Dose Score)
Tested material
Average nephropathy scores3
Phase 1 Phase 2 Phase 3
2,2,5-Trimethylhexane 94/96b
2,3-Dimethylbutane 77/73
2-Methylpentane 49/65
Methylcyclopentane c/36
2,2,4-Trimethylpentane
2,3-Dimethylpentane
2-Methylhexane
Light alkylate naphtha
220-280°F Distillation fraction
145-220°F Distillation fraction
Light catalytic-cracked naphtha
Heavy catalytic-cracked naphtha
Light catalytic-reformed naphtha
280°F-End distillation fraction
Reference unleaded gasoline 92/91
Saline (negative control) 30
80/97
63/67
47/38
64/55
39
101/94
99/104
60/89
37/53
36/34
c/39
42/c
84/83
27
a See Table 7-10 for explanation of scores.
b Represents low-dose/high-dose scores.
c Not statistically higher than saline controls.
SOURCE: Adapted from Haider et al., 1985.
7-28
-------
At present, the mechanism of nephrotoxicity of gasoline components and/or their
metabolites in male rats has not been elucidated; however, several lines of inquiry using
2,2,4-trimethylpentane, reviewed below, are being pursued.
• Charbonneau et al. (1987a) studied the nephrotoxicity of 2,2,4-trimethylpentane
(TMP) and some of its putative metabolites in male Fischer 344 rats. The rats
were given a single oral dose (4.4 mmol/kg) of each of the following
compounds: TMP, 2,2,4-trimethyl-l-pentanol, 2,2,4-trimethyl-I-pentanoic acid
(2,2,4-TMP acid), 2,4,4-trimethyl-l-pentanol, 2,4,4-trimethyl-l-pentanoI acid,
or2,4,4-trimethyl-2-hydroxy-l-pentanoic acid. Nephrotoxicity was assessed by
measurement of the following parameters: increase in renal droplets (hyaline
droplets), renal alpha-2-microglobulin concentration, and renal cell proliferation.
All compounds tested led to an increase in protein droplet accumulation and renal
alpha-2-microglobulin concentration. Only with exposure to TMP and 2,2,4-
TMP acid was renal cell proliferation observed.
• In other studies, Lock et al. (1987b) observed that when Fischer 344 rats were
dosed orally with 3H-TMP (500 mg/kg), radioactivity from TMP became
reversibly associated with alpha-2-microglobulin (a protein specific to male rats)
from the kidney cytosol of male rats. Female rats showed no such effect.
• In further work, Charbonneau et al. (1987a, b) identified 2,4,4-trimethyl-2-
pentanol as the metabolite reversibly bound to alpha-2-microglobulin in the
kidney cytosol of male Fischer 344 rats.
• As a result of in vitro studies with kidney slices from male rats, Lock et al.
(1987a) has reported that 2,4,4-TMP acid is a substrate for the renal organic
anion transport system, suggesting that carrier-mediated transport of TMP-acid
metabolites may contribute to the nephrotoxicity of TMP.
In previous work as reviewed above, Haider et al. (1984,1985) were able to
induce kidney nephropathy in male rats exposed to totally vaporized samples of whole
gasoline, pure branched alkanes, or gasoline fractions rich in these compounds. These
exposures, however, were deemed by the authors to be poorly representative of exposure
to gasoline vapors in the occupational setting where exposure is predominantly to the more
volatile components of gasoline. In particular, Haider et al. (1986a), in a survey of several
7-29
-------
work sites, reported that four C4/C5 gasoline components (n-butane, isobutane, n-pentane,
and isopentane) comprised 90 to 92 percent of all the C4/C5 vapor components and about
61 to 67 percent by weight of the total vapors. Thus, based on what was considered to be
a more realistic assessment of the occupational exposure, a series of inhalation studies
(reviewed below) were undertaken to assess the nephrotoxic potential of mixtures of C4/C5
components of gasoline and of a low boiling unleaded gasoline distillation fraction. While
these fractions may be more representative of ambient gasoline exposures in and around
service stations, actual exposures to gasoline or gasoline vapors in residential environments
may be quite different
In one series of studies, Haider et al. (1986b) evaluated the nephrotoxic potential of
a blend consisting of 25 percent (w/w) each of n-butane, n-pentane, isobutane, and
isopentane in Sprague-Dawley rats. Three groups of rats (10 males, 10 females per group)
were exposed by inhalation to total vapors of the hydrocarbon blend at levels of 0.12 mg/L
(44 ppm), 1.15 mg/L (432 ppm), and 11.80 mg/L (4,437 ppm); a fourth group served as
the control and was exposed to filtered air only. Exposure was for 6 hours/day, 5
days/week for 3 weeks, for a total of 15 exposures. The animals were monitored daily for
mortality, morbidity, and adverse clinical signs. At study termination, all animals were
necropsied with the brain, heart, liver, spleen, kidneys, adrenals, and gonads weighed
before fixation. Histopathologic analysis of the above tissues was conducted for the
control animals and the high-dose group. For the two lower dose groups, only kidney
histopathology was evaluated.
No treatment-related pathologic lesions were noted upon either gross or
microscopic examination of tissues. In particular, no evidence of nephrotoxicity was noted
at any of the dose levels tested. No treatment-related effects were observed in body and
organ weights, hematology, or serum chemistry.
The IFT Research Institute (1985) conducted a subchronic inhalation toxicity study
of the 0 to 145°F gasoline distillate fraction. Male and female Fischer 344 rats were
exposed to the gasoline distillate at 4,500 and 1,000 ppm for 6 hours/day, 5 days/week for
13 weeks. The kidneys did not reveal any hydrocarbon nephropathy in the rats, nor did
they contain significant lesions when compared to the filtered air control rats.
In another series of studies, Aranyi et al. (1986) and Aranyi (1984,1985) assessed,
in Fischer 344 rats, the nephrotoxic potential of two alkane mixtures and of a distillation
cut, boiling below 145°F, of an unleaded gasoline blend. The two alkane mixtures were
50/50 weight-percent mixtures of n-butane/n-pentane and isobutane/isopentane,
respectively. The distillation cut was deemed to approximate reasonably well the
composition of gasoline vapors observed in occupational settings. For each test material.
7-30
-------
groups of 20 male and 10 female rats were exposed by inhalation to two concentrations
indicated below, 6 hours/day, 5 days/week, for up to 91 days (66 exposures).
Test materials Time-Weighted Average Concentration (ppm)
High Low
n-Butane/n-pentane (50%/50%)
Isobutane/isopentane (50%/50%)
0-145°F Distillation cut
4,498
4,458
5,229
1,017
996
1,209
a Controls were exposed to filtered air.
Necropsies were performed on half of the male rats from each treatment group after
the 10th exposure (day 28) and on the remaining animals at the end of the study. Major
tissues were collected and fixed. Organ weights were determined only for kidney and
liver, and only kidneys were examined microscopically. Except for a slight but statistically
significant increase in relative kidney weights in female rats treated with the distillation cut,
no statistically significant difference in kidney or liver weights was observed for any of the
treatment groups. Histologic assessment of nephrotoxicity at the end of the exposure
period, using the scoring procedure shown in Table 7-10, produced nephropathy scores
that were not significantly different from controls for any of the treatment groups.
However, determination of nephropathy scores at the 28th day interim sacrifice revealed a
significantly elevated nephropathy score in males exposed to the low dose of the
isobutane/isopentane mixture. The significance of this transient elevation of the
nephropathy score is not known.
Although the Aranyi studies detected only mild kidney effects at relatively high
gasoline exposure concentrations, the finding of a kidney response only in the female rats
differs from findings of other investigators. Usually, kidney effects have been seen in only
male rats (cf., the Haider studies). This finding thus raises the possibility that apparent sex
specificity in the kidney responses to gasoline vapor exposure may be significantly
influenced by the exposure mixture and the test protocol.
7-31
-------
Pulmonary Toxicitv
Kuna and Ulrich (1984) exposed Sprague-Dawley rats and squirrel monkeys to
leaded and unleaded gasoline vapors at concentrations of 0,0.42, or 1.53 mg/L (0,103, or
374 ppm) and 0,1.57, or 6.35 mg/L (0, 384, or 1,552 ppm), respectively. The exposures
were for 6 hours/day, 5 days/week for 13 weeks. At the high-dose level for both fuels,
pulmonary function data indicated increased minute volumes in the male monkeys (four per
group). Female monkeys (four per group) exposed to the high concentration of leaded fuel
exhibited a significant decrease in tidal volume, while the high-level unleaded females
exhibited significant decreases in respiratory rate. Statistically significant changes in organ
weights were noted for both fuels in both species, but these were not corroborated by
histopathologic changes. The only other change observed in this study was histopathologic
evidence of regenerative epithelium and dilated tubules in the kidney of the unleaded fuel
high-level male rats. This was a well-conducted study, despite the fact that no consistent
patterns of lexicological significance were established.
Lykke and Stewart (1978) demonstrated that subchronic exposure of rats to
commercial leaded gasoline via inhalation can induce pulmonary toxicity. Rats (40 males,
strain not specified) were exposed to 100 ppm "Super" grade leaded gasoline (98 percent
octane rating) 8 hours/day, 5 days/week, for up to 12 weeks. None of the animals exposed
for less than 6 weeks showed unequivocal evidence of structural damage. Of those rats
exposed for 6 to 12 weeks, 22 of 28 exhibited lung pathology ranging from minor scattered
foci to widespread sclerosis. Hypertrophy and hyperplasia of type 2 pneumocytes (lung
surfactant-producing cells) was seen in the early periods of exposure, but their numbers
decreased in lungs with more marked focal sclerosis. There were no lung changes evident
in the control animals. It is difficult to assess the quality of this study since many details of
the experimental protocol were not given.
Lykke et al. (1979) further characterized the pulmonary changes associated with
subchronic inhalation of whole leaded gasoline vapors. An unspecified number of
"batches" of 20 female rats were exposed for 8 hours/day, 5 days/week, for up to 12
weeks to 100 ppm "Super" grade (octane rating 98 percent) whole leaded gasoline vapors.
A high incidence (frequency unspecified) of ultrastructural damage to lung parenchyma was
observed in the exposed animals. Initial changes appeared more frequently after the sixth
week of exposure and included degeneration of the endothelium and interstitial fibrosis.
Between 6 and 10 weeks, there was hypertrophy of type 2 pneumocytes. Finally, at 9 to
12 weeks of exposure, irregular foci of fibrosis became more frequent and were associated
with alveolar distortion and collapse.
7-32
-------
Further studies conducted by this laboratory (LeMesurier et al., 1980; LeMesurier
et al., 1979) showed that surfactant yield in rats was significantly affected by gasoline
exposure. These studies also provided evidence that changes in pulmonary surfactant
levels signified early and sensitive indicators of ultrastructural changes in type 2
pneumocytes. These ultrastructural changes were also correlated with alveolar collapse and
the pathogenesis of fibrosing alveolites in the lower lung.
Female Wistar rats were exposed to 100 ppm of leaded gasoline vapor for 8
hours/day, 5 days/week, for up to 45 days. Five rats were sacrificed at intervals of 5 days
following commencement of gasoline exposure. Surfactant levels dropped by half (21 mg
to 10 mg) during the first 5 days of exposure. At 15 days, surfactant levels were at 6 mg.
This was followed by a steady increase in surfactant levels through 30 days of exposure,
when surfactant levels reached about 14 mg, followed by a drop to about 9 mg for the
remainder of exposure.
The investigators observed that the initial drop in surfactant levels seemed to
correlate with the morphological evidence of hypertrophy and hyperplasia in the type 2
pneumocytes (LeMesurier et al., 1979). In the subsequent recovery phase, the
pneumocytes showed large numbers of dense osmiophilic surfactant lamellae (LeMesurier
et al., 1979). The last phase correlated with intracytoplasmic degenerative changes in the
pneumocytes, characterized by vacuolation and poor definition of the cytoplasmic surface
lamellae, and with evidence of irregular alveolar collapse as well as emphysematous
changes and fibrosis (LeMesurier et al., 1979).
Surfactant is essential for normal lung function in adult and newborn animals, as it
prevents the collapse of the alveoli upon expiration. Deficiencies in pulmonary surfactant
have been found in infants who died of respiratory distress syndrome (RDS) (Rooney,
1984). RDS is the leading cause of morbidity and mortality in prematurely bom infants,
accounting for about 20,000 deaths per year in the United States. The syndrome results
from the atelectasis caused by collapsing alveoli, and the infant becomes cyanotic and
hypoxic. As RDS is a developmental disorder caused by immature lungs, control of
surfactant synthesis has been studied as a key element in fetal lung maturation (Rooney,
1984). Surfactant may also be important in the control of the immune response to
mitogenic stimulation within the lung, thereby preventing inflammatory reactions from
occurring in areas of the lung required for gas exchange (Anstield et al., 1979).
The surfactant system may be particularly susceptible to pulmonary toxicants, either
through direct effects on the surfactant system or as a secondary effect caused by damage to
the alveolar pneumocytes (Rooney, 1984). In addition to petroleum distillates, agents
identified as toxic to type 1 and type 2 pneumocytes include nitrogen dioxide, ozone,
7-33
-------
oxygen, cadmium, and metal salts, and drugs such as bleomycin and the
hypercholesteremic agent AY9944 (Shami et al., 1984). Consequently, co-exposure to
agents such as these could augment the toxic effects of gasoline on the respiratory system.
In contrast to the findings of Lykke et al. (1979) and LeMesurier et al. (1979),
MacFarland et al. (1984) found no evidence of progressive focal interstitial fibrosis in rats
exposed to 0,67,292, or 2,056 ppm of wholly vaporized gasoline over a 12 month
period. Rats did exhibit mild, multi-focal pulmonary inflammatory responses that were
characterized by accumulations of alveolar macrophages in the alveolar spaces of the lungs.
The incidence of these macrophage aggregates was 19,5,43, and 38 percent in males, and
40,46, 34, and 67 percent in females in the control, low, medium, and high-dose group,
respectively. There is no easily discemable reason for this difference in the pulmonary
toxicity findings. It is possible that the difference may be due to the different compositions
of gasoline vapor mixtures used in the studies. Support for this possibility comes from
other subchronic studies on rats. In a study by Haider et al. (1985), for example, the 21-
day exposures of up to 6,396 mg/m^ of the heavy catalytic-reformed naphtha (which
consists primarily of aromatic compounds) caused some lung irritation which led to
interstitial pneumonitis and pulmonary edema. A 21-day exposure to a blend of n-butane,
isobutane, n-pentane, and isopentane at concentrations up to 11,800 mg/m^, however,
yielded no pulmonary toxicity (Haider et al., 1986b).
Also, few details were given regarding the control and monitoring of the gasoline
exposures in the Lykke et al. and LeMesurier et al. studies. On the other hand, the
pulmonary surfactant studies provide a sensitive indicator of gasoline-induced lung
damage. Because the lung is a known target organ for gasoline toxicity, these findings
warrant consideration.
Neurotoxicity
Spencer (1982) studied pathological changes in the nervous system of Fischer 344
rats following inhalation of unleaded gasoline motor fuel. Groups of six rats (three/sex)
were exposed via inhalation to 2,056 ppm gasoline, 6 hours/day, 5 days/week, for 8 to 18
months. Histopathological examination of samples of the peripheral nervous system and
spinal cord from a limited number of test animals revealed no distal axonal neuropathy
associated with n-hexane. No functional changes were observed in any animal.
Electroencephalographic studies of unleaded and leaded gasoline intoxication were
conducted in rats (Saito, 1973). Bipolar electrodes were implanted on the brain surface
between the frontal and occipital lobes of the left hemisphere. Rats were divided into two
groups. They were interperitoneally administered 1 ml of either unleaded or leaded
7-34
-------
gasoline (containing 1,000 ppm of tetraethyl lead) per 100 g body weight. The
electrocorticogram (ECoG) was observed for 10 days. The ECoG, during excessive
tension and excitement from the sixth or seventh day after leaded gasoline injection,
showed remarkable increases in the alpha and theta waves. Rats receiving unleaded
gasoline did not show such symptoms, and their ECoGs were very similar to those of the
controls. The author suggested that the striking increases in the alpha and theta waves were
due to a direct action of tetraethyl lead.
Lead, toluene, and n-hexane, the typical components in leaded gasoline, have
produced visual evoked response (VER), suggesting an impairment in the reaction of the
eyes of the exposed animals (Kuna and Ulrich, 1984). Although the use of VER in the
study of neurotoxicity in the central nervous system (CNS) is still being developed, it is a
relatively sensitive non-invasive test to assess the functional capacity of the optic nerve and
the cerebral cortex. Squirrel monkeys exposed to 1,552 ppm unleaded or 374 ppm leaded
gasoline vapors 6 hours/day, 5 days/week, for 90 days showed negative response to the
VER test, suggesting that the fuels, under the conditions of the test, produced no
neurotoxicity in the animals (Kuna and Ulrich, 1984).
Hematoxicitv
Subchronic studies on blood effects have revealed equivocal treatment-related
patterns of toxicity associated with the inhalation of whole gasoline vapors. Kuna and
Ulrich (1984) exposed both Sprague-Dawley rats and squirrel monkeys to leaded and
unleaded gasoline vapors at concentrations of 0,0.42, or 1.53 mg/L (0,103, or 374 ppm);
total leaded gasoline vapor containing 1.94 g lead/gallon; or 0,1.57, or 6.35 mg/L (0,384,
or 1,552 ppm) total unleaded fuel vapor. The animals (20 rats/sex/concentration; four
monkeys/sex/concentration) were exposed 6 hours/day, 5 days/week, for 13 weeks. The
following parameters were measured: body weight; hematology; CNS response;
pulmonary function; urinalysis; deposition of IgG in the renal glomerulus; lead levels in
blood, urine, and tissue; organ weights; organ-to-body weight ratios; and histopathology.
Hematological data indicated decreased mean corpuscular hemoglobin concentration in male
rats, increased hematocrit and mean corpuscular volume in female rats, and decreased white
blood cell count in female rats at the high leaded gasoline level. Hematological changes in
rats exposed to unleaded fuel vapors included elevation of thrombocyte counts in males and
reticulocyte counts in females. No hematological changes were noted in the monkeys. It
should be noted, however, that the sample size was quite small (four monkeys/group). It
is unknown to what extent the lead in the gasoline may have contributed to these effects.
These findings are similar, however, to the hematological effects reported by Sterner
7-35
-------
(1941) for a cohort of painters exposed to vapors released during spray painting with
gasoline diluted paints.
As a concern for the animals dying during the inhalation exposure studies, the
possibility of auto-antibody deposition was investigated in 90-day and lifetime inhalation
toxicity studies (Kuna and Ulrich, 1984; and Wilson, 1983, respectively). (The Wilson
(1983) study is described in the effects of chronic exposure section.) Five groups of rats
and monkeys were exposed to 384 and 1,552 ppm atomized unleaded gasoline vapor or 74
and 103 ppm atomized leaded gasoline vapor, 6 hours/day, 5 days/week, for 90 days.
Renal imrnunofluorescence evaluations of all male and female rats and monkeys displayed
no evidence of IgG deposition in the kidneys following the 90-day exposure period to
gasoline vapors. Since the lungs have also been reported as a possible target organ for
Goodpasture's syndrome by Beime and Brennan (1972, as cited in Kuna and Ulrich,
1984), lung tissues were also studied. However, no evidence of IgG deposition was
found.
Liver Toxicirv
Long-term inhalation exposure to leaded and unleaded gasoline vapors produces
hepatotoxic effects in laboratory animals, with leaded gasoline producing more severe
responses than unleaded gasoline. Soviet studies (cited in U.S. EPA, 1988) noted
gasoline-induced decreases in blood protein, total globulin, and gamma globulin levels.
Leaded gasoline vapors also caused decreased blood albumin and ceruloplasmin levels. In
rabbits exposed to 60 mg/L leaded or unleaded gasoline vapors for 4 hours/day, 6
days/week, for 6 months, decreases were found in blood serum and liver monamine
oxidase activity (Przybylowski et al., 1977, as cited in U.S. EPA, 1988). In a similar
study (Przybylowski et al., 1977, as cited in U.S. EPA, 1988), 60 mg/L inhalation
exposures to leaded or unleaded gasoline produced no changes in serum concentrations of
aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, acid
phosphatase, or ceruloplasmin. It is difficult to interpret these Soviet studies because they
were presented only in abstract form.
7-36
-------
7.1.3 Effects of Chronic Exposure to Gasoline
7.1.3.1 Human Studies
Kidney Toxicity
Mikulski et al. (1972) studied changes in urinary excretion in 51 ship painters
exposed to xylene (84 to 93 percent), toluene (6 to 16 percent), and benzene (not detectable
to 21 percent). A group of 21 unexposed workers was used for comparison. Workers
were divided into two exposure groups: those exposed to <100 ppm (<434 mg/m^) and
those exposed to >200 ppm (>868 mg/m3) xylene during an 8-hour shift. The authors
reported that creatinine excretion was increased in workers exposed to high solvent levels.
Urinary glucuronic acid, hippuric acid, and phenol were also increased in a dose-related
manner. A dose-related decrease in uric acid excretion was noted and was attributed to a
reduction in glycine available for conjugation. No other xylene-related effects were
reported.
Askergren (1981,1982) and Askergren et al. (1981) evaluated renal function in
workers exposed to organic solvents. A total of 134 workers were studied; only 40 of the
subjects were exposed to xylene and toluene in the manufacturing of paint A group of 48
unexposed workers was used as the control. Workers exposed to xylene and toluene had a
significantly (p <0.05) increased mean urinary albumin concentration (13.7 mg/L or 13.7
ppm) as compared with that of the control group (3.7 mg/L or 2.7 ppm). The median
values for the exposed and control workers were reported to be comparable (3.2 and 3.1
mg/L, respectively). These data indicated that individuals excreting high levels of albumin
were more numerous in the exposed group, but not all exposed workers had elevated
levels; osmolality of the urine was comparable between control and exposed groups
(Askergren, 1981).
Francini et al. (1983) also studied renal damage associated with solvent exposure.
A group of 118 painters (predominantly males) was exposed to toluene, xylene, and other
benzene homologs for an average of 9.1 years. Xylene exposure levels were not
estimated. Two control groups, one consisting of 50 females and 30 males and the other of
16 females and 65 males, were used for comparison. Urinary albumin concentration was
not elevated in the exposed group. Urinary B-glucoronidase levels were significantly (p
<0.01) higher in the exposed group than in either control group.
7-37
-------
Pulmonary Toxicity
No chronic effects studies of gasoline on the human respiratory system were
identified in the literature.
Neurotoxicitv
No data are known regarding the chronic oral exposure of humans to liquid
gasoline. However, several reports of chronic exposure to gasoline vapors were found.
Long-term effects that have been reported to be associated with chronic gasoline sniffing
include loss of appetite and weight, neurological and psychological symptoms, muscular
weakness and cramps, and possible liver and renal damage (Poklis and Burkett, 1977).
Machle (1941) reported that no signs of chronic gasoline poisoning were revealed
in 2,300 refinery workers observed for 10 to 12 years. Many of these men were employed
for a few hours per week to mix ethyl fluid with gasoline and to perform other activities;
gasoline exposures of all types and degrees of severity were encountered. Likewise, no
signs of chronic intoxication were observed in studies of large numbers of filling station
attendants, tank wagon drivers, and garage mechanics. No specifics were given regarding
exposure levels, numbers exposed at each level, or duration of exposure.
A survey of 51 workers handling gasoline at retail filling stations in India was
conducted by Pandya et al. (1975). Benzene contents of the petrol ranged from 10 to 17
percent, and phenol levels in urine were measured as an index of benzene and hydrocarbon
exposure. Workers reported symptoms such as headache, fatigue, sleep disturbance,
memory loss, giddiness, and generalized weakness. Since the authors did not report on the
level of gasoline present at the time exposures took place, it is difficult to establish a
relationship between reported symptoms and exposures.
Sterner (1941) described symptoms in a group of painters exposed to vapors
released during spray-painting with gasoline-diluted paints. The chief symptoms were
headache, nausea, weakness, mental depression, anorexia, and inability to sustain attention
and activity. In addition, a significant decrease was noted in hemoglobin, erythrocytes,
and blood cell volumes, with an increase in corpuscular hemoglobin, mean corpuscular
volume, and reticulocyte counts. Although the authors attributed these effects to the
aromatic hydrocarbons (mostly toluene and xylene with very little benzene) which averaged
about 5 to 10 percent of the total hydrocarbons, the lack of controls and better exposure
characterization render interpretation of the reported results questionable. These workers
had been exposed for 10 years or more and had used several types of respirators.
Kurppa and Husman (1982) studied the effects of solvent exposure on serum
enzyme activities, neurophysiological function, and ophthalmological parameters. A group
7-38
-------
of 102 male car painters exposed to toluene (306 ppm or 1,150 mg/m^), xylene (5.8 ppm
or 25 mg/m3), and other solvent compounds (butyl acetate, 5.7 ppm; white spirit, 4.9
ppm) for periods of 1 to 40 years were studied and compared to a group of 102 age-
matched controls. Abnormally slow motor and sensory conduction velocities and
neuromuscular latencies were observed in 12 of 59 car painters who underwent
neurophysiological evaluation. Ophthalmological examinations revealed lens opacities in
48 of 92 men from the solvent exposure group. The results of serum enzyme activity
determinations were comparable between the exposed and control groups.
Hematoxicity
McLean (1960) described an oil company employee who developed hemolytic
anemia and myelofibrosis after 12 months' exposure to gasoline vapor resulting from
spills. Machle (1941) also reported thrombocytopenicpurpura in a man who had cleaned
metal pans in gasoline for 2 years. The benzene content of gasoline in the first case was <1
percent and in the second instance may have been as high as 10 percent
An examination of 39 female workers exposed to gasoline during the production of
rubber products revealed effects such as mild hypochromic anemia, lymphocytosis, and
granulocytosis with Mommsen toxic granules in one-third of the women, generally in those
with the longest employment (Amorati et al., 1952). All of the subjects except four had
been exposed for at least 1 year, and 15 of the women had worked with gasoline for 6 to 9
years. The authors believed significant gasoline exposure had occurred during the last
three steps in the production of sanitary rubber products.
Available data suggest a tentative link between benzene exposures and anomalies in
immunoglobulin, complement, and T-lymphocyte levels. However, no reports were found
of immunological anomalies attributable solely to benzene exposures. Serum
immunoglobulin levels (Lange et al., 1973a) and leukocyte agglutinins (Lange et al.,
1973b) were studied in a group of 35 workers with a history of exposure to benzene,
toluene, and xylene. The concentration of these compounds in the air ranged from 0.011 to
0.17 mg/L, 0.08 to 0.23 mg/L, and 0.12 to 3.0 mg/L, respectively, and the duration of
exposure ranged from 1 to 21 years. Serum IgG and IgA levels were found to be
significantly lower in the solvent-exposed workers than in nonexposed controls, although
IgM levels tended to increase (Lange et al.. 1973b). It was also reported that 10 of the 35
workers had leukocyte agglutinins for autologous leukocytes and demonstrated an increase
of leukoagglutination liter in human sera after incubation with benzene, toluene, and
xylene; this suggested that some workers exposed simultaneously to these aromatic
compounds may exhibit allergic blood dyscrasias (Lange et al., 1973a). In another study,
7-39
-------
79 workers exposed to benzene, toluene, and xylene showed a significant (p<0.01) mean
reduction in compliment level (Smilok et al., 1973). It should be noted that in all of the
studies discussed above, the specific solvent(s) responsible for the changes was not
identified.
The effect of exposure to mixed organic solvents (benzene, toluene, and xylene) on
the hematopoietic system was studied by Moszczynski (1981,1982) and Moszyczynski
and Lisiewicz (1983a, b, 1984). Increased reticulocyte counts were noted in workers
exposed for 1 to 122 months. Workers exposed for 55 to 122 months exhibited decreased
leukocyte and total lymphocyte counts, lowered mean corpuscular hemoglobin levels, and
increased monocyte counts. The complex nature of solvent exposure in this group of
workers made it difficult to determine an etiological agent These investigators also found
that occupational exposure to benzene, toluene, and xylene (up to 379 mg/m^, 580 mg/m^,
and 560 mg/m^, respectively) for periods greater than 55 months resulted in a significant
decrease (p <0.01) in the T-lymphocyte count with no disturbance in T-lymphocyte
function or increase in respiratory or urinary infection susceptibility in exposed individuals
(Moszczynski and Lisiewicz, 1983b).
In another study, Askergren (1981) reported the excretion of erythrocytes and
leukocytes in the urine of men exposed to various solvents (styrene, toluene, and/or
xylene). The average excretion of erythrocytes and leukocytes in the urine of the exposed
workers was significantly elevated as compared with the controls. This increase in urinary
blood cells was attributed to glomerular damage by the various solvents. Askergren (1982)
suggested that increased excretion of albumin and blood cells in the urine could be caused
by increased permeability of the glomerular basement membrane, while proximal and distal
tubular function (as measured by urine concentration ability) did not appear to be affected.
Sukhanova et al. (1969) studied the effects of occupational exposure to organic
solvents (including xylene) on immunocompetency in humans. A slight but statistically
significant decrease in neutrophilic glycogen content was observed in xylene-exposed
workers. A correlation between length of employment and magnitude of effect was
reported. No information was found in the literature on the immunotoxic effects of xylene
in study animals.
Anti-glomerular basement membrane (GMB) antibody disease in humans is often
rapidly fatal. Beime and Brennan (1972, as cited in Kuna and Ulrich, 1984) have
implicated some petroleum products (solvent) in antibody-mediated glomerulonephritis
(Goodpasture's syndrome). They retrospectively analyzed the occupational exposure
history of eight patients with Goodpasture's syndrome. Six of the eight patients had
positive histories of exposure to various industrial hydrocarbon solvents, hair sprays, and
7-40
-------
painting solvents. Most of the patients were heavily exposed to vapors or a fine mist of
heated solvents. In only one case was a specific reference to jet fuel exposure made.
Liver Toxicirv
Little information exists concerning the effects of chronic gasoline exposure on the
liver. Przybylowski et al. (1978b, as cited in U.S. EPA, 1988) found a "slight" toxic liver
lesion and a reduction in alpha-2-macroglobulin, presumably in workers exposed to
gasoline. No data were provided, however, on the exposure conditions or the number of
persons studied.
7.1.3.2 Animal Studies
Kidnev Toxicitv
The histopathology of gasoline vapor-induced non-neoplastic renal lesions was
studied by Busey and Cockrell (1984). Groups of male Fischer 344 rats exposed for 6
hours per day, 5 days per week, to either room air or 67,292, or 2,056 ppm (groups I
through IV, respectively) of unleaded gasoline vapors were sacrificed at 3,6,12, 18, and
24 months (for details, refer to MacFarland et al., 1984, in Carcinogenicity section). The
histopathologic examination of the kidneys from the exposed rats indicated that the
incidence and severity of both regenerative epithelium and tubular dilation increased in an
exposure- and time-related fashion. Male rats exposed to 292 or 2,056 ppm of unleaded
gasoline for 12, 18, and 24 months had striking linear mineral deposits in their renal
medulla, which served to distinguish them from the controls.
Incidences of this mineralization in Group I through IV were 0,5,63 and 91
percent, respectively. These deposits were shown to contain calcium and phosphates and
probably resulted from mineralization of cellular debris. Busey and Cockrell (1983)
suggested a progressive pathological process in the epithelial cells of the proximal
convoluted tubules including cell death and sloughing with accumulation at the
corticomedullary junction leading to tubular dilation. Sloughed cells are replaced by
regenerating cells (U.S EPA, 1988). Several exposed male rats over one year of age
demonstrated karyomegaly, renal tubular epithelial cells with very large nuclei protruding
into the tubule lumen. In addition, renal tubular epithelial hyperplasia was observed in
several male rats over 18 months of age. The authors suggested that these alterations,
although not commonly observed, may be preneoplastic in nature (U.S EPA, 1988). The
exposure-related lesions consisted of increased foci of regenerative epithelium in the renal
conex and dilated tubules with intratubular protein at the corticomedullary junction.
7-41
-------
Kidneys from female rats lacked the exposure-related renal lesions with the exception of
two high-dose animals, which, at the 24-month sacrifice, had medullary mineralization.
Pulmonary Toxicitv
No chronic exposure effects studies of gasoline on the respiratory systems of
laboratory animals were identified in the literature.
Neurotoxicitv
No chronic exposure effects studies of gasoline on the nervous systems of
laboratory animals were identified in the literature.
Hematotoxicitv
Wilson (1983) reported on immunopathological studies conducted using the serum
and tissue samples of kidney and lung from exposed rats and mice dying before the
termination of a 2-year chronic study by the International Research and Development
Corporation (IRDC). The techniques employed in these studies were direct
immunofluorescence assay, indirect immunofluorescence assay, and radioimmunoassay.
These studies provided no convincing evidence of hydrocarbon-associated formation of
anti-basement antibodies. Sampling was incomplete because in some test groups either
serum was lacking or kidney tissue was unsatisfactory for a meaningful
immunofluorescence study.
7.2 BENZENE
7.2.1 Effects of Acute Exposure to Benzene
7.2.1.1 Human Studies
The clinical signs of acute exposure to high concentrations of benzene include deep
anesthesia with resultant narcosis, coma, and death from respiratory arrest or myocardial
sensitization. A single exposure to 66,000 mg/m3 of benzene was fatal within 5 to 10
minutes (IARC, 1982). Nausea, headache, and a feeling of euphoria that may lead to
unconsciousness can be seen after mild exposure to benzene. Signs of acute poisoning
seen at autopsy include inflammation of the respiratory tract, lung hemorrhage, and kidney
congestion (Drozd and Bockowsk, 1967; Dubois, 1967; Snyder and Kocsis, 1975a; Winek
and Collom, 1971). In acute poisonings, (Sax, 1984) individuals become confused and
dizzy, and complain of tightening of the leg muscles and pressure over the forehead. They
7-42
-------
then pass into a stage of excitement and, if exposure continues, become stupefied and lapse
into coma. When applied to the skin, benzene has a strong irritating effect (Sax, 1984). It
produces erythema and burning. In more severe cases, edema and even blistering may
result.
7.2.1.2 Animal Studies
Lethality
Oral acute LDsQs are in the range 3.8 to 6.87 g/kg in adult rodents; one
investigation, however, reported a value of 0.93 kg/kg/body weight in Sprague-Dawley
rats. Acute inhalation LC$QS are in the range of 10,000 to 14,122 ppm in mice and 9,536
to 13,700 ppm in rats (Bonnet et al., 1982; Fielder, 1982; IARC, 1982; NRC, 1986a;
NTP, 1986b).
Neurotoxicity
Benzene was reported to be moderately irritating to the skin (500 mg for 24 hours)
and severely irritating to the eyes (2 mg for 24 hours) (RTECS, 1983-84). Benzene
produced no significant sensory or pulmonary irritation in mice exposed to vapor
concentrations as high as 8,500 ppm for 30 minutes (Neilsen and Alarie, 1982).
Acute inhalation or oral intake of lethal amounts of benzene induces anesthetic
effects, limb paralysis, and convulsions in rodents and rabbits (Fielder, 1982; NRC,
1986b).
Benzene vapors at an atmospheric concentration of 6,780 ppm (lowest
concentration tested) induced restlessness and hypersynchronous amygdaloid activity in
cats (sex not indicated). At atmospheric concentrations of 30,000 to 39,000 ppm, benzene
induced generalized convulsive tonic-clonic seizures within minutes of exposure initiation
as well as ataxia and a lack of responsiveness to environmental stimuli. Animals recovered
from ataxia 12 minutes after the intoxication was discontinued. Tolerance developed with
subsequent exposures (Contreras et al., 1979).
Hematoxicity
Several investigations have been conducted to assess the changes benzene may
cause in blood cell structure and function. These investigations were conducted as a result
of epidemiological findings that linked benzene exposures to leukemia, particularly acute
non-lymphocytic leukemias, and various hematological disorders. Such findings include
reductions in the number of circulating blood cells, chromosomal aberrations, bone marrow
7-43
-------
damage, and aplastic anemia. The subjects of these studies were exposed to benzene for
periods of several months to several years, although few were exposed for periods longer
than 10 years. The animal studies attempted to derive a more complete clinical pattern of
benzene-induced changes in these blood cells.
Animal studies regarding benzene exposure have focused on the leucopenic effects
of this compound. In addition, because the bone marrow produces cells involved in the
immune response, experiments were also conducted to investigate the effects of benzene on
specific immune system cells and functions.
Limited information is available concerning the hematologjcal effects of acute
benzene exposure on laboratory animals. Toft et al. (1982) studied the effects of acute
benzene inhalation exposures (95 and 201 ppm) on male NMRI mice. Specifically, they
evaluated the effect of exposure on: the number of nucleated cells per tibia (cellularity) and
the number of colony forming granulopoietic stem cells per tibia (CFU-C). In mice
exposed to 95 ppm of benzene for 6 to 8 hours, total cellularity was reduced by
approximately 50 percent, and CFU-C numbers were decreased to about 20 to 30 percent
of the controls. Exposure to 201 ppm produced more marked changes in these parameters
(Toft et al., 1982). Cellularity and CFU-C were reduced to between 1 and 5 percent of
controls after 8 hours, with a significant decrease (50 percent) in cellularity observed after
only 2 hours of exposure. All hematological effects for both exposures were related to
exposure duration, although CFU-C did not show a consistent downward trend until about
4 hours into the exposure.
Uyeki et al. (1977) exposed female BDF mice to 4680 ppm of benzene for 4 to 8
hours. Significant depletions (60 percent) in the numbers of bone marrow colony forming
cells (CFC) were observed 1 day after the inhalation exposure. The CFC levels returned to
about 80 percent of controls by the seventh day after exposure, indicating a return to
steady-state. Total cellularity was unchanged. Spleen weight and cellularity were initially
decreased to 80 to 90 percent of controls on the first day after exposure. They then became
slightly elevated (about 140 percent of controls) on the fourth day following exposure,
although these values returned to the control levels by the seventh day. CFC content in the
spleen was unchanged relative to controls at all monitoring times. The authors noted that
this finding contrasts with other experiments they conducted in which significant elevations
in splenic CFC content were evident 1 week after exposure. To explain these differences,
they suggested that the toxic effects of benzene might include inhibition of stem cell
migration to the spleen, or that the degree of damage may not be significant to trigger
repletion events in the spleen. Peripheral white blood cell counts were not measured. A
similar study by Gill et al. (1980), however, showed significant (50 to 70 percent)
7-44
-------
decreases in peripheral white blood cell counts in C57B1 mice exposed to 500 or 1,000
ppm of benzene for 24 hours.
Uyeki et al. (1978) also studied female BDF mice exposed repeatedly to 4,680 ppm
of benzene. One group (four mice) were exposed for 3 days, and another group (four
mice) were exposed for an additional 4 hours on day 4. Bone marrow colony forming cells
were cultured on day 4 for both groups. The CFC content in the bone marrow was
depressed to 45 percent of controls in the 3-day exposure group, and to 13 percent of
controls in the 4-day exposure group. Both spleen weight and bone marrow cellularity,
however, were reduced in the 4-day exposure group only. This may reflect the lack of time
these animals had to recuperate from the exposure, and thus may represent an acute effect
more than an effect of repeated exposures.
7.2.2 Effects of Subacute and Subchronic Exposure to Benzene
7.2.2.1 Human Studies
The association of protracted exposure to benzene and aplastic anemia has been
known for many years (Santesson, 1987; Selling, 1916), as has the association of
exposure to benzene and increased risk of leukemia (Delore and Borgomano, 1928).
DeGowin (1963) described a case of aplastic anemia that developed into acute myeloid
leukemia about 15 years later in a painter who had used benzene for 13 years. Other case
reports linking exposure to benzene with leukemia include Goguel et al. (1967), Goldstein
(1977), Hunter (1964), Mallory et al. (1939), and Vigliani and Saita (1964). Other blood
disorders associated with long- and short-term exposures to benzene include
erythromyelosis (DiGuglielmo and lannacione, 1958; Fomi and Moreo, 1969; Galavotti
and Troisi, 1950; Goldstein, 1977; Nissen and Ohlsen, 1953; Rozman et al., 1968).
Subchronic exposure periods are often difficult to quantify in epidemiological
studies. Epidemiological studies on workers exposed to benzene in occupational settings
have primarily utilized the retrospective method of study. Past employment records and
death certificates were used to establish a study cohort whose exposure to benzene may
have ranged from less than 1 year to 15 years (Aksoy, 1980; Ott et al., 1978; Rinsky et al.,
1987). These study designs have allowed a latency period of up to 15 years (API, 1986b;
NRC, 1986a; Rinsky et al., 1987) and a follow-up period of 25 years (Aksoy, 1980;
Aksoy et al., 1978; Bond et al., 1986b; Ott et al., 1978). Studies involving Subchronic
exposures to benzene are thus addressed below in the chronic toxicity section.
Carcinogenic effects of benzene exposure are reviewed separately.
7-45
-------
7.2.2.2 Animal Studies
Lethality
Increased mortality was reported following subchronic inhalation of benzene by
mice (Cronkite, 1986; Cronkite et al., 1985; Green et al., 1981).
Neurotoxicity
Changes in earing, grooming, sleeping, and resting behavior were reported in mice
exposed to 300 or 900 ppm benzene vapor in a scheme of repeated alternating cycles of five
daily inhalation exposures followed by 2 weeks of no exposure (Evans et al., 1981).
Groups of four male Sprague-Dawley rats were exposed 6 hours/day for 3
consecutive days to benzene vapors at atmospheric concentrations of 0 or 1,500 ppm.
Benzene induced significant (p <0.05) increases in the levels and/or turnover of dopamine
and noradrenaline in some regions of the hypothalamus (Anderson et al., 1983).
Groups of six rats (strain and sex not given) were exposed to benzene vapor at
atmospheric concentrations of 4 or 20 ppm, 6 hours/day, 6 days/week, for 5.5 months.
The control group was not described. When conditioned reflex activity was measured,
there was a delay in response time in rats treated with 20 ppm benzene but not in those
administered 4 ppm benzene (Novikov, 1956).
Horiuchi et al. (1967) investigated the effects of benzene exposure on spontaneous
behavior in mice. In groups of five male NA 2 mice, inhalation of benzene vapors at
atmospheric concentrations of 10 or 100 ppm 6 hours/day for up to 20 days induced
significant (p <0.05) decreases in cumulative wheel-turning activity compared to the air
exposed controls. However, it was not determined whether or not this effect was
reversible (Horiuchi et al., 1967).
Hematotoxicitv
The hematological effects of benzene exposures at levels above 100 ppm have been
well documented (Johnston et al., 1979; Gill et al., 1980; Ward et al., 1985; Toft et al.,
1982; Green et al., 1981; Cronkite et al., 1985), and will not be reviewed here unless they
add to the understanding of lower dose effects. Instead, this assessment will focus on
effects associated with benzene exposures of 100 ppm and below.
Early studies found leucopenic effects in laboratory animals following subchronic
benzene exposures to less than 100 ppm (Wolf et al., 1956; Deichmann et al., 1963). Wolf
et al. (1956) exposed guinea pigs (five to ten per exposure group), rats (10 to 15 per
exposure group), and rabbits (one to two per exposure group) to 80 to 88 ppm of benzene,
7-46
-------
7 hours/day, 5 days/week. Rats were exposed over a 204 day period; guinea pigs were
exposed over either a 32-day or 269-day period; rabbits were exposed over a 243 day
period. Slight leucopenia was observed in all animals at all exposure durations. Spleen
weights were increased in rats and in guinea pigs exposed for 269 days. The latter group
also showed evidence of histopathological changes in the bone marrow. An oral dose of
10 mg/kg/day in female rats administered over 132 days also resulted in slight leucopenia.
A dose of 1 mg/kg/day over the same period was without effect
Deichmann et al. (1963) conducted 10 inhalation experiments on male and female
Sprague-Dawley rats (40 rats per exposure group). The following experiments are
specifically relevant to this assessment: (1) exposure of rats to 65 pm of benzene over a
39-day period, 5 days/week; (2) exposure of rats to 47 ppm of benzene, 7 hours/day, five
days/week over 245 days; (3) exposure of rats to 44 ppm of benzene, 7 hours/day, for 8
weeks; (4) exposure of rats to 31 ppm of benzene, 7 hours/day, for 17 weeks; and (5)
exposure of rats to 15 ppm of benzene, 7 hours/day, 5 days/week, for 28 weeks.
Leucopenia was observed in the 65 ppm exposure group after 2 weeks. In the 44 ppm and
47 ppm exposure groups, leucopenia became apparent in female rats after 3 to 7 weeks of
exposure, and in male rats after 7 to 8 weeks of exposure. No leucopenic effects were
observed for either the 31 ppm or 15 ppm exposure groups. Histopathological analysis of
the spleen, however, identified mild to moderate excesses in hemosiderin pigments in the
15 ppm exposure group. The effect was more significant in females than in males.
Deichmann et al. (1963) also compared the extent of leucopenia associated with the various
exposure regimens. They found that the degree of leucopenia induced was similar in rats
exposed to 830 ppm to that in rats exposed to 65 ppm, although exposure to the higher
concentration produced an earlier response. The leucopenia was less severe in the 47 and
44 ppm exposure groups. These findings imply that the leucopenic effects in animals are
mediated by a metabolite of benzene.
Bone marrow cellularity and colony cell forming cell assays in mice were conducted
by Toft et al. (1982). The authors exposed male NMRI mice (five per exposure group)
either intermittently or continuously to benzene concentrations ranging from 1 to 200 ppm
for up to 10 days. Specifically, they evaluated the effect of exposure on: (1) the number of
nucleated cells per tibia (cellularity), and (2) the number of colony forming granulopoietic
stem cells per tibia. At 95 ppm, the number of nucleated cells per tibia was near zero after
4 days and the number of colony forming granulopoietic stem cells per tibia was near zero
after 2 days of continuous exposure. Continuous exposure to 50 and 21 ppm produced
less severe, but still significant depressions in bone marrow cellularity. Continuous
exposure to 50 ppm of benzene produced results similar to the 95 ppm exposed group after
7-47
-------
5 days of exposure. Exposure to 21 ppm of benzene reduced the CFU-C/tibia to about 5
percent of the control values after 10 days of exposure. At this exposure concentration, the
number of nucleated cells per tibia was reduced to about 25 percent of controls after 4 days
of exposure, followed by a slight increase through the tenth day. Exposure to 10 or 1 ppm
of benzene, however, produced no reduction in CFU-C levels (data were not shown).
This finding implies either a low sensitivity in the assay or a steep dose-response curve.
Toft et al. (1982) also conducted intermittent exposures (8 hrs/day, for 2 weeks).
Intermittent exposures (8 hours/day, 5 days/week) produced depressions in CFU-C at
concentrations of 21 ppm and higher, and depressions in bone marrow cellularity at 50
ppm and higher. No effects on these parameters were observed at either 10 or 1 ppm.
Toft et al. (1982) also found that for equivalent doses, continuous exposures were
more toxic than intermittent exposures, and that intermittent exposures of longer durations
were more toxic than those of shorter durations. For example, the authors noted that a 24
hour exposure to 95 ppm (2,280 ppm-hr) or a 96 hour exposure to 21 ppm (2,016 ppm/hr)
produced severe toxicity, while exposure to 95 ppm of benzene, 2 hours/day, 5 days/week,
for 2 weeks produced almost no effects. They also noted that a 2-week exposure to 50
ppm of benzene for 8 hours/day (4,000 ppm) produced more severe toxicity than a 201
ppm, 2 hr/day (4,020 ppm/hr) exposure.
Cronkite et al. (1985) investigated the effects of benzene in male and female C57B1
mice. The mice (eight per exposure group) were exposed 6 hours/day, 5 days/week, for 2
to 16 weeks to 10, 25,100, 300, or 400 ppm of benzene. They examined blood counts,
total cellularity of the bone marrow, numbers of pluripotent stem cells in the bone marrow,
colony formation unit (CPU), the fraction of the stem cells involved in DNA synthesis, and
tumor incidence. Inhalation of 100 ppm of benzene over a 2 week period depressed
marrow cellularity, the CPU level in the femur, and levels of blood lymphocytes and
granulocytes. The only effect observed below 100 ppm was a decrease in lymphocyte
levels after a 10 day benzene exposure to 25 ppm. No effects were observed at 10 ppm.
Cronkite et al. (1985) also examined the reversibility of the stem cell and
lymphocyte depression in the animals exposed to 300 ppm of benzene. Peripheral
lymphocyte levels returned to control levels by 8 weeks post-exposure in mice exposed for
2 to 16 weeks. CPU levels returned to normal by 4 weeks after the end of the benzene
exposure in the mice for either 2 or 4 weeks. CPU levels returned to normal by 16 weeks
after the end of the benzene exposure in mice for 8 weeks. In the mice exposed for 16
weeks, CPU levels were at 60 percent of controls 16 weeks after the end of exposure. One
sample taken from these mice 25 weeks after exposure indicated a CPU level that was 92
percent of the controls. The authors did not speculate on whether the effects with this
7-48
-------
exposure regimen are completely reversible. These findings indicate that lymphocyte
depressions resulting from acute or subacute benzene exposures are reversible. The
reversibility of this effect during longer exposure periods, however, is unknown.
Rozen et al. (1984) observed changes in peripheral red blood cell lymphocyte levels
in C57B1 mice exposed to either 10,30,100, or 300 ppm of benzene. The animals (seven
to eight per group) were exposed 6 hours/day, for 6 days. Assays were conducted
immediately after the last exposure. Lymphocyte levels were depressed to about 60 percent
of the controls in the 10 and 30 ppm exposure groups, and to about 40 percent of the
controls in the 100 and 300 ppm groups. Red blood cell levels rose slightly in the 10 ppm
group, and then fell to about 80 percent of controls in the 100 and 300 ppm groups. All
benzene concentrations reduced mitogen induced bone marrow B-lymphocyte colony
forming ability. Lipopolysaccharide-induced B-lymphocyte colony forming ability was
depressed to about 30 percent of controls at all exposure concentrations. These
investigators also found that splenic T-Iymphocytes phytohemaglutinin-induced
blastogenesis was significantly reduced at 31 ppm. Marrow B-lymphocyte numbers and
splenic T-lymphocytes were reduced in the 100 and 300 ppm exposure groups.
The findings of Rozen et al. (1984) are supported by cell culture studies. Mitogen
responsiveness was also reduced in vitro when rat spleen cells and human lymphocytes
were exposed to a 10 micromolar solution of two benzene metabolites, hydroquinone and
benzoquinone (Irons and Pfeifer, 1984).
Green et al. (1981) exposed male CD-I mice (four per exposure group) 6 hrs/day
for 5 days to 1.1, 9.9, 103, 306, 603, 1,276, 2,416, and 4,862 ppm of benzene. The
mice were also exposed to 9.6 ppm of benzene at the same weekly exposure regimen for 10
weeks and to 302 ppm for 26 weeks. They observed changes in the spleen which occurred
during the lowest exposures (5-day exposure to 1.1 or 9.6 ppm; 50-day exposure to 9.6
ppm). Depressions in spleen weight (85 percent of controls) and splenic granulocyte
counts (50 percent of controls) were found at all benzene exposure levels during the 5-day
exposure. Only the 1.1 ppm, 5 day exposure showed a significant depression for splenic
granulocyte counts; however, dose-related decreases in both of these parameters became
apparent as these responses were compared to responses at higher dose levels (103 through
4,862 ppm). Significant increases in spleen weight and cellularity were found following a
10-week exposure to 9.6 ppm. The authors noted that spleen effects are likely to be quite
variable, since the rodent spleen is a labile organ which is naturally involved in
erythropoiesis and is often a first line of defense against prevailing hematopoietic stress.
On the other hand, these findings show that benzene exposures as low as 1 ppm can affect
7-49
-------
the functioning of the hematopoietic system in rodents, and may provide a sensitive
indicator of benzene-induced hematotoxicity in human beings.
Green et al. (1981) also observed other measures of toxicity to the bone marrow,
spleen, and peripheral blood (granulocytopenia, lymphocytopenia) in the mice after a 5-day
exposure to 103 ppm of benzene. Exposure to 9.6 ppm for 10 weeks (an equivalent
administered dose to 103 ppm for 1 week) produced no detectable toxicity to the bone
marrow and peripheral blood The findings of Green et al. (1981) agree with those of
Cronkite et al. (1985), but differ from those of Rozen et al. (1984), who found significant
lymphocyte depression in response to 10 ppm of benzene, and of Toft et al. (1982), who
found that longer exposures to equivalent doses increased toxicity. The Green et al. (1981)
findings also differ from those of Baarson et al. (1984), who found effects in vitro on bone
marrow cells as a result of exposures of mice to 10 ppm of benzene. These exposure
durations were longer than those in the Cronkite et al. (1985) study. Five male C57B1
mice were exposed to 10 ppm of benzene, 6 hours/day, 5 days/week, over 178 days.
Erythroid colony forming units (CFU-E) in the bone marrow showed a steady decline
throughout the exposure. By the end of the exposure, they were only 5 percent of the
control values. Marrow erythroid burst forming units (BFU-E, progenitors of CFU-E)
were reduced to 55 percent of control values at 66 days, but returned to control values by
the end of exposure. In the spleen, CFU-E, cellularity, and the number of nucleated red
cells were significantly reduced by the end of the exposure period. BFU-E in the spleen
was slightly elevated throughout the exposure period. Levels of circulating red cells and
lymphocytes were also depressed in the exposed mice.
Baarson et al. (1984) noted that while the inability of erythroid colony forming
units to differentiate in culture may not reflect in vivo conditions, the study indicates that at
least one functional characteristic of these cells had been changed by benzene exposure.
They also noted that changes in CFU-E represent a less severe response than changes in
BFU-E, in that damage to BFU-E could result in significant depressions in CFU-E. Thus,
the depression of the BFU-E during the exposure was interpreted as a significant finding.
Finally, effects on spleen cells also indicated that the effects of benzene exposure extended
to this secondary erythropoietic organ.
Aoyami (1986) exposed BALB/c male mice to 50 ppm of benzene for either 7 or 14
days. The ratio of the organ weight to body weight decreased at both exposure durations
for the spleen (15 percent and 60 percent) and at the 14-day exposure duration for the
thymus (50 percent). The ratio and absolute numbers of T- and B-lymphocytes were
depressed after 7 days of exposure, with the B-cell depression being dose-dependent and
more intense than the T-cell depression. This depression occurred without a decrease in
7-50
-------
total lymphocyte levels, a finding that the authors attributed to a possible benzene-induced
alternation of surface antigens on the lymphocytes. In addition, as the ability of B-
lymphocytes to form antibodies was suppressed, the author concluded that benzene is
selectively toxic to B-lymphocytes at this exposure level. The Aoyami (1986) findings as
well as the Rozen et al. (1984) study are significant in that they show that benzene may
affect not only the numbers of circulating immune system cells, but also the integration of
the immune response among these cells. The Aoyama (1986) study is also significant in
that it shows benzene effects on the thymus. The thymus is a critical organ in the early
development of the immune system. The effects of benzene on the newborn, therefore,
warrant special attention.
Dempster et al. (1984) investigated the behavioral and hematological changes
caused by inhaled benzene in C57BL mice. Mice (30 to 40 animals per group) were
exposed to 0,100, 300, 1000, and 3000 ppm of benzene for 6 hours/day, for the number
of days necessary to achieve a minimum concentration times product of 3,000 ppm-days.
Hematological changes occurred at all exposures. The most sensitive behavioral index,
milk-licking, was affected at the 100 ppm and 300 ppm exposures. Licking significantly
increased during the first week of exposure to 100 ppm, and then gradually returned to
control values during the succeeding weeks. At 300 ppm, licking increased significantly
by the end of the first week of exposure, and continued throughout the second week.
(Food intake, by contrast, decreased significantly in this exposure group during the first
week but returned to control values during the second week, indicating that licking was not
solely attributable to hunger. Milk-licking behavior at this exposure did, however follow
the same time course as the hematological changes, suggesting similar toxicological
mechanisms.) There was a marginally significant increased trend (p=0.056) in milk-licking
for the 3-day, 1000 ppm exposure. No changes in licking were observed after the 1-day,
3000 ppm exposure. However, other signs of central nervous system toxicity, such as
body tremors were observed. Significant decreases in hind limb grip strength (an
indication of peripheral neuropathic effects) were observed in the 3,000 and 1,000 ppm
exposure groups, but not in the 300 ppm group. Retesting at 60 to 70 days post-exposure
indicated complete recovery from the effects of subacute dosing. The neurobehavioral
alterations observed in the subacute study have been attributed to alterations in brain
monoamine levels, particularly to imbalances in the concentrations of catecholamines and
indolamine (Hsieh et al., 1988).
Locomoter activity was unaffected by benzene exposure when Dempster et al.
(1984) compared their findings to those of Horiuchi et al. (1967), who found decreased
wheel-turning in mice exposed to 10 ppm of benzene. They suggested that because wheel
7-51
-------
turning is a more strenuous exercise than simple locomotion, it may represent a more
severe health endpoint.
The effects of benzene on regional brain monoamine neurotransmitter levels in CD-
1 mice were examined by Hsieh et al. (1988). Mice were fed benzene in drinking water for
4 weeks at concentrations of 0, 31,166, and 790 mg/L (corresponding to 0, 8,40, and
180 mg/kg/day). Dose related increases in dopamine, serotonin, and norepinephrine were
observed in the hypothalamus. Increases were frequently associated with increases in
metabolites of these neurotransmitters, indicating an elevated turnover. The authors
concluded that the data do not predict a no effect level for benzene in drinking water.
Hematotoxic effects have been observed in CD-I mice following a 4-week
exposure to benzene in drinking water at concentrations of 31,166, and 790 mg/L (five
animals per group) (Hsieh et al., 1988). These concentrations corresponded to body doses
of 8,40, and 180 mg/kg/day. Benzene-treated water produced dose-related increases in
spleen weight and kidney weight. Benzene exposure caused a significant dose-response
reduction in peripheral blood leucocytes, lymphocytes, and erythrocytes, and resulted in
severe macrocytic anemia. Splenic lymphocyte proliferation to both TS cell and T cell
mitogens was enhanced at the lowest dose and depressed at the higher doses. Effects on
cell-mediated injury followed a similar biphasic pattern. Antibody production was
suppressed at the two highest dose levels. The authors concluded that their findings
suggest a biologically significant immunotoxic effect on both cellular and humoral
responses at the dose levels tested.
7.2.3 Effects of Chronic Exposure to Benzene
7.2.3.1 Human Studies
Hazards of chronic occupational benzene exposure include cytogenetic damage,
hematologic cytopenias (including aplastic anemia and pancytopenia), and leukemia
(Aksoy, 1980; Aksoy et al., 1971,1972a, b, 1976; IARC, 1982; NRC, 1986a).
Malignant lymphoma and myeloid metaplasia have been associated with benzene
exposures to a lesser extent (Aksoy et al., 1975; Snyder et al., 1984; U.S. EPA, 1978a).
Aplastic anemia as a chronic effect of occupational benzene exposure was
recognized clinically when it developed among individuals who used solvents containing a
high concentration of benzene. Individuals were identified in the leather, rubber, printing,
chemical manufacturing, and petrochemical industries (Adamson and Seiber, 1981; Aksoy,
1980; Arp et al., 1983; Austin et al., 1986; Bond et al., 1986b; Chekoway et al., 1984;
7-52
-------
Green et al., 1979; Infante et al., 1977a, b, 1978; Ott et al., 1978; Rinsky et al., 1981,
1987; Tsai et al., 1983).
Most recent studies (from 1970) on the hematological effects of protracted human
exposure to benzene were conducted by Aksoy and his colleagues in Turkey. A review of
these and earlier studies was conducted by Goldstein (1977). The subjects of the Aksoy
investigations were mainly shoeworkers in small manufacturing shops. High exposures to
benzene occurred when adhesives containing this solvent were used. Benzene
concentrations ranged from 15 to 30 ppm during non-working hours, and from 30 to 210
ppm when the adhesives were used (Aksoy et al., 1976; Aksoy et al., 1971). Rarely,
benzene concentrations as high as 640 ppm were recorded (Aksoy et al., 1976). These
investigations determined that an elevated incidence of acute granulocytic leukemias
(particularly acute myeloblastic leukemias) occurred in these workers and made
comparisons between this disease and various measures of hematological dysfunction.
Pancytopenia is considered to be a secondary immunodeficient disease by virtue of the
reduced numbers of immunocompetent cells that result from bone marrow dysfunction
(Luster and Rosenthal, 1986). It has been suggested that individuals with aplastic anemia
have an increased risk of developing myelogenous leukemia (Goldstein, 1985). Aksoy et
al. (1975) reported the case of a leather worker, previously diagnosed with benzene-
induced pancytopenia due to an 8-year occupational exposure, who later developed
myelofibrosis and myeloid metaplasia after a 9-year latency period. Paroxysmal nocturnal
hemoglobinuria was reported in another individual with benzene-induced aplastic anemia
(Aksoy et al., 1975). However, these case reports are insufficient to document a causal
relationship between the reported physiological anomalies and exposure to benzene (U.S.
EPA, 1978a).
Aksoy et al. (1972b, 1976, 1980) made several comparisons between acute
leukemia and other clinical findings in patients with leukemia. One investigation was of
four shoemakers with acute leukemia (Aksoy et al., 1972b). All workers were exposed to
high benzene concentrations (150 to 210 ppm) for 7 to 14 years. Three of the four workers
developed acute myeloblastic leukemia. The other developed acute monocytic leukemia.
Aplastic anemia, a condition characterized by a decrease in all cells formed in the bone
marrow, preceded the cancers in two of the three myeloblastic leukemia cases and in the
monocytic leukemia case. No lymphadenopathy or hepatosplenomegaly was observed in
any of the patients. Thrombocythemia, in association with numerous megakaryocytes, was
observed in the monocytic leukemia case.
In subsequent studies, Aksoy et al. (1976,1980) observed a preceding
pancytopenic period in about 25 percent of the shoeworkers with acute leukemias. A
7-53
-------
summary of the distribution of pancytopenia among the leukemic patients is presented in
Table 7-12. The interval between the onset of leukemia and the pancytopenic period varied
from 6 months to 6 years. The clinical and hematological findings of pancytopenia often
improved considerably or even disappeared completely, despite the fact that leukemia
developed later. The authors also noted, however, that re-exposure to benzene following
the findings of pancytopenia was common in these patients.
In their clinical description of pancytopenia, Aksoy et al. (1972b) emphasized that
this condition is an insidious one. When first seen, most patients had moderate or severe
pancytopenia; however, the duration of symptoms attributable to pancytopenia before the
first visit or examination was very short. Thus mild pancytopenia may go undetected in
many workers who eventually develop leukemia. This possibility is difficult to verify,
since pre-exposure blood counts are rarely available for benzene-exposed workers
(Goldstein, 1977).
In addition to changes in peripheral blood cell numbers, Aksoy et al. (1980) also
noted structural changes in red blood cells, platelets, bone marrow and, rarely, myeloid
metaplasia. Anemia was found to be more common in patients with hyperplastic or
hypoplastic bone marrows. Platelet changes were characterized by cytomegaly. Changes
in red blood cells were evidenced by the identification of erythroblastemia (the elevation of
immature red blood cells), increased osmotic fragility in red blood cells, and a rare disorder
called paroxysmal nocturnal hemoglobinuria (PNH). Increased osmotic fragility (a sign of
membrane changes in red blood cells) was a common finding in the leukemia patients
(Aksoy et al., 1972b). PNH is characterized by the onset of hemolysis and
hemoglobinuria during sleep and is thought to be caused by clonal expansion of an
abnormal stem cell. Aksoy et al. (1980) speculated that it may be due to complement lysis
of defective red blood cells.
In addition to the Aksoy studies on red blood cell changes, two case studies
focused on benzene exposure and the development of erythroleukemias (Rozman et al.,
1968; Galavotti and Troisi, 1950). Rozman et al. (1968) studied a worker with a 30-year
history of working in a closed, benzene-rich environment (concentrations not specified),
while Galavotti and Troisi (1950) studied a man with only 5 and one-half months of
occupational exposure to benzene (concentrations not specified). HepatosplenomegaJy was
observed in both cases, as were nuclear abnormalities (polyploid and arrested mitosis) in
erythroblasts, and abnormalities in the size and shape of red blood cells.
Aksoy et al. (1971,1972a, 1980) also studied pancytopenic workers without
leukemia in the shoemaking shops. Among a cohort of 44 workers with pancytopenia, six
were found to later develop leukemia (Aksoy, 1980). Changes in red blood cells were also
7-54
-------
TABLE 7-12
OCCURRENCE OF PANCYTOPENIC PERIOD PRECEDING LEUKEMIA
IN 42 CASES WITH LONG-TERM EXPOSURE TO BENZENE
Types of leukemia
No. of leukemic patients
with preceding pancytopenic
period (total number in
parenthesis)
Percentage
Acute myeloblastic leukemia
Preleukemia
Acute erythroleukemia
Acute monocytic leukemia
Acute undifferenriated leukemia
Acute lymphoblastic leukemia
Acute promyelocytic leukemia
Chronic myeloid leukemia
Total
3(16)
4(7)
2(8)
1(3)
KD
0(4)
0(1)
0(2)
11(42)
19
57
25
33
100
0
0
0
26
SOURCE: Aksoy, 1980.
7-55
-------
observed in many of these workers. Such changes included erythroblastemia, increased
osmotic fragility, mild hyperbilirubinemia, increased total hemoglobin, and increased
excretion of fecal urobilinogen. The authors noted that the changes could be explained
either by the presence of a hemolytic component or by ineffective erythropoiesis.
Aksoy et al. (1971) investigated the incidence of blood abnormalities among 217
apparently healthy workers who had been exposed to 30 to 210 ppm of benzene for
exposure periods ranging from 3 months to 7 years. Fifty-one (23.5 percent) of the
workers had hematological abnormalities, primarily related to decreased cell numbers.
Leucopenia often occurred in combination with thrombocytopenia. Lymphocytosis was
observed in one case, although the authors also noted other studies in which lymphopenia
in association with leucopenia was seen. The incidence of abnormalities was not correlated
with the exposure period (Aksoy et al., 1971). In addition, there was apparently no
correlation between age and the incidence of the abnormalities (Aksoy et al., 1971).
This latter finding differs from that of Doskin (1971). In this study, 365 workers
employed for 3 years in a chemical plant were evaluated for hematological abnormalities.
The author found that for equivalent exposure durations (3 years), men aged 18 to 21 had
higher incidences and more severe cases of hematological disturbances (e.g., low cell
counts, decreased phagocytic capability) than other workers. The Doskin study (1971) is
particularly relevant to the assessment of low level benzene exposures. Although he did
not state the exposure concentrations, the author noted that the benzene concentrations
exceeded the maximum exposure limit by two to eight times in 64 percent of the samples
taken from the first year of sampling, 37 percent the second year, and 3 percent in the third
year. Assuming that the maximum exposure limit was the current exposure limit of 1.6
ppm, this means that adverse hematological effects were observed in humans at
concentrations at or below 13 ppm (which signifies the lowest concentration associated
with benzene induced effects in humans). Forty percent of the workers exposed to 7 to 13
ppm benzene for 1 year displayed mild hematologic abnormalities. The decreased benzene
exposure levels parallel a decreased incidence of hematological abnormalities. These
findings are especially important given the lack of epidemiological information to
characterize the dose-response relationship for benzene from the occupational studies,
especially at exposure levels currently found in occupational environments. On the other
hand, the actual exposure levels in this study are suspect in that they cannot be determined
with certainty and represent only occasional random measurements (NRC, 1986a; ODW,
1985).
Occupational studies have also identified abnormal biochemical parameters in the
white blood cells of benzene-exposed workers (Goldstein, 1977; Moszczynski and
7-56
-------
Lisiewicz, 1984; Moszczynski, 1983). There are reports of altered immune function,
including leucocyte agglutinins, altered serum immunoglobulin and complement levels, and
antibodies against white blood cells, red blood cells, and platelets in patients chronically
exposed to benzene (Goldstein, 1977). Abnormal granulocyte function was found in cases
of severe aplastic anemia (Goldstein, 1977). It may also occur in mild cases of benzene
hematotoxicity, and may precede cases of overt pancytopenia (Goldstein, 1977).
Indicators of benzene-induced toxicity in granulocytes are decreased alkaline phosphatase
levels, alterations in osmotic fragility and in the fluorescent characteristics of the nuclei, and
the presence of Pelger-Hurt anomaly (Goldstein, 1977). Damage to lymphocyte
lysosomes, as evidenced by an exposure-related increase in the number of acid phosphatase
positive lymphocytes, was found in workers exposed to 0 to 370 mg/m3 of benzene for 1
to 122 months (Moszczynski, 1983). This exposure also resulted in decreased numbers of
lymphocytes having intact N-acetyl-beta-D-glucosaminadose (NAG) positive lysosomes
(Moszczynski and Lisiewicz, 1984). The authors comment that this change represents an
early marker for benzene exposure.
7.2.3.2 Animal Studies
Lethality
Increased mortality was reported during chronic oral and inhalation exposure of rats
and mice to benzene (NTP, 1986b; Snyder et al., 1978,1980,1982a).
Hematotoxicitv
Chronic studies performed on rodent populations were primarily directed towards
the observations of leukemias and other cancers (Snyder et al., 1980; Snyder et al., 1984;
Snyder, 1982a). The benzene exposure concentrations in these studies were either 100
ppm or 300 ppm. The hematological effects observed in these studies (reduced cellularity,
leucopenia, reductions in red blood cells, abnormal changes in red blood cells and platelets,
increased splenic hemosiderin pigments, splenomegaly, bone marrow hyperplasia, and
hypoplasia) were similar to those observed in the subacute and subchronic studies.
7-57
-------
7.3 TOLUENE
7.3.1 Effects of Acute Exposure to Toluene
7.3.1.1 Human Studies
The acute toxic effects of toluene in humans have been evaluated primarily by
studying individuals exposed to toluene via inhalation in experimental or occupational
settings or during episodes of intentional abuse. Central nervous system dysfunction
represents the primary health effect of concern. Acute experimental and occupational
exposures to toluene ranging from 200 to 1,500 ppm (750 to 5,600 mg/m3) have resulted
in dose-related CNS changes such as fatigue, nausea, muscular weakness, confusion, and
incoordination, as well as impairments in reaction time and speed (U.S. EPA, 1983b).
Exposures to high concentrations of toluene (i.e., levels approaching air saturation
concentrations of 30,000 ppm), either accidentally or deliberately, initially result in such
CNS stimulatory effects as exhilaration, light-headedness, dizziness, and delusions. As
the duration of exposure increases, narcotic effects indicative of CNS depression
progressively develop, followed, in extreme cases, by collapse, loss of consciousness, and
death (U.S. EPA, 19835). Toluene exerts a stronger irritant effect than benzene on the
skin and mucus membranes (Patty's, 1983). Severe dermatitis may result from its drying
and defatting action, although it has not been found to be a dermal sensitizer (Patty's,
1983).
Very little dose-response data exists regarding the sensory irritation and central
nervous system effects of toluene exposure. Often cited studies (e.g., Von Oettingen et al.,
1942) are almost 50 years old (see Table 7-13). The estimated dose-response relationships
for the acute effects of single short-term exposures to toluene (U.S. EPA, 1983b) are
summarized below.
<37 ppm: Probably perceptible to most humans.
50 to 100 ppm: Subjective complaints (fatigue or headache) but probably no
observable impairment of reaction time or coordination.
100 to 300 ppm: Detectable signs of incoordination during exposure periods
up to 8 hours.
200 ppm: Mild throat and eye irritation.
300 to 800 ppm: Gross signs of incoordination may be expected during
exposure periods up to 8 hours.
400 ppm: Lacrimation and irritation to the eyes and throat
7-58
-------
TABLE 7-13
DOSE-RESPONSE RELATIONSHIP FOR 8-HOUR EXPOSURES TO
TOLUENE
(n = 3)
0 mg/rrH (control) - occasional moderate tiredness explained by lack of physical
exercise, unfavorable illumination, and monotonous noise from fans.
188 mg/m^ no subjective complaints from one subject and drowsiness with very mild
headache in the second subject. No after-effects.
375 mg/m^ moderate fatigue and slight headache on one occasion. No after-effects.
750 mg/m3 muscular weakness, confusion, paresthesias of the skin, impaired
coordination, dilation of pupils, impaired light accommodation, repeated
headache, and nausea at the end of exposure. After-effects included fatigue,
general confusion, moderate insomnia, and restless sleep.
1,125 mg/m3 severe fatigue, headache, muscular weakness, incoordinate, slight pallor.
After-effects were fatigue, headache, skin paresthesias, and insomnia.
1,500 mg/m3 fatigue, mental confusion, headache, skin paresthesias, muscular weakness,
dilated pupils. After-effects were fatigue, headache, skin paresthesias, and
insomnia.
2,250 mg/m3 extreme fatigue, mental confusion, exhilaration, nausea, severe headache,
and dizziness after 3 hours exposure. Eight-hour exposures show
incoordination and staggering gait. After-effects included nervousness and
some confusion.
3,000 mg/m3 severe fatigue, extreme nausea, confusion, lack of self-control, considerable
incoordination, and staggering gait after 3-hours exposure. After-effects
included moderate to severe insomnia lasting several days.
SOURCE: Von Oettingen et al., 1942.
7-59
-------
1,500 ppm: Probably not lethal for exposure periods up to 8 hours.
> 4,000 ppm: Would probably cause rapid impairment of reaction time and
coordination. Exposures of 1 hour or longer might lead to
narcosis and possible death.
10,000 to 30,000 ppm: Onset of narcosis within a few minutes. Longer exposures
may be lethal.
Wahlberg (1984) applied 1.5 and 0.1 ml of pure toluene (and other solvents) to the
forearms of human volunteers. The 1.5 ml toluene application involved the use of a glass
ring fastened to the forearm with rubber bands. The 0.1 ml toluene was applied by pipette
and allowed to evaporate. The 1.5 ml application resulted in an immediate increase in
blood flow (a measure of erythema), whereas a similar exposure to methyl ethyl ketone,
ethanol, propylene glycol, and water did not. The 0.1 ml application did not alter blood
flow, presumably due to evaporation.
Anderson et al. (1983b) noted mild nose and eye irritation, as well as symptoms of
headaches, dizziness, and feelings of intoxication in subjects exposed to 375 mg/m^ of
toluene for 6 hours. No adverse effects were noted during 6-hour exposures to either 150
mg/m3 or 37.5 mg/m^, although all three exposures were associated with a perceived
deterioration of the air quality. No significant changes in nasal mucus flow or pulmonary
function parameters were observed in the Anderson study. No statistically significant
decrements in psychometric performance tests were noted in subjects exposed to 375
mg/m3 of toluene for 6 hours (Anderson et al., 1983). The authors did note, however, that
the subjects considered the tests to be more difficult and strenuous during exposures to 150
mg/nA 37.5 mg/m^, or clean air.
The lowest acute effect level for toluene was reported in a Russian article. Gusev
(1965, as cited in WHO, 1981) reported changes in electroencephalograhic activity in
subjects exposed to 1 mg/m^ of toluene for 5 to 6 minutes. Exposure to 0.6 mg/m^ had no
effect In a review of this study, the World Health Organization (1981) concluded that the
methodology and health significance of the results need further study.
Persons who deliberately inhale toluene to produce intoxication may be exposed to
concentrations up to 10,000 ppm for several minutes (Hinman, 1984).
7.3.1.2 Animal Studies
As with humans, the most pronounced acutely toxic effects of toluene in study
animals are on the CNS. Acute inhalation exposure to high concentrations of toluene has
been associated with CNS depression as evidenced by reduced activity. Exposure to levels
7-60
-------
below 1,000 ppm (~3,800 mg/m3) has minimal to no effect on gross behavioral
manifestations, although more sensitive assay methods (e.g., detection of changes in
cognition and brain neuromodulator levels) have indicated effects at lower levels. Toluene
is only slightly toxic by the oral route: the oral L£>50 ranges from 6.0 to 7.5 g/kg. The
inhalation LCso has been reported to be 500 to 700 ppm (-1,900 to 2,600 mg/m3) in mice
and 4,000 ppm fclS.OOO mg/m3) in rats (U.S. EPA, 1983b).
There is some evidence that repeated exposures to high toluene concentrations may
alter tolerance to certain neurobehavioral effects. Rats exposed to 10,000 ppm of toluene,
two times per day for 15 minutes, for 6 weeks developed tolerance to ataxia, hind limb
myoclonus, and inhibition of rearing, but reversed tolerance to headshakes and increased
locomotor activity (Himnan, 1984). Thus, for high-exposure situations, the relationship
between duration and neurological effect can be variable. The dynamics of toluene
exposure on neurotoxicity at low exposures have not been well studied.
Taylor and Evans (1985) studied the effects of toluene exposure on behavioral
changes and expired carbon dioxide levels in female Macaque monkeys (Taylor and Evans,
1985). Carbon dioxide levels were measured to provide a sensitive indicator of combined
behavioral, irritant, sensory, and metabolic effects. The monkeys (six per exposure group)
were exposed to 0, 100, 200, 500, 1,000, 2,000, 3,000, and 4,500 ppm of toluene for 20
to 50 minutes. Monkeys exposed to concentrations of 2000 ppm and above showed
significantly reduced performance on cognitive and motor skills tests, particularly tests
involving attention and visual-motor abilities. The animals' response time, while being
significantly impaired following a 25 minute exposure to 2,000 ppm, displayed a tolerance
as it improved during the second half (25 to 50 minutes) of the session. However, no
tolerance was observed for the other tests of cognitive/motor function (e.g., accuracy in
delayed matching, probability of responding, responding during delay). Also, no
behavioral measure exhibited either cumulative effects or tolerance to 4,500 ppm of toluene
during two 3-day exposures.
Expired CO2, the most sensitive measure, increased following a 20 minute
exposure to 100 ppm of toluene. The response was an invened U-shaped curve. The
authors noted that this type of curve also described behavioral changes in the response rates
of pigeons and rats during acute toluene exposure, suggesting behavioral stimulation at
lower doses and behavioral sedation at higher doses.
7-61
-------
7.3.2 Effects of Subacute and Subchronic Exposure to Toluene
7.3.2.1 Human Studies
Wilson (1943, as reported in U.S. EPA, 19835) reported the following effects of
repeated occupational toluene exposure over a period of several weeks in 100 workers.
• 50 to 200 ppm (approximately 60 percent of the patients) - Headache, lassitude,
and loss of appetite. These symptoms were so mild that they were considered to
be due primarily to psychogenic and other factors rather than to toluene fumes.
• 200 to 500 ppm (approximately 30 percent of the patients) - Headache, nausea,
bad taste in the mouth, anorexia, lassitude, slight but definite impairment of
coordination and reaction time, and momentary loss of memory.
• 500 to 1,500 ppm (approximately 10 percent of the patients) - Nausea, headache,
dizziness, anorexia, palpitation, and extreme weakness. Loss of coordination
was pronounced and reaction time was definitely impaired.
No other acceptable data were found regarding the subchronic effects of toluene in
humans.
7.3.2.2 Animal Studies
Neurological and hematological effects have been observed in subchronic animal
studies on toluene. Slight nasal and ocular irritation and motor incoordination that preceded
paralysis of the extremities were observed in two dogs exposed to 2,000 ppm pure toluene
by inhalation for 4 months and subsequently to 2,660 ppm for 2 months. Death occurred
on days 179 and 180. No effect was seen on body weight gain, bone marrow, adrenal,
thyroid, or pituitary glands. Pulmonary congestion, hepatic hemorrhage, a decrease of
lymphoid follicles, hemosiderosis in the spleen, hyperemic glomeruli of the kidney, and
albumin in the urine were observed (Fabre et al., 1955). No deleterious effects on
hematology or organ pathology were noted after subchronic treatment of rats (Von
Oettingen et al., 1942) and of rats, guinea pigs, dogs, and monkeys (Jenkins et al., 1970;
Smyth and Smyth, 1928) at concentrations of 200 and 1,085 ppm toluene, respectively.
Rats and mice exposed to 7,500 mg/m^, 8 hours/day, for 1 to 3 weeks showed
slight increases in liver weight relative to body weight (Ungvary et al., 1982). However,
very little subcellular damage was observed. The authors attributed the increase in liver
7-62
-------
weight to a functional hypertrophy caused by proliferation of the endoplasmic reticulum.
Slight reductions in the body weights of male rats were observed after 3-month exposures
to 3,750 mg/m3 of toluene (CUT, 1980; Rhudy et al., 1978).
Animal data regarding the pulmonary effects of toluene are very limited. Effects of
toluene have been investigated, however, in efforts to assess the effect of this chemical on
pulmonary host defenses (Aranyi et al., 1985). These effects have been found to be
sensitive indicators of pulmonary toxicity in investigations of common atmospheric
pollutants (e.g., ozone).
Aranyi et al. (1985) investigated the effects of single and repeated exposures to 1,
2.5, 5, 10, 25, 50,100, 250, and 500 ppm of toluene on pulmonary host defenses. CD-I
mice (135 to 405 animals per group) were exposed to toluene for 3 hrs/day, for 1 day.
Mice were also exposed to 1 ppm of toluene for 5 days (3 hours/day) and 20 days. A
single exposure to toluene at or above 2.5 ppm caused a dose-related increase in mortality
from respiratory infections in mice. Exposure to 1 ppm did not increase susceptibility after
a 1,5, or 20 exposures. Bactericidal activity was decreased in mice following a single
exposure to 100 ppm and above. The response at lower concentrations was variable,
although a single exposure to 2.5 ppm and 5 exposures to 1 ppm produced significant
decreases. The authors concluded that susceptibility to respiratory infections occurred in
mice following a single 3-hour exposure to toluene concentrations at or above 2.5 ppm,
and that the 1 ppm exposure was without effect.
Neurotoxic and hematotoxic effects have been observed in mice exposed to low
toluene concentrations for 20 days. Horiguchi and Inoue (1977) exposed adult male HA2
mice (six mice, weighing 20 grams, per exposure group) to 0,1, 10,100, and 1,000 ppm
of toluene for 6 hours/day for 20 days. Cumulative wheel-turning rounds (a measure of
spontaneous wheel-turning behavior in mice) was significantly depressed after the sixth to
eighth exposure day in the groups of mice exposed to 10 to 1000 ppm (relative to air-
exposed controls), and after the tenth day in the group of mice exposed to 1 ppm of
toluene. The authors stated that the neurobehavioral effects of the 10 ppm toluene exposure
in the mice were approximately the same as those of a 100 ppm exposure to benzene. No
follow-up measurements were taken to investigate the reversibility of this effect. Horiguchi
and Inoue (1977) also measured the hematological changes that occurred in association
with these toluene exposures. They found that red blood cell counts were depressed in the
10 ppm (15 percent decrease) and 1,000 ppm (24 percent decrease) exposure groups (46
percent decrease in the mice exposed to 1,000 ppm, and 25 percent decrease in the mice
exposed to 10 ppm) and that thrombocyte counts were depressed in all exposure groups
except the controls and 1 ppm exposed animals. The 100 ppm exposure (but not lower
7-63
-------
exposures) was also associated with slight decreases in the density of bone marrow cells,
in megakaryocytes, and in elements of red cells without maturity disturbance. White blood
cell counts were transiently elevated in all exposure groups midway through exposure.
They approached control values by the end of the exposure except in the group exposed to
1000 ppm, in which they dropped to about 50 percent of the control values. No follow-up
measurements were taken to investigate the reversibility of this effect.
Von Euler et al. (1989) exposed adult male Sprague-Dawley rats (weighing 600
grams) to 80 ppm of toluene for 6 hours/day, 5 days/week, for 3 months. The rats were
relatively old (15 months) compared to other experimental animals. The authors chose
these animals because they were thought to have increased vulnerability to toluene due to
their reduced synaptic plasticity. The authors postulated that this condition also exists in
adult humans.
Norepinephrine levels tended to be reduced in the hypothalamus while serum
prolactin levels were increased following toluene exposure. The exposure significantly
altered receptor binding in selective regions of the brain. Binding of the peptide
neurotensin was significantly reduced in parts of the orbital cortex. However, binding of
etorphine in the nucleus accumbens and of vasoactive intestinal polypeptide in parts of the
medulla oblongata was significantly increased. Acute treatment with an irreversible
monoamine receptor antagonist (EEDQ) increased the binding of neurotensin in the orbital
cortex of toluene exposed animals as compared to reduced binding in the air-exposed
controls treated with the antagonist Toluene treatment obliterated the increased vasoactive
intestinal polypeptide binding in the area postrema that was induced by EEDQ. These last
two findings indicate that toluene may interfere with receptor-receptor binding.
The exposed animals did not differ from the controls in measures of two behavioral
effects: locomotor activity and passive avoidance. Although no significant changes were
observed between the exposed and control animals for passive avoidance, there was some
suggestion that the reaction times of the exposed animals were slower than those of the
controls. Also there was a lag time between the last exposure and the behavioral tests.
Locomotor activity was assessed 14 days and 21 days following the last toluene exposure.
Passive avoidance was measured 2 weeks after the last exposure. Thus, while these
behavioral tests were negative, it is possible that toluene could have induced reversible
neurobehavioral changes during the exposure or that more sensitive tests could have
revealed subtler, more persistent effects.
Toluene exposure influenced protein phosphorylation in the fronto-parietal cortex.
Exposure induced a decrease in calcium-regulated protein phosphorylation and an increase
in cAMP-regulated protein phosphorylation (Von Euler et al., 1989). The authors noted
7-64
-------
the importance of protein phosphorylation processes for synaptic transmission and
metabolic events that are likely to be important in learning, memory, and other types of
high mental functions. They also noted that the passive avoidance test may have been too
simple to reveal possible alterations in learning and memory.
These results indicate that toluene causes selective alterations between peptide and
monoamine receptors in affected areas of the brain, as well as changes in neurotransmitter,
protein phosphorylation, and serum prolactin levels, at levels similar to occupational
standards (80 ppm) in many European countries (Von Euler et al., 1989). The significance
of these findings is unclear. They were not well correlated with behavioral changes, and
no comparative dose-response measurements were taken.
No adverse effects were reported when rats were orally administered toluene at a
dose of 590 mg/kg/day for 6 months (Wolf et al.. 1956).
7.3.3 Effects of Chronic Exposure to Toluene
7.3.3.1 Human Studies
Evidence of neurotoxicity was observed following repeated occupational exposures
to toluene over a period of several years at concentrations of 200 to 400 ppm (£740 to
1,500 mg/m3) (U.S. EPA, 1983b). Impaired performance on tests for intellectual and
psychomotor ability and muscular function have resulted following chronic exposure to
mixtures of solvent vapors predominantly containing toluene at levels of 30 to 100 ppm
fclOO to 400 mg/m3) (U.S. EPA, 1983b). Residual or permanent CNS effects can result
from prolonged abuse of toluene or solvent mixtures containing toluene (U.S. EPA,
1983b).
Hanninen et al. (1976), Raitta et al. (1976, as cited in U.S. EPA, 1983), and
Seppalainen et al. (1978) studied a cohort of automobile painters exposed to a solvent
mixture containing 112 mg/m3 (30 ppm) of toluene for 15 years. The composition of the
mixture is described in Table 7-14. The actual exposure concentrations were estimated to
be approximately 30 percent of the threshold limit value (TLV) for the solvent mixture
(Seppalainen, 1978). Symptoms of CNS depression and impaired performance on tests of
intelligence, memory, and visual and verbal ability were also noted. In addition,
Seppalainen et al. (1978) observed slower nerve conduction velocities in several of the
workers relative to the reference control population. Electroencephalograms (EEGs) were
not significantly different between the exposed and control groups. Yet, the EEGs in both
groups were abnormally high (32 percent in the exposed and 37 percent in the controls)
compared to historical controls (10 percent) (Seppalainen et al., 1978). Raitta et al. (1976)
7-65
-------
TABLE 7-14
MEAN CONCENTRATIONS OF ORGANIC SOLVENTS IN THE
BREATHING ZONE OF 40 CAR PAINTERS
Solvent Mean concentration
(ppm)
Toluene
Xylene
Butyl acetate
White spirit
Methyl isobutyl ketone
Isopropanol
Ethyl acetate
Acetone
Ethanol
Total
30.6
5.8
6.8
4.9
1.7
2.9
2.6
3.1
2.9
Percentage
oftheTLV3
15.3
5.8
4.5
2.5
1.7
0.7
0.7
0.3
0.3
31.8
TLVa
(ppm)
200
100
150
200
100
400
400
1,000
1,000
aTLV = threshold limit value in Finland in 1976.
NOTE: Sampling period = 1 hr, number of car repair garages = 6; number of samples = 54
SOURCE: Seppalainen, et al., 1978.
7-66
-------
observed greater incidences of lens opacities and nuclear sclerosis (possible early indicators
of cataract development) in the solvent exposed workers; in 27 instances, more lens
changes were apparent in the exposed group than in the controls, while the converse was
true in only four instances.
Using a different cohort of spray painters, Elofsson et al. (1980) found significant
decreases in reaction times, manual dexterity, perceptual speed, and short-term memory.
In this exposure group, the average toluene concentration was approximately 15 mg/m-*.
Slight increases in the incidence of lens changes were also noted. No differences were
noted with respect to verbal, spatial, or reasoning abilities between the control and exposed
groups. As with the previous cohort, the actual exposure concentrations in the Elofsson
study were estimated to range between 20 and 30 percent of the TLV for the solvent
mixture (Elofsson et al., 1980).
In an effort to separate the nervous system effects of toluene from the effects of
solvent mixtures, Iregren (1982) compared a cohort of automobile painters to a cohort of
rotogravure printers exposed exclusively to toluene. Exposure to toluene at concentrations
near the Swedish TLV of 300 mg/m3 resulted in significant decreases in simple reaction
time, through other parameters did not differ significantly from controls. No dose-
response correlations were found between test performance and length of employment.
Changes in leucocyte enzyme activities and lymphocyte morphology have been
observed in the case of exposure ranging from 160 to 3000 mg/m3 of pure toluene (Cohr
and Stockholm, 1979). Prolonged occupational exposure to toluene appears to increase
lysosomal activity in neutrophils relative to unexposed control populations (Friborsko,
1973, as cited in U.S. EPA, 1983b). Matsushita et al. (1975, as cited in U.S. EPA,
1983b) observed a significantly increased number of "Mommsen's" toxic granules in the
neutrophils of workers chronically exposed to approximately 375 mg/m3 of toluene.
7.3.3.2 Animal Studies
A chronic inhalation toxicity study was conducted for the Chemical Industry
Institute for Toxicology (CUT) for 24 months in male and female Fischer 344 rats.
Toluene was administered by inhalation at 30,100, or 300 ppm (113,377, or 1,130
mg/m3, respectively) to 120 rats of each sex for 6 hours/day, 5 days/week. An equal
number of animals served as controls. Clinical chemistry, hematology, and urinalysis
testing was conducted at 18 and 24 months. None of the parameters measured at the
termination of the study differed significantly from control values except for a dose-related
reduction in hematocrit values in females exposed to 10 ppm and 300 ppm of toluene.
Also, the mean corpuscular hemoglobin concentration was slightly, but significantly,
7-67
-------
increased in females exposed to 300 ppm toluene (CUT, 1980). This was a well-conducted
study; a large number of animals/sex were tested in each of the three dose groups, many
parameters were studied, and interim necropsies were performed. This study is the only
acceptable chronic toxicity study on toluene found in the literature.
7.4 XYLENE
7.4.1 Effects of Acute Exposure to Xvlene
7.4.1.1 Human Studies
Early reports on the acute toxic effects of xylene generally involve incidences of
exposure to solvents and thinners in the workplace. Since solvents and thinners also
contain appreciable amounts of benzene and toluene as well as other aromatic and
nonaromatic compounds, the demonstrated effects attributed to xylene in these early
investigations are questionable. NIOSH (1975) has reviewed and evaluated much of the
early literature. More recently, Von Burg (1982) summarized the adverse effects of xylene
exposure.
Signs and symptoms following acute inhalation of xylene vapors include fatigue,
headache, nausea, dizziness, dry throat, irritability, pulmonary edema, and coma. Xylene
appears to be more toxic in this regard than either benzene or toluene (Sandmeyer, 1983).
Ingestion of xylene will cause severe gastrointestinal distress (Sandmeyer, 1983). If
aspirated into the lungs, xylene will cause chemical pneumonias, pulmonary edema, and
hemorrhage (Sandmeyer, 1983). The nervous system effects associated with acute xylene
poisonings are similar to those produced by exposure to benzene, toluene, and other
anesthetic gases (Sax, 1984): a preliminary stage of excitation followed by depression and,
if exposure continues, coma and death. Death may occur from respiratory arrest or
myocardial sensitization (Sandmeyer, 1983). Irritation of the mucous membranes of the
eyes and upper respiratory airways and renal and hepatic impairment have also been
reported. Direct contact with the skin can produce a burning sensation.
In humans, an exposure to 110 to 460 ppm (472 to 1,980 mg/m3) of xylene for 3
to 5 minutes has been associated with irritation to the eyes, nose, and throat (Sandmeyer,
1983). Respiratory irritation (concentrations and exposure durations not specified) has
been associated with occupational exposure to xylene, and the chemical has been classified
as a cilia toxin and a mucus coagulating agent (Sandmeyer, 1983).
7-68
-------
Conjunctivitis and corneal bums of the eyes, as well as irritation and defatting of
the skin leading possibly to dryness, cracking, blistering, or dermatitis, has been observed
in humans following direct contact with xylene (Sandmeyer, 1983).
Goldie (1960) reported that eight men experienced headache, vertigo, gastric
discomfort, dryness of the throat, and feelings of slight drunkenness while painting the
interior of a gun tower on a ship. The paint contained 80 percent xylene and 20 percent
methylglycolacetate; at the time the symptoms appeared, a strong smell of xylene was
noticed within the tower. One painter, who had been working with paints for 2 months
and had a history suggesting an epileptic focus, experienced pronounced cerebral
symptoms that were interpreted as an epileptiform seizure 1 hour after he left the
workplace. Hospital examination revealed normal reflexes and a normal
electroencephalogram. The author suggested that xylene exposure may act as an
aggravating or provoking agent in patients with latent epilepsy. Arthur and Curnock
(1982) described an adolescent boy with a history of absence seizures who regularly
suffered major and minor epileptic seizures while using a xylene-based modeling glue.
Morley et al. (1970) reported three cases of xylene poisoning after the subjects
(three men) were exposed to prolonged inhalation of paint fumes while working in an
enclosed space. Analysis showed that xylene comprised more than 90 percent of the
solvent in the paint, and the estimated exposure to xylene was 10,000 ppm. One man died
and the autopsy revealed congested lungs and liver. The brain showed microscopic
hemorrhaging in both gray and white matter, and there was evidence of anoxic neuronal
damage. The other two men remained unconscious for up to 24 hours; after regaining
consciousness, they were mentally confused and amnesic for events that occurred during
the 24 hours preceding the incident. Both men showed some evidence of liver cell damage,
and one showed evidence of impaired renal function. Both men subsequently recovered.
Klaucke et al. (1982) reported an incident in which 15 employees of a pathology
laboratory at a small hospital were exposed to xylene vapors after about 1 L xylene had
been discarded in a sink. Air concentration and duration of exposure were not reported.
The symptoms reported by the workers included headache, nausea, vomiting, vertigo,
dizziness, and irritation of the eyes, nose, and throat.
7.4.1.2 Animal Studies
NTP (1986a) conducted acute toxicity studies of mixed xylenes in Fischer 344 rats
and B6C3Fi mice. Groups of five animals of each sex and species were administered
single oral doses of xylene in com oil, by gavage, at levels of 500,1,000, 2,000,4,000,
or 6,000 mg/kg. All rats that received 6,000 mg/kg and three of five male rats that received
7-69
-------
4,000 mg/kg died within 48 hours. Three of five male mice and four of five female mice
receiving 6,000 mg/kg died within 32 hours. Lack of coordination, prostration, and
hunched posture were observed within 24 hours in rats receiving 4,000 mg/kg, and
tremors and/or slow breathing were observed in mice surviving at 6,000 mg/kg and in
those receiving 4,000 mg/kg. An LCso in rats exposed to 6,350 ppm (27,305 mg/m^) for
4 hours has been reported (Sandmeyer, 1983).
DiVincenzo and Krasavage (1974) evaluated solvent-induced liver injury using
serum ornithine carbamyl transferase (OCT) activity in male guinea pigs. Solvents were
administered intraperitoneally and enzyme activity was measured after 24 hours. After a
single intraperitoneal injection of 1,000 mg/kg of xylene, the OCT activity of four guinea
pigs averaged 18.4 International Units (IU) compared to a mean OCT activity of 2.0 IU in
the controls. At a dose of 2,000 mg/kg, three of four animals died and the mean OCT
activity in one surviving animal was 25.2 IU. Microscopic examination of the liver
revealed moderate accumulation of lipid in the hepatocytes at both dose levels.
Hepatocellular necrosis was not observed.
7.4.2 Effects of Subacute and Subchronic Exposure to Xvlene
7.4.2.1 Human Studies
No information was found in the available literature on the effects of subchronic
exposure of humans to xylene.
7.4.2.2 Animal Studies
Jenkins et al. (1970) conducted subchronic inhalation toxicity studies with o-xylene
in groups of 15 rats, 15 guinea pigs, 2 dogs, and 3 monkeys. Animals of each species
were exposed to either 779 ppm (3,358 mg/m3) for 8 hours/day, 5 days/week for 30
repeated exposures, or to 78 ppm (337 mg/m^) continuously for 90 days. Pre- and post-
exposure body weight, hematology, and mortality data were reported. In the 30-day study,
mortality occurred in 3 of 15 rats and 1 of 3 monkeys; 1 dog from the 30-day exposure
regimen exhibited tremors. In the 90-day study, death occurred in one rat. No significant
changes were observed in body weight or hematologic data. Microscopic examinations did
not reveal any treatment-related changes.
Carpenter et al. (1975) conducted a 13-week inhalation toxicity study with vapors of
mixed xylenes in male rats and beagle dogs. The test material contained 65.0 percent m-,
7.6 percent o-, and 7.8 percent p-xylene, and 19.27 percent ethylbenzene. Groups of 25
rats or 4 dogs were exposed to xylene vapors at measured mean concentrations of 0.77,
7-70
-------
2.0, 3.5 mg/L (180,460, and 810 ppm) for 6 hours/day, 5 days/week, for 13 weeks.
Clinical pathology (hematology, blood chemistry, and urinanalysis) and histopathology
evaluations were performed on groups of four rats sacrificed at 15 days, 35 days, and at
termination, and on 4 dogs at termination. No treatment-related effects were observed in
any of the test parameters. No gross or microscopic lesions were attributed to inhalation of
the mixed xylenes.
NTP (1986a) conducted subchronic studies on mixed xylenes in Fischer 344 rats
and B6C3F1 mice. Groups of 10 male and 10 female rats were treated by gavage with
mixed xylenes in corn oil at 0,62.5, 125, 250, 500, or 1,000 mg/kg/day, 5 days/week, for
13 weeks. Groups of 10 male and 10 female mice were similarly treated at dosages of 0,
125,250,500, 1,000, or 2,000 mg/kg/day. There were no mortalities or signs of toxicity
in the rats. At 13 weeks, mean body weights were 15 and 8 percent lower in male and
female rats receiving 1,000 mg/kg, respectively, than in the corresponding vehicle controls.
Two female mice treated at 2,000 mg/kg died before the end of the study. Weakness,
lethargy, short and shallow breathing, tremors, and paresis were observed in male and
female mice receiving 2,000 mg/kg. These signs persisted for 15 to 60 minutes after
dosing. Mean body weight gains of the high-dose male and female mice were 7 and 17
percent lower, respectively, than those of the corresponding vehicle control group. No
treatment-related gross or microscopic pathologic lesions were observed in rats or mice.
Early studies found associations between xylene exposure and hematological
effects. Mice exposed to 10 to 50 ppm of xylene, 2 to 4 hours/day, for 1 year exhibited
hematologic and immunologic effects (Sandmeyer, 1983). Other, more recent studies have
generally not found effects at xylene exposures below 1,000 ppm (Sandmeyer, 1983). A
possible explanation for this disparity may have been benzene contamination of the xylene
test substance in the earlier studies (Sandmeyer, 1983).
7.4.3 Effects of Chronic Exposure to Xvlene
7.4.3.1 Human Studies
Several reports were found in the literature on the effects of long-term occupational
exposure to xylene. However, all of these studies were complicated by exposure to
multiple solvents; thus, it is difficult to make any accurate interpretations of the data.
Because many of the solvents identified in these studies (especially the aromatic
compounds) are also contained in gasoline, summaries of these studies are included in the
gasoline chronic health effects section.
7-71
-------
7.4.3.2 Animal Studies
A 2-year bioassay of mixed xylenes (60.2 percent m-xylene, 13.6 percent p-xylene,
9.1 percent o-xylene, and 17.0 percent ethylbenzene) was conducted by the NTP (1986a).
Groups of 50 male and 50 female Fischer 344 rats and 50 male and 50 female B6C3F1
mice were administered, by gavage, xylene in corn oil. The chemical was administered to
rats 5 days/week for 103 weeks at doses of 0,250, or 500 mg/kg/day and to mice at 0,
500, or 1,000 mg/kg/day. After week 59, average body weights of the high-dose male rats
were decreased by 5 to 8 percent compared with controls; this was considered to be an
indication of slight toxicity. No differences in body weight were noted for female rats or
male and female mice. There was a dose-related decrease in survival of male rats; the
difference between the high-dose and the control group was statistically significant. Many
of the high-dose deaths were attributed to gavage error, however, the possibility that the
rats resisted dosing by gavage because of a behavioral effect was not ruled out. No
significant differences in mortality were observed for female rats or male and female mice.
High-dose male and female mice were observed to be hyperactive for 5 to 30 minutes after
dosing from week 4 to 103. There was no significant increase in the incidence of
microscopic lesions at any site in either species.
Tatrai et al. (1980,1981) reported that chronic exposure to o-xylene via inhalation
at 1,090 ppm (4,750 mg/m3) for 8 hours/day, 7 days/week, for a 1-year period caused
several toxic responses in male CFY rats. Water and food consumption of xylene-exposed
rats were increased throughout the study. Body weight gains of the treated rats, however,
were reported to be decreased. Hepatic enzyme induction and hepatoxicity were observed
in the treated rats. Hepatic cytochrome P-450 and B-5 concentrations, NADPH-
cytochrome reductase activity, aniline hydroxylase activity, and aminopyrin-N-demethylase
activity were increased, and bromosulfophthalein retention time in the liver was decreased.
Hepatomegaly was reported; however, the livers of the exposed rats appeared histologically
normal when examined by light microscopy. Ultrastructural examination of centrilobular
hepatocytes revealed moderate proliferation of smooth endoplasmic reticulum and depletion
of glycogen, with occasional mitrochondrial damage and increased number of autophagous
bodies and peroxisomes. Proliferation of the endoplasmic reticulum is consistent with
increased enzymes of the mixed function oxidase system. Tatrai et al. (1980,1981)
concluded that o-xylene was an inducer of hepatic enzyme metabolism; liver enlargement
was attributed to functional hypertrophy during chronic exposure.
7-72
-------
7.5 SUMMARY
The toxicity of gasoline blends and its components, including benzene, toluene, and
xylene as well as selected alkanes and aromarics, has been investigated for both short- and
long-term exposures in humans and animals. The human studies generally involve
occupational exposures to relatively high concentrations of gasoline or mixtures of other
solvents. Few long-term human studies on unleaded gasoline blends are available. The
human data on gasoline suggest neurological and pulmonary effects with short- and long-
term exposures. Animal studies suggest a sex and species specific nephrotoxic effect from
exposure to a component of gasoline, trimethylpentane; however, possible renal toxicity is
occasionally reported in female rats and mice. Allergic contact dermatitis, a common
response to many organic solvents, is also reported for gasoline. Behavioral studies in
animals show differences between exposed and controls. NOAELs and LOAELs for
gasoline generally range from 10 to 100 ppm.
With respect to the aromatic components of gasoline (benzene, toluene, and
xylene), the well-established hematological effects of benzene have been studied in detail in
both animals and humans. The NOAELs for benzene action on the blood-forming organs
are in the range of 10 ppm. Other actions for the aromatics include neurological responses
at higher doses and irritation of the skin and respiratory system. All components discussed
have indications of chronic toxicity in both humans and animals with exposures
approaching ambient or occupational levels.
7-73
-------
8. REPRODUCTIVE AND DEVELOPMENTAL TOXICITY
8.1 INTRODUCTION
According to the U.S. EPA (1987a), approximately 3 percent of newborn children
are found to have one or more significant congenital malformations at birth, and by the end
of the first year, about 3 percent more are found to have serious developmental defects. Of
these human developmental defects, it is estimated that 20 percent are of known genetic
transmission, 10 percent are attributable to known environmental factors, and the remainder
result from unknown causes. Approximately 7.4 percent of children have reduced birth
weight at birth. Fetal weight is a sensitive and reversible fetotoxic effect It may also be
indicative of potentially more significant, irreversible developmental toxicity.
The central nervous system is a target of gasoline toxicity in adults, it is possible
that toluene may also produce neurologic effects during development. The brain is the first
organ to differentiate, as well as the fastest growing throughout gestation and childhood.
After birth, the brain doubles in size during the first 6 months, and doubles again by about
4 years of age. This is accomplished mainly through increases in the size and branching of
neural processes, by increases in the number of glia, and by producing myelin. Thus, the
brain is anticipated to be a sensitive target organ during both prenatal and postnatal
development. Few attempts, however, have been made to study the effects of gasoline or
its components on neurological development
Reproductive toxicity may be defined as the adverse effects observed in the male
and female reproductive systems in organisms exposed to chemical or physical agents. The
adverse effects in females may include alterations in ovary and accessory glands, endocrine
control, sexual behavior, onset of puberty, fertility, gestation, parturition, lactation, or
premature reproductive senescence; in males, these alterations may occur in the testis and
accessory glands, endocrine control, sexual behavior, and reproductive outcomes.
Developmental toxicity may be defined as the adverse effects observed in the
developing organism from parental exposure prior to conception, during prenatal
development, or postnatally to the time of sexual maturation. The adverse developmental
effects may include death of the developing organism, structural abnormalities, altered
growth, and/or functional deficiency.
This chapter will summarize the major human and animal studies that have
investigated the potential for gasoline, benzene, toluene, and xylene to cause adverse
reproductive or developmental effects.
8-1
-------
8.2 GASOLINE
8.2.1. Reproductive Effects in Humans
Anecdotal studies from the Soviet Union, available in the abstract form only, of
female workers occupationally exposed to gasoline and possibly other organic solvents
including benzene, toluene, and xylene, reported a higher frequency of gynecological
disorders including chronic inflammation of the uterus and amenorrhea (Vozovaya, 1974;
Zhibura, 1974; Mirinski, 1979). Zhibura (1974) suggested that gasoline exposure possibly
decreased ovarian estrogen production and disturbed pituitary secretion, which caused the
menstrual disorders.
Michon (1965, as cited in U.S. EPA, 19860 compared a group of 500 Polish
women, aged 20 to 40 years, employed in a leather and rubber shoe factory, who were
exposed to a mixture of benzene, toluene, and xylene vapors to a control group consisting
of 100 women from the same plant. Although the exposure levels were not reported, the
study reported that the levels were no greater than permissible occupational exposure limits
established in Poland (31 ppm of benzene; 67 ppm of toluene, and 58 ppm of xylene). The
results of the study indicated that the exposed group exhibited prolonged and more intense
menstrual bleeding than was found in a non-exposed group.
8.2.2 Developmental Effects in Humans
Exposure to organic solvents during pregnancy was studied by Holmberg (1979) in
mothers of children from Finland with and without congenital central nervous system
effects (case mothers and controls, respectively). Of 120 cases identified and their
controls, 14 case mothers and 3 control mothers had been exposed to organic solvents
including benzene, toluene, xylene, and mixed aromatic hydrocarbons. One mother was
exposed to gasoline of an unspecified nature in the metal products manufacturing industry.
Twelve of the 14 case mothers were exposed during work. However, the duration of
exposure was not reported. The study found that significantly more case mothers than
control mothers were exposed to organic solvents during the first trimester of pregnancy
(p<0.01). Of the 14 case mothers exposed to organic solvents, five delivered stillborn
offspring with anencephaly (36 percent). Other CNS defects noted in children from these
case mothers included hydrocephalus and meningomyelocele.
A gasoline-abuse case study reported teratogenic effects in children of mothers
intentionally exposed to relatively high concentrations of gasoline vapor (Hunter et al.,
1979, as cited in U.S. EPA, 1988). However, the results of the study are difficult to
8-2
-------
assess because of the possibility that these exposures were confounded by alcohol abuse by
the mother.
There is some indication that maternal exposure to fat soluble solvents (such as
xylene, one of the many aromatic compounds contained in gasoline) may be a risk factor in
the development of sacrococcygeal agenesis in humans. Kucera (1968) identified nine
cases of this syndrome from a registry of malformations in Czechoslovakia for the years
1959 to 1966. The mothers of six of these infants were exposed to chemicals during
pregnancy. In five of the six, exposure was to organic solvents. Because this was not a
case-control study, and information on exposure was not presented, it is difficult to
evaluate the lexicological significance of these findings. On the other hand, the etiology of
most developmental effects in human beings are not known.
8.2.3 Reproductive Effects in Animals
Several Soviet studies have reported adverse reproductive effects from exposure to
gasoline or a mixture of gasoline and dichloroethane on laboratory animals, but these
studies are only available in abstract form. The following data are available from the
abstracts.
Male rats exposed to 300 mg/m3 of Grade BR-1 gasoline for 2.5 months showed
decreased spermatogenesis and symptoms described as general intoxication. Further
examination indicated that gasoline did not induce dominant lethal mutations in sperm
(Feller, 1972).
Similar findings were reported in a dominant lethal study sponsored by the
American Petroleum Institute. This study was conducted to determine if UG is able to
reach the developing sperm cells and induce genetic damage in CD-I mice exposed by
inhalation (Litton Bionetics, 1980). Male mice were exposed to either 400 or 1,600 ppm
gasoline vapor, 6 hours/day, 5 days/week for 8 weeks. Results indicated a non-significant
increase in pre- and post-implantation loss of embryos as compared to control.
Lipovskdi (1978) observed increased tonus and serotonin content, which accelerated
the rate of uterine myometrial contractions in rats that inhaled gasoline vapors (300 mg/m3)
for 4 hours/day for 30 to 34 days. The authors concluded that these results suggest a
disruptive effect of gasoline and gasoline vapors on the uterine contractility.
Abnormal excitation of the estrous cycle was studied by Lovchikov et al. (1978) in
419 sexually mature female rats exposed daily to 300 mg/ml gasoline vapors, 4 hours/day,
for 1 month. CNS excitability in unexposed rats (controls) decreased from diestrus to
proestrus, reaching its lowest level during estrous and increased to its maximum during
8-3
-------
metaestrous. In gasoline exposed rats, CNS excitability initially declined followed by
gradual restoration to the appropriate excitability level for a given stage of the cycle.
Vozovaya (1975a) found that exposure of rats to a mixture of dichloroethane and
gasoline (30:1200 mg/m^) for 6 months, 4 hours/day, decreased the frequency of
conceptions. Inhalation of dichloroethane vapors (57 mg/m^) or the mixture of
dichloroethane and gasoline decreased fertility and increased the rate of stillbirths. The
body weight of newboms and maternal weight gains were decreased. A decrease in
phagocytic activity of leukocytes and in muscular endurance of the pups was also
observed. The vapor mixture caused perinatal mortality in second-generation pups. In
addition, male offspring exposed in utero had decreased weight of the heart, spleen,
thyroid, adrenals, and testes. Exposure to 1,783 mg/m^ of gasoline vapors had no adverse
effect on the reproductive function of females.
Under the same test conditions, Vozovaya (1975b) reported that a 4-month airborne
exposure to a mixture of gasoline and dichloroethane resulted in abnormal sexual cycle,
blood and organ indices, immunological reactivity, and embryonal development in female
rats. Vozovaya reported that the effects from exposure to a mixture of gasolines and
dichloroethane were greater than from exposure to them individually.
The results of the studies by Lipovskii (1978), Lovchikov et al. (1978), and
Vozovaya (1975a, 1976) are inconclusive because the data and the validity of conclusions
cannot be ascertained from the available abstracts.
8.2.4 Developmental Effects in Animals
The only known teratogenicity study of gasoline vapor was sponsored by API in
which groups of 25 pregnant rats (CRL:COBS CD(SD)BR) were exposed to airborne
concentrations of atomized unleaded gasoline vapors at 0,400, or 1,600 ppm for 6
hours/day on days 6 through 15 of gestation (Litton Bionetics, 1978d). The vapor of the
test material was generated by metering it into a wanned, glass flask and passing
compressed air through it. This concentrated vapor was then diluted with room air as it
entered the dams' cages. Control dams were placed in similar cages, but were exposed to
room air only. Dams were weighed and food consumption was measured throughout the
gestational period. Dams were observed daily for changes in general appearance, behavior,
and condition. On day 20 of gestation, all dams were sacrificed, and visceral and thoracic
organs were grossly examined. The implantation sites, live and dead fetuses, and
resorption sites were examined. One-third of the fetuses were examined for changes in the
soft tissue of organs and the remaining for skeletal anomalies.
8-4
-------
No significant differences in body weight and food consumption between control
and treated pregnant rats was observed. The test material did not produce any effects in the
uterus of treated rats compared to controls and no abnormalities in the fetuses were
observed. However, according to the investigators, all fetuses of one litter in the highest
dose group (1,600 ppm) were "extremely small in size" (0.78 g) and difficult to examine.
An increase in fetuses with "unusual skeletal variations" in the form of retarded bone
ossification (not malformations) were also reported at the highest dose group. According
to the authors, most of these skeletal variations are frequently observed in 20 day old rat
fetuses of this strain and source. The authors concluded from these data that there was no
evidence of variation in sex ratio, embryotoxicity, inhibition of fetal growth and
development, or teratogenic effects resulting from exposure of the dams to unleaded
gasoline. Although the weight reduction was not judged by the authors to be related to the
treatment, this effect may be a sensitive indicator of developmental toxicity which requires
further investigation.
8.3 BENZENE
8.3.1 Reproductive and Developmental Effects in Humans
No data are available to assess the developmental or reproductive effects in humans
from exposure to benzene. However, there are studies available that evaluated exposure to
benzene in a mixture with other constituents of gasoline. These studies have been reviewed
in section 8.1.
8.3.2 Reproductive Effects in Animals
Male mice exposed via inhalation to 300 ppm benzene 6 hours/day, 5 days/week,
for 13 weeks exhibited testicular atrophy and degeneration, decreased numbers of
epididymal sperm, and abnormal sperm morphology. In addition, ovarian cysts were
observed in females following exposure to 300 ppm (Ward et al., 1985).
A single dominant lethal assay conducted with benzene consisted of exposing male
rats (20 males/group) by inhalation to 1, 10, 30, and 300 ppm benzene for 6 hours/day, 5
days/week for a 10 week treatment period (Bio/dynamics Inc., 1980). Following
treatment, males were placed with untreated females for two consecutive 7-day mating
intervals. Pre- and post-implantation loss was evaluated in this study. The mean number
of corpora lutea were comparable to control in all exposure groups except the 10 ppm
group, which had a non-significant decrease in mean number of implants at week 1. A
non-significant increase in mean number of dead implants were observed at week 1 for the
8-5
-------
300 ppm dose group. A similar increase was observed at week 2 at each dose level,
although it was not considered unusually high when compared to historical controls.
During week 2, the mean mutagenic ratios (number of dead implants/number of
implantation sites) for females exposed to treated males was higher than the control value,
although it was within the historical range for this rat strain. The authors concluded that
benzene was not clastogenic in this test system, however, the number of
females/dose/mating week (20) casts doubts on the statistical power provided by the sample
size used. In addition, data regarding the proportion of females with one or more dead
implants were not presented or analyzed statistically. Therefore, the study should be
considered inconclusive evidence of a non-clastogenic effect on male germinal cells.
8.3.3 Developmental Effects in Animals
Watanabe and Yoshida (1970) first reported teratogenic effects in mice
subcutaneously exposed to a single dose of 3 ml/kg benzene on days 11 to IS of gestation.
No control group was included in the study. The teratogenic effects consisted of cleft
palate, agnathia (no lower jaw), and microagnathia in mice exposed only on day 13 of
gestation. No maternal toxicity was reported. Other gavage studies conducted by Nawrot
and Staples (1979; 0.5 ml/kg) and Lyon (1975; 0.3 to 1.0 ml/kg) failed to show any
teratogenic effects in rats and mice.
Fetotoxic effects have been observed in the progeny of rats and mice exposed by
inhalation to benzene. The most prevalent effects were reduced fetal body weight, skeletal
anomalies, and increased resorptions (Green et al., 1978; Hudak and Ungvary, 1978;
Vozovaya, 1975b, 1976; Hazelton Laboratories, 1977; Litton Bionetics, 1978
(unpublished); Tatrai et al., 1979; Murray et al., 1979; and Kuna and Kapp, 1981). The
following is a summary of a review by Mehlman et al. (1980) of these data. The data are
summarized in Tables 8-1 and 8-2.
In a study by Green et al. (1978), reduced fetal body weights and crown-to-rump
lengths were observed in the offspring of Sprague-Dawley rats exposed to 2,200 ppm
benzene for 6 hours/day on days 6 to 15 of gestation. Missing sternebrae were also
observed in fetuses exposed to 100 ppm and 2200 ppm benzene, but not at the 300 ppm
dose level (Green et al., 1978). Maternal toxicity, as indicated by decreased body weight,
was observed at the 2200 ppm dose level. Continuous exposure of CFY rats to 313 ppm
during days 9 to 14 of gestation resulted in reductions in fetal body weights and skeletal
anomalies (Hudak and Ungvary, 1978). Reduced maternal body weight was also observed
at this dose level. Vozovaya (1975b, 1976) reported that progeny of rats exposed to 116
and 559 ppm benzene prior to and throughout gestation exhibited reduced body weights.
8-6
-------
TABLE 8-1
SUMMARY OF BENZENE INHALATION TERATOLOGY
oo
Contractor Species
Hazelton, 1977 . Rat
Green et al.. 1978 Rat
Murray et al.. 1979 Mouse
Rabbit
Litton Bionetics, Rat
1978
Strain
Sprague-
Dawley
Sprague-
Dawley
CF-1
New Zealand
COBSTM
Sprague-
Dawley
Inhalation
Exposure
Level
(PPM)
0
10
50
500
100
300
2.200
500
500
10
40
Duration
day 6-
day!6
of
gestation
day 6-
day!6
of
gestation
day 6-
daylS
of
gestation
day 6-
daylS
of
gestation
day 6-
day!6
of gestation
Decreased
Maternal
Body
Weight
_
-
yes3
yes3
_
_
yes3
..
_
_
-
Decreased Decreased
Fetal Crown-
Body Rump
Weight Distance
- -
yes3
yes3 yes3
_ _
_ _
yes3 yes3
yes3
_ _
_ _
-
Comments
or
Observations
—
—
Malformations1
Missing stemebrae
(most in females)
Missing stemebrae
(most in females)
Missing stemebrae
delayed skull ossification
unfused occipital
Extra ribs^. lumbar
spur(s); fused ribs, fused
thoracic vertebrae (1 pup
each/ 2 litters)
gastroschisisO pup)
Increased resorplion
Increased resorption
Exencephaly, angulated ribs, non-sequential ossification of forefeet
Occurred less often in litters of rabbits exposed to benzene.
Statistically significant (p<0.05).
SOURCE: Mchlman etal., 1980.
-------
TABLE 8-2
REVIEW OF BENZENE INHALATION TERATOLOGY STUDIES
Exposure
Study Species Level
Route and Decreased Decreased
Duration Maternal Fetal Comments
of Weight Weight or
Exposure Gain Gain Observations
Golfmekler, Rat
1968.
Pushkina Rat
et al., 1968.
Vozovaya, Rat
1976.
6.3-210 ppm Inhalation
24 hr/day
10-15 days
prior to
impregnation
l-670mg/m3 Inhalation
(208 ppm)
throughout
pregnancy
1783 mg/m3
(116 ppm)
Hudak & Rat 1000 mg/m3
Ungvary, Mouse
1978.
Inhalation
4 mo prior
plus pregnancy
Inhalation
(310 ppm)
24 hr/day 1-14
days of pregnancy
Decreased
litter size
Decreased
litter size
Yes - No malformations
yes
No malformations
SOURCE: Mehlman et al., 1980.
8-8
-------
No maternal toxicity was reported at these dose levels. Hazelton Laboratories (1977)
observed skeletal anomalies (exencephaly, angulated ribs, and ossification of forefeet out
of sequence) and reduced fetal weight in offspring of Sprague-Dawley rats exposed to 50
and 500 ppm benzene from days 6 through 16 of gestation. Maternal body weight was
decreased at both of these doses. However, a low fertility index in the control group and
the maternal nutritional status in the exposed group were cited as problems with this study
(Mehlman et al., 1980). CF-1 mice exposed to 500 ppm also had decreased fetal weight,
missing stemebrae, delayed skull ossification, and unfused occipital bones (Murray et al.,
1979). Maternal toxicity was not observed in CF-1 mice exposed to 500 ppm of benzene.
Murray et al. (1979) also exposed pregnant New Zealand rabbits to 500 ppm benzene on
days 6 through 18 of gestation and reported extra ribs, lumbar spur(s), fused ribs, fused
thoracic vertebrae (1 pup each/2 litters), and gastroschisis (1 pup/19 liners).
An increase in the incidence of resorptions was reported in the following studies:
in rats exposed to 2,200 ppm for 6 hours (Green et al., 1978); in rats exposed to 500 ppm
for 7 hours (Hazelton Laboratories, 1977); in rats exposed to 10 and 40 ppm for 6 hours
(Litton Bionetics 1978, unpublished); and in CFY rats continuously exposed to 150 ppm
benzene from days 7 through 14 of gestation (Tatrai et al., 1979). No maternal toxicity
was observed at these doses.
In the study conducted by Kuna and Kapp (1981), pregnant Sprague-Dawley rats
were exposed to 10,50, and 500 ppm benzene 7 hours/day from days 6 through 15 of
gestation. Maternal weights in the mid- and high-dose groups were significantly decreased
compared to the air exposed control group. Mean fetal weights were significantly
decreased in the mid- and high-dose groups compared to control fetuses. Skeletal
abnormalities observed in the mid-dosed group were delayed ossification of the rib cage
and extremities. In the high-dose group, these abnormalities consisted of exencephaly (one
fetus), angulated ribs (another fetus), ossification of forefeet out of sequence (two fetuses),
and lagging ossification of the skull, vertebral column, rib cage, and pelvic girdle. These
skeletal abnormalities observed in the high-dose group have been judged to be teratogenic
in rats (Kuna and Kapp, 1981). Mean number of caudals were significantly lower in the
high dose group. A dose-related decrease in the mean number of metacarpels and
phalanges per foot was determined by regression analysis. Slight dilation of the ventricles
in the brain were observed in five fetuses in the mid-dose group and four fetuses in the
high dose group. In the latter group, three fetuses exhibited both dilated lateral and third
ventricles.
According to the authors, the historical incidences of these effects in laboratory
control animals are: exencephaly (2/4938), angulated ribs (19/2737), dilated lateral and
8-9
-------
third ventricles of the brain (2/2737), and forefeet ossification out of sequence (0/2737).
These data indicate that the skeletal abnormalities observed in the treated group would not
be considered spontaneous in this strain.
Alterations in the hematopoietic system of mice were investigated by Keller et al.
(1986). Swiss-Webster mice were exposed in utero to 5,10, or 20 ppm benzene (5 to 10
animals/group) from days 6 through IS of gestation. Hematotoxic and other responses
were determined in fetuses on day 16 of gestation (two males and two females per litter), 2-
day-old neonates (two male and two female offspring per litter) and 6-week-old progeny
(one male and one female offspring per liner). The longer term effects of in utero exposure
to 10 ppm benzene was determined in 10-week-old progeny re-exposed to 10 ppm benzene
for 2 weeks. Hematotoxic effects were assessed by determining changes in the BFU-E and
CFU-E erythroid precursor cells, and granulocytic/macrophage colony-forming cells (GM-
CFU-C).
Litter sizes and weights, and the number of dead, resorbed, or malformed fetuses
were within control limits for all exposure concentrations. Erythroid stem cells in the
fetuses were affected at all exposure levels. The response in the more differentiated CFU-E
cells was more pronounced than the response in the BFU-E cells. This is consistent with
the fact that CFU-E changes represent more sensitive indicators of hematological alterations
than changes in BFU-E cells.
In the fetuses, significant increases in the number of BFU-E and CFU-E levels
were observed at 5 and 10 ppm exposures. By contrast, CFU-E levels were significantly
decreased at 20 ppm. This bimodal response in cell numbers was attributed to continued
benzene-induced suppression in the colony-forming ability of the 20 ppm exposure group.
The effects at 5 ppm did not persist past parturition. Male 2-day-old neonates of the mice
exposed to 20 ppm had a significant increase in CFU-E (a signal of recovery from the in
utero exposure). GM-CFU-C levels were significantly increased in the 2 day old male and
female neonates of the 20 ppm exposed mice. In the progeny of the 10 ppm exposed mice,
2 day old neonates exhibited variable responses in the CFU-E. Male adult progeny of the
10 ppm exposed mice exhibited significantly decreased CFU-E.
When 10 week-old male progeny exposed in utero to 10 ppm of benzene were re-
exposed to 10 ppm for 2 weeks, significant reductions in bone marrow CFU-E were
observed relative to mice exposed in utero to only air. No significant effect was observed
in the female mice similarly exposed. Decreases in splenic GM-CFU-C (but not bone
marrow GM-CFU-C) in both sexes were observed in both in utero air- and benzene-
exposed progeny re-exposed to 10 ppm of benzene at 10 weeks of age. The response was
more pronounced, however, in the in utero benzene-exposed mice. These findings indicate
8-10
-------
that possible long-term damage to stem cells may occur in mice exposed in utero to
concentrations as low as 10 ppm. As benzene re-exposure at lower levels was not
conducted, it is possible that even lower benzene concentrations may produce longer term
effects in this assay.
Persistent changes in rat brain catecholamine function following neonatal exposure
to benzene were reported by Tilson et al. (1980). Tilson et al. dosed male and female
Fischer 344 rats with 550 mg/kg of benzene or a corn oil vehicle on days 9,11, and 13
post-partum. Spontaneous motor activity was found to be elevated in the exposed rats
relative to the controls when the animals were tested at 100 to 130 days of age. The
exposed animals were also less sensitive to the increasing motor activity effects of d-
amphetamine when tested as adults. Body weights and performance in a battery of tests to
assess neurotoxicity (on days 45,60, and 100 postpartum) were not affected by neonatal
benzene exposure. These authors concluded that the data indicate that postnatal exposure to
benzene can produce significant alterations in the motor activity of rats when tested during
adulthood and that the type of effect depended on the procedure used. They also concluded
that changes in the sensitivity of benzene exposed animals to d-amphetamine suggest long-
term alterations in catecholaminergic function.
8.4 TOLUENE
8.4.1 Reproductive Effects in Humans
Syrovadko (1977, as cited in U.S. EPA, 1986b) reported a high percentage of
menstrual disorders in women employed in the manufacture of electric insulation materials
and exposed to average concentrations of 6 to 93 ppm toluene and other solvents. In
addition, the offspring of this group were more often underweight, and experienced more
frequent fetal asphyxia and belated onset of nursing.
Dysmenorrhea was reported in 19 out of 38 Japanese female shoemakers exposed
for over 3 years to mean toluene concentration of 60 to 100 ppm and in "a few working
places" to 20 to 50 ppm gasoline (Matsushita et al., 1975, as cited in U.S. EPA, 1983b).
8.4.2 Developmental Effects in Humans
No data are available to assess the developmental effects in humans from exposure
to toluene. However, there are studies available that evaluated exposure to toluene in a
mixture with other constituents of gasoline. These studies have been reviewed in section
8.1.
8-11
-------
8.4.3 Reproductive Effects in Animals
In a dominant lethal assay, sponsored by API, by Litton Bionetics (1981), CD-I
male mice were administered 100 ppm and 400 ppm toluene, 6 hours/day, 5 days/week for
8 weeks. Toluene did not cause significant increases in either pre- or post-implantation
loss of embryos compared to the control group. Exposure of two consecutive generations
of mice to 2000 ppm toluene did not effect survival or reproductive parameters in the
progeny (API, 1985). Smith (1983) also found no evidence of toluene induced
embryotoxicity in a study of 270 mice whose mothers were exposed to 2,370 mg/kg/day of
toluene.
8.4.4 Developmental Effects in Animals
Narwot and Staples (1979) reported a significant increase in embryo lethality in
female mice orally exposed to 261,440, and 870 mg/kg/day toluene on days 6 through 15
days of gestation and 870 mg/kg from days 12 to 15 of gestation. Increased incidence of
cleft palate, reportedly not due to growth retardation, was observed in the highest dose
group and a significant reduction in fetal body weight was reported for mice exposed to
440 and 870 mg/kg toluene exposed on days 6 through 15 of gestation. No maternal
toxicity was noted in animals receiving toluene on days 6 through 15 of gestation.
However, decreased maternal weight was observed in the dams exposed on days 12 to 15.
Hudak and Ungvary (1978) exposed CFLP mice and CFY rats to toluene at various
dose levels and stages of development The dosing schedule is presented below.
Parameters examined in this study were maternal weight gain, fetal loss, mean fetal weight,
placental weights, and percentage of weight-retarded fetuses. In this study, pregnant mice
exposed to 399 ppm (1,500 mg/m^) toluene died within 24 hours. Exposure to pregnant
mice under the same conditions to 133 ppm (500 mg/m3) toluene from days 6 to 13 of
gestation did not produce maternal lethality, but significantly reduced fetal weights. No
external or visceral malformations in the fetuses were found at this dose level.
8-12
-------
Species
Rats
Mice
Exposure Duration
9-14 days of gestation
24 hr/day
1-21 days of gestation
8 hr/day
1-8 days of gestation
24 hr/day
6-13 days of gestation
24 hr/day
Exposure Dose
mg/m3
1,500 mg/m3
l,000mg/m3
l,500mg/m3
l,500mg/m3
500 mg/m3
ppm
399
266
399
399
133
SOURCE: Hudak and Ungvary, 1978.
Rats continuously exposed to 399 ppm (1,500 mg/m3) on days 9 through 14 of
gestation exhibited an increased incidence of skeletal anomalies (i.e., fused stemebrae and
extra ribs). No maternal or fetal toxicity was observed at this dose level. There was an
increased frequency of signs of retarded skeletal development (i.e., unossified stemebrae,
bipartite vertebrae centra, and shortened thirteenth ribs) in the second group of rats exposed
to 266 ppm (1,000 mg/m^) for 8 hours per day until the twenty-first day of gestation. No
maternal deaths or fetal weight loss was reported in this group. A high incidence of
maternal lethality (5/9) was observed in the third group of dams exposed to 399 ppm
(1,500 mg/m3) on days 1 through 8 of gestation. In the surviving pups, retarded skeletal
development and a decrease in fetal weights were observed in this group.
Von Euler et al. (1989) examined the effects of neonatal toluene exposure on the
regional brain catecholamine levels and utilization, as well as on serum levels of
hypophysial and adrenocortical hormones in the adult (8-week-old) male rat. The rats (five
per group) were exposed to air or 80 ppm of toluene (the threshold limit value in many
countries) 6 hours a day, during days 1 through 7 of gestation, and to either air or toluene
for 3 days (6 hours a day) during the eighth week of gestation. Neonatal toluene exposure
produced a selective reduction in dopamine levels and utilization (80 percent of controls) in
the olfactory tubercle and substantia nigra of the adult rat. Noradrenaline levels and
utilization were reduced selectively in the substantia nigra and increased selectively in the
median eminance and the paraventricular hypothalamic nucleus. Although circulating
serum hormone levels were not affected by neonatal toluene exposure, the reductions in
dopamine levels were reflected in the inhibition of prolactin induced by this
neurotransmitter.
8-13
-------
Neonatal toluene exposure also rendered the rats insensitive to subacute toluene
exposure during adulthood. In some cases, the exposure even counteracted the effects of
toluene exposure observed in animals exposed neonatally to only air.
A drinking water study on mice exposed perinatally to toluene provides further
evidence that neurological effects may occur as a result of toluene exposure (Kostas and
Hotchin, 1981). Pregnant female mice were exposed to 16,80, or 400 mg/L of toluene in
their drinking water. While toluene treatment did not affect maternal fluid consumption or
offspring mortality, the 72 mg/kg/day treatment significantly reduced habituations to open
field behavior at 35 days of age. Rotorod performance was significantly reduced in all
exposure groups at 45 to 55 days. Acute exposure to toluene (14.4 or 72 mg/kg/day) via
intraperitoneal injection at 35 days produced no significant effect on open field activity.
Rotorod performance was not examined in the acutely exposed mice.
Litton Bionetics (1978b) conducted a study under the sponsorship of API to
determine the teratogenic potential of toluene. Groups of 25 pregnant rats were exposed to
airborne concentrations of atomized toluene at 0,100, or 400 ppm for 6 hours/day on days
6 to 15 of gestation. Control dams were placed in similar cages, but were exposed to room
air only. Dams were weighed and food consumption was measured throughout the
gestation period. Dams were observed daily for changes in general appearance, behavior,
and condition. On day 20 of gestation, all dams were sacrificed, and visceral and thoracic
organs were grossly examined. The implantation sites, live and dead fetuses, and
resorption sites were examined. One-third of the fetuses were examined for changes in the
soft tissue of organs and the remaining for skeletal anomalies. Administration of toluene
by inhalation to female rats at doses of 100 and 400 ppm produced no effect in the pregnant
dams. There was no evidence of variation in fetal sex ratio, embryo toxicity, inhibition of
fetal growth, and development or teratogenic effect induced by toluene at these airborne
concentrations.
In a study by Courtney et al. (1986), CD-I mice were exposed to 200 or 400 ppm
of toluene, 7 hours/day from days 7 through 16 day of gestation. In addition to measuring
parameters indicative of maternal and fetal toxicity, Courtney et al. measured lactic
dehydrogenase (LDH) activity in mice during both prenatal and postnatal development
Organ damage can result in an increase of LDH and a change in the isoenzyme profile.
Prenatal development was assessed in this study by assaying the LDH activity and LDH-
isoenzyme profile in major organs of pregnant dams, non-pregnant mice and their
respective controls exposed to air. For the postnatal study, LDH activity in 21-day old
neonates from pregnant mice in the high dose group (400 ppm) and air exposed control
mice was determined.
8-14
-------
In the prenatal study, the maternal weights of dams was not effected by exposure to
toluene, but liver-to-body weight ratios in dams at both concentrations were significantly
reduced. No difference in the the average number of implantation sites, live fetuses, fetal
mortality, or fetal body weight in toluene-treated mice were observed compared to pregnant
air-control dams. No toxicity was observed in the non-pregnant treated mice.
A statistically significant increase in the incidence of enlarged renal pelves compared
to the control group and high dose group were observed in the low dose group. Because
both the kidney and rib profile were affected in the lower dose group, the authors suggest
that this may be due to a desynchronization of growth and maturation between the low and
high dose groups. An increase in the number of fetuses with 14 ribs was also observed in
the low dose group. In the high dose group, a statistically significant increase in the
number of fetuses with 13 ribs was observed compared to the control group.
The heart, lungs, brain, liver, and kidneys of treated dams and pregnant and non-
pregnant air-control mice were assayed for total LDH activity and LDH-isoenzyme profiles.
A statistically significant increase in the LDH activity in the brain of treated dams in the
high dose group was observed. In non-pregnant treated mice, a statistically significant
increase in LDH activity in the liver and kidney was observed in the high dose group. The
authors suggested that the increase of LDH activity in the brains of pregnant treated mice,
and kidney and liver of non-pregnant treated mice may indicate that toluene is distributed
differently in pregnant mice than in non-pregnant mice. There were no significant
differences in the LDH-isoenzyme profile dams and non-pregnant mice for the heart, brain,
lungs, or kidneys. LDH-3 was significantly depressed in non-pregnant mice exposed to
the low dose level compared to the control and high dose group.
A non-significant increase (almost two times) in the mortality of neonates at birth in
the toluene-exposed group compared to the air-control groups was observed. There was
no significant difference in the LDH activities between the treated and control neonates 21
days after birth, except for a slightly greater brain LDH activity level.
8.5 XYLENE
8.5.1 Reproductive and Developmental Effects in Humans
No data are available to assess the developmental or reproductive effects in humans
from exposure to xylene. However, there are studies available that evaluated exposure to
xylene in a mixture with other constituents of gasoline. These studies have been reviewed
in section 8.1.
8-15
-------
8.5.2 Reproductive Effects in Animals
No reproductive effects were observed in male and female CD rats exposed by
inhalation to 500 ppm xylene, 6 hours/day for 130 days before mating (Bio/dynamics,
1982).
8.5.3 Developmental Effects in Animals
Mice exposed by inhalation to xylene (no information on composition reported) at
levels of 500, 1,000, 2,000, or 4,000 ppm for 6 hours/day from days 6 through 12 of
gestation exhibited reduced fetal body weights, a dose-related trend toward delayed
ossification, and development of extra ribs (14th) at concentrations of 2,000 and 4,000
ppm. In one-third of the mice from this study allowed to deliver, neonates were examined
for abnormalities. Offspring from mice exposed to 4,000 ppm exhibited decreased weight
gain and delayed body hair and tooth development (Shigeta et al., 1983, as cited in U.S.
EPA, 1986f).
Slight increases in post-implantation loss (numbers of resorbed and dead fetuses)
were noted in pregnant CFY rats continuously (24 hours/day) exposed to a mixture of
xylenes (50,20,10, and 20 percent o-, m-, and p-xylene and ethylbenzene, respectively) at
a level of 230 ppm (1,000 mg/m3) during days 9 through 14 of gestation (Hudak and
Ungvary, 1978). There were statistically significant increases (p < 0.005) in incidences of
fused sternebrae, extra ribs, and an increased trend of retarded skeletal development in
fetuses of treated rats compared to control animals.
In an inhalation studies by Ungvary et al. (1980), pregnant rats were continuously
(24 hours/day) exposed to o-, m-, or p-xylene at levels of 35, 350, or 700 ppm (150,
1,500, or 3,000 mg/m3, respectively) during days 7 through 14 of gestation. An
additional group of rats was exposed to these three xylene concentrations for 2 hours only
on the 18th day of gestation. Results of this study are presented in Table 8-3.
Ungvary et al. determined xylene crosses the placenta and is present in fetal blood
and amniotic fluid. Maternal toxic effects occurred in groups exposed to the mid and high
concentrations of xylene and were different with each isomer. Exposure to the mid- and
high-dose levels of o-xylene (350 and 700 ppm) was characterized by an increase in liver
weights whereas exposure to the highest dose level of m-xylene (700 ppm) caused an
increase in maternal mortality (4/30). No maternal effects were observed from p-xylene
exposure.
Exposure to 35 ppm o-xylene and all three concentrations of p-xylene (35,350, or
700 ppm) caused a decrease in placenta weight. An increase in placenta weight was
8-16
-------
oo
TABLE 8-3
DATA OF THE FETUSES OF ORTHO-, META-, AND PAJM-XYLENE TREATED CFY RATS
Inhalation Air
on days 7- 14
of pregnancy
24h/day
Number of litters 13
Number of live fetuses 168
External malformations-
agnathia
Number of fetuses 84
dissected
Internal malformations
hydrocephalus
iniemus
pulmo polycysticus
dystopia renis
pyelactasia 1 1
ectasia ves. urin
Number of Alizarine-
red S-siained fetuses 84
Skeletal retardation signs 13
Skeletal anomalies
fused sternum -
extra ribs
Skeletal malformation-
agnalhia
Ortho-;
fmp/m
ISO
17
217
1
107
—
1
-
7
1
110
27
3
2
1
nylene
L3)
1500
18
242
_
118
1
—
1
10
5
123
30
4
3
3000
17
234
_
113
_
-
-
17
9
121
48a
—
3
Air
23
284
_
139
_
5
-
5
3
142
6
1
2
Meta-x
(mp/nr
ISO
IS
198
_
85
_
3
-
2
5
102
4
_
1
ylene
L3)
1500
19
247
_
118
_
4
—
2
-
123
6
_
3
3000
18
196
_
94
_
—
—
8
5
97
15
_
8a
Air
18
226
_
110
1
1
-
3
-
116
16
1
6
Azra-xylene
(tnti/m^)
ISO 1500
1216 6
157 197
_ _
80 96
_ _
— —
— —
3 11
- -
77 100
24a 38a
1
1 9
3000
51
multiplex:!
25
_
—
—
3
1
26
15a
2
10b
-- ____ __ --multiplex:!
a P<0.05; b P<0.01; (Mann-Whitney U-test)
-------
00
oo
TABLE 8-3
(continued)
Inhalation
on days 7- 14
of pregnancy
24h/day
Number of females
treated
died
non-pregnant
total resorption
Number of liners
Maternal wl gain in %
of starting body wt
Liver wt/body wt.
ratio (%)
Liver wt/reduced body
wt ratio (%)
Number of fetuses
live
dead
resorted
Air
15
-
2
—
13
57.5
3.88
4.54
168
3
14
Of/io-xylene
(mflmi)
150
20
—
3
-
17
60.0
4.008
4.68a
217
-
17
1500
20
—
2
—
18
57.2
4.24a
4.99a
242
1
14
3000
20
—
1
2
17
54.3
4.18a
4.88a
234
-
35
Air
25
—
2
-
23
65.8
4.27
5.02
284
-
27
A/ew-xylene
(mg/m^)
150
20
_
5
-
15
67.3
4.35
5.06
198
—
17
1500
20
_
1
—
19
64.7
4.45
5.21
247
3
14
3000
30
4
8
-
18
48.3a
4.67
5.36
196
1
11
Air
20
_
2
-
18
52.9
4.12
4.86
226
_
9
/Vira-xylene
(mg/ni^
150
20
_
8
-
12
58.2
3.98
4.72
157
_
8
1500
20
_
4
-
16
54.5
4.08
4.79
197
1
13
3000
20
_
7
7
6
41.3
4.56
5.05
51
_
US
Mean implantation/ dam 14.23 13.76 14.28 14.16
13.52 14.27 13.89 11.441
13.06 14.08 13.44 12.77
Fetal loss in % of total
implantation sites
9.2
7.3
5.8 13.0
6.7 7.9
6.4
5.3
4.0
9.0
8.0 69.06
-------
TABLE 8-3
(continued)
Inhalation
on days 7-14
of pregnancy
24h/day
Air Ortto-xylene
ISO 1500 3000
Air Afew-xylene
fmp/m3)
150 1500 3000
Air Pora-xylene
150 1500 3000
00
I—•
NO
Mean liuer size
Mean fetal wt (g)
Mean placenta! wt (g)
Wt retarded fetuses in %
of living fetuses
12.9 12.8 13.4 13.8
4.07 4.02 3.72C 3.75C
0.63 0.61f 0.65f 0.38h
2.0 0.0 12.011 17.0
12.4 13.2 13.0 10.9
3.97 3.85 3.94 3.60b
0.65 0.65 0.65 0.69
3.2 8.8 3.6 27.5d
12.6 13.1 12.3 8.5C
4.05 4.02 3.97 3.56C
0.73 0.648 0.66h 0.61"
4.0 9.6
2.0 20.0d
a P<0.05; b P<0.01; c P<0.001; (variance analysis): + S.E.M.d P<0.05; c P<0.01; (Mann-Whitney U-test); f P<0.05; 8 P<0.01;n P<0.001; (x2-lest);
+ S.E.M.' P<0.05: (t-test); + S.E.M. 3 Reduced body wt + maternal wt on day 21 of pregnancy - (total wt of fetuses = placentas).
SOURCE: Ungvary etal., 1980.
-------
observed in groups exposed to the mid- and high doses of oxylene. No effects on
placenta weight were observed with m-xylene exposure.
There was a 31 percent decrease in the number of rats found pregnant at autopsy
(preimplantarion loss) in the group exposed to the highest concentration of m-xylene and a
20 to 40 percent decrease in the group exposed to the highest concentration of p-xylene.
Reduced fetal weight was observed in rats exposed to o-xylene and m-xylene at the mid
and high doses. This reduction in fetal weight corresponded to the dams that had decreased
food consumption.
The incidence of skeletal retardation was increased in the groups exposed to the
highest concentration of o-xylene and all three concentrations of p-xylene that were not
toxic to the dams. However, no malformations were induced by these isomers. The
incidence of extra ribs was significantly higher in the m-xylene and p-xylene groups
exposed to the highest concentration. Because there was no increase in the incidence of
extra ribs in the o-xylene exposure group, the authors noted that the increase in extra ribs
observed in other studies from exposure to lower doses of the xylene mixture is probably
due to the combined effect of m-xylene and p-xylene.
Several enzymes associated with kidney function were decreased in fetuses exposed
to o-xylene and p-xylene at 700 ppm. Succinic dehydrogenase and glucose-6-phosphatase
activities were reduced in the liver and thymus cells of rats exposed to all three isomers at
the highest concentration.
In another inhalation study (Balogh et al., 1982, as cited in U.S. EPA, 1986f),
pregnant rats were exposed to 230,1,900, or 3,360 mg/m^ (52, 436, or 772 ppm) for 24
hours per day from days 7 through 14 of gestation. Delayed bone development and
increased incidences of extra ribs and resorptions (post-implantation loss) were reported at
exposure levels of xylene at which no evidence of maternal toxicity was noted.
In contrast to these findings, a study by Litton Bionetics (1978c) exposed pregnant
rats to 0, 100, and 400 ppm xylene, 6 hours/day on days 6 through 15 of gestation showed
no evidence of variation in fetal sex ratio, embryotoxicity, or inhibition of fetal growth and
development. The fetuses of three litters in the high dose group were small and markedly
retarded in development, but these changes were not significantly different from controls.
Pregnant Wistar rats were exposed to 2,12, and 115 ppm (10, 50, and 500 mg/m3)
of xylene for 6 hours/day, 5 days/week until the 21st day of pregnancy (Mirkova et al.,
1983). Incidence data were reported as percentages in this study. Post-implantation loss at
the late stages of development was increased by 168 percent at the highest dose group and
94 percent in the 50 mg/m^ group. Mean fetal weight was also decreased. An increased
incidence of fetal hemorrhages were reported in the mid- and high-dose groups. The
8-20
-------
hemorrhages appeared to occur primarily in the cervical area. A significant increase in the
incidence of anomalies in the internal organs of fetuses exposed to the highest concentration
was reported. The incidence of abnormal ossification of fetal skeletons were reported for
the two highest doses including the absence and impaired formation of bone and skull.
Visceral anomalies were also observed but not specified. At 10 mg/m^, no significant
increase in skeletal effects was observed. Exposure to 50 and 500 mg/m^ caused a
significant decrease in fetal weight of the newborn on days 7 and 21 after birth. These data
are difficult to assess because of possible methodological problems associated with the
animal care that were noted with this study (U.S. EPA, 1986f)-
There is also some evidence that oral exposure to xylene results in teratogenicity in
mice at maternally toxic doses. In a study reported by Marks et al. (1982), offspring of
CD-I mice receiving oral doses of 0.6,1.2,2.4,3.0, 3.6, and 4.8 ml/kg/day mixed
xylenes (60.2, 13.6,9.1, and 17.0 percent of m-, p-, and o-xylene and ethylbenzene,
respectively) during days 6 through 15 of gestation exhibited increased incidences of cleft
palates in 2.4, 3.0, and 3.6 ml/kg/day dose groups. A statistically significant (p <0.001)
dose-related increase in the average percent of malformed fetuses based on the incidence of
multiple wavy ribs in fetuses in groups exposed to 2.4 ml/kg-day and greater.
Developmental toxicity was evidenced by a significant, dose-related increase (p <0.05) in
resorptions at the highest dose and significant reductions (p <0.05) in fetal body weights at
dose levels of 2.4 ml/kg/day and greater. Maternal toxicity was observed at levels of 2.06
g/kg/day or greater.
Similarly, increased incidences of resorptions and cleft palates were observed in
CD-I mice given oral doses of 1.96 or 2.61 g/kg/day of o- or p-xylene during days 6
through 15 of gestation (Nawrot and Staples, 1981, as cited in U.S. EPA, 1986f).
Maternal toxicity was noted at these doses in mice exposed to o- or p-xylene. All xylene
isomers administered at the highest dose during days 12 to 15 of gestation caused an
increase in prenatal deaths. It should be noted that spontaneous cleft palates are observed
in some strains of mice, including CD-I mice.
There is some evidence that dermal exposure to xylenes may result in sb'ght
developmental toxicity. Decreased fetal size and body weights and increased prenatal death
reportedly occurred in hamsters after dermal application of xylene between days 7 and 11
of gestation (Overman, 1981, as cited in U.S. EPA, 1986e). "Extensive effects" on
maternal skin were also reported. Further details of this study were not specified.
8-21
-------
8.6 SUMMARY
Occupational exposure studies have reported adverse reproductive effects in female
workers exposed to gasoline vapors or constituents of gasoline, including benzene,
toluene, and xylenes. The results of an epidemiological study suggest that women
occupationally exposed to gasoline may lead to congenital CNS effects in their children.
There is limited animal data on the effects of gasoline exposure; only one study has
investigated the potential for developmental effects in rodents. The study reported reduced
size of fetuses in the high dose group. Benzene and xylene (o-, m-, or p-xylene or a
mixture of the three isomers) have been shown to be teratogenic in rats at maternally toxic
levels after inhalation or oral exposure, and toluene was teratogenic in mice below
maternally toxic levels after oral exposure. Evidence has also been presented that benzene,
toluene, and xylene cause developmental toxicity in laboratory animals. Increased post-
implantation loss (increased resorptions), reduced fetal body weights, and delayed
ossification or retardation of skeletal development were observed in rats and mice after oral,
dermal, or inhalation exposure to these gasoline constituents.
8-22
-------
9. GENETIC TOXICITY
The genotoxicity of unleaded gasoline (UG) has been evaluated in a number of in
vivo and in vitro biological test systems using bacteria, yeast, rodents, insects, and human
tissue. The major genotoxic endpoints evaluated in these studies are gene mutation,
chromosomal aberrations, and DNA damage in bacteria and mammalian cells.
This chapter summarizes the results of the major genotoxicity studies of unleaded
gasoline and benzene, toluene, and xylenes.
9.1 GASOLINE
Table 9-1 summarizes the genotoxicity studies of unleaded gasoline. Selected
studies from this table are discussed in detail below.
9.1.1 Gene Mutation Studies
9.1.1.1 Bacterial Cells
Unleaded gasoline was evaluated for genetic activity in a series of in vitro microbial
assays in the presence and absence of metabolic activation. Generally, UG is not
mutagenic under the conditions of these assays. Technical problems associated with
genotoxicity testing of chemical mixtures have been reported for gasoline. These
problems, including sub-threshold metabolic activation or limited bioavailability of the
genotoxic components, are believed to have suppressed the mutagenic response of
genotoxic components in the chemical mixture.
In an attempt to overcome some of these technical difficulties, Dooley et al. (1987)
tested UG (PS-6), a DMSO extraction of UG, and an evaporative residue of UG for their
ability to induce mutation in a modified Ames bioassay performed with and without
metabolic activation. According to chemical analysis of the test material, the UG extraction
contained a high concentration of alkylated benzenes (73 percent) and relatively low
concentrations of non-aromatics, alkylated benzenes, and indans. In contrast, the
evaporative residue contained a high concentration of alkylated naphthalenes as well as low
concentrations non-aromatics, alkylated benzenes, and indans.
No increase in revertant colonies were observed for UG or the UG extraction. A
reduction in the revertant colonies at the highest doses indicated that the UG extraction was
toxic, but not mutagenic. The evaporative residue induced a less than two-fold increase in
the mutant colonies, and therefore, was not considered mutagenic in this assay. However,
9-1
-------
TABLE 9-1
GENOTOXICITY STUDIES OF UNLEADED GASOLINE
TEST TYPE
bacterial
mutation assay
bacterial
mutation assay
mutation assay
yeast culture
mutation assay
TEST SPECIES
S. typhimurium
TA98
with and without
metabolic activation
(modified assay using
hamster S9 cofaclor)
S. typhimurium
TA98.TA100, TA1535
TA1537, TA1538
with and without
metabolic activation
S. typhimurium
TA98. TA 100, TA 1535
TA 1 537, TA 1538
with and without
metabolic activation
S. cerevisiae
D4
with and without
metabolic activation
CONCENTRATION
UG
UG extraction
suspension lest
5-200 ul/plale
Evaporative residue
50 to 10.000 ue/olate
suspension test
0.375-3.0% in DMSO
plate test
0.001-5.0 ug/plate
0.001-5.0 ug/plate
for plate assay
0.625-5.0% in DMSO
for suspension assay
RESULTS
No increase in reverlant colonies were observed
The number of reverlant colonies was not significantly
different from controls except at the highest concentration
where a slight reduction was observed. The UG
extraction was considered toxic, but not mutagenic.
Inconclusive, results were reproducible but less than
two-fold increases in mutant colonies were observed.
Slight increases in mutation frequency at the highest dose
without metabolic activation in strains TA100, TA1537
and TA1538. Significant increases were observed in
TA98 but were not reproducible
The results of the assays with metabolic activation were
considered negative because of anomalous reductions in
viable cell counts.
No incease in reverlant colonies were observed.
No increase in revertam colonies were observed.
REFERENCE
Dooley et al., 1988
Litton Bioneiics, 1977
Litton Bioneiics, 1977
VJD
-------
TABLE 9-1 CONTINUED
TEST TYPE
mammalian cell
mutation
mammalian cell
mutation
mammalian cell
mutation
mammalian cell
mutation
mammalian cell
mutation
TEST SPECIES
mouse lymphoma
(IS 1 78 Y cell line)
thymidine kinase (TK)
locus
with and without
metabolic activation
mouse lymphoma
(15178Y cell line)
thymidine kinase (TK)
locus
with and without
metabolic activation
mouse lymphoma
(15 1 78 Y cell line)
thymidine kinase (TK)
locus
with and without
metabolic activation
mouse lymphoma
(IS 1 78 Y cell line)
ihymidine kinase (TK)
locus
with and without
metabolic activation
CONCENTRATION
Eight concentrations
ranging from 0-300
ug/ml
UG-0.06S-1.04 ul/ml
UG-0.045-0.07 ul/ml
Extraction of UG
Evaporative residue
RESULTS
Statistically significant increase in the mutation frequency
at the TK locus with or without metabolic activation
An increase in the number of mutants was observed
at two concentrations with activation.
A two-fold or greater increase in mutant frequency
with and without activation was observed at cone
yeilding total growth of less than 10%.
No significant increase in mutant frequency at
concentrations with total growth of >10% with and
without activation.
Dose-related increases in mutation frequency without
metabolic activation
Dose-related increases in mutation frequency with
metabolic activation.
REFERENCE
Phillips Petroleum,
1984
Litton Bionetics, 1977
Dooley el al., 1987
Dooley et al., 1987
NO
-------
TABLE 9-1 CONTINUED
TEST TYPE
mammalian cell
mutation
Sister chromatic!
exchange
Sister chromatic!
exchange
Unscheduled DNA
synthesis (UDS)
Replicative DNA
synthesis (RDS)
TEST SPECIES
human lymphoblasts
with and without
metabolic activation
Chinese hamster ovary
(CHO) cells
with and without
activation
human lymphoblasts
with and without
metabolic activation
in vivo/in vitro kidney
cell of Fischer rats
in vivo/in vitro kidney
cell UDS Fischer rats
CONCENTRATION
UG and TMP in final
concentrations of
5% to medium and
volatile components o
10mlUGfor3hours
Five doses with the
highest dose of 1600
ug/ml
UG, TMP and
volatile components
of UG (see above)
UG by gavage with
2000-5000 mg/kg for
1 to 4 days
UG-inhalationof2000
ppm for 18 days
see exposure as above
RESULTS
No significant increases in the mutation frequency at the
TK locus with or without activation.
Cytoloxic response at high dose levels caused
lymphoblasts to dissolve
No statistically significant increase in SCEs except at the
highest dose with metabolic activation.
No increase in SCE frequency was observed.
No UDS activity in the kidney was observed in this assay
No increase in UDS in male or female rats, although the
amount of the repair activity occuring in the kidney may
may be below the detection limit of the assay.
Significant increases in RDS activity were observed in
male rat kidneys but not female rat kidneys.
REFERENCE
Richardson etal., 1986
Phillips Petroleum,
1984
Richardson el al., 1986
Loury et al., 1987
-------
TAIILE 9-1 CONTINUED
TEST TYPE
Unscheduled DM*
synthesis (UDS)
Replicative DNA
synthesis (RDS)
Mutation of rat
bone marrow
cells
TEST SPECIES
in vivo/ in vitro
rat hepaiocyies
in vivo/ in vitro
mouse hepaiocyies
in vitro rat hepaiocyies
in vitro mouse
hepatocytes (male only)
in vitro human
hepaiocyies
in vivo/ in vitro
rat hepatocytes
in vivo/ in vitro
mouse hepaiocyies
Acuie sludy:
rats received UG
intraperitoneally (ip)
Subchronic sludy:
UG was adminisled ip
daily for 5 days
CONCENTRATION
0.1-0.5g/kgUGor
0.5 g/kg TMP
0.02 g/kg
0.01-0.10%
0.01-0.05%
0.024, 0.08 or 0.24
ml/rat
0.01, 0.03 or 0.1
ml/rat daily for 5 days
RESULTS
No increase in DNA activity was observed.
A statistically significant increase in UDS activity was
observed in male and female mice 12 hrs after treatment
with 2000 mg/kg
UG stimulated a dose-dependent increase in UDS. No UDS
was observed in hepatocytes exposed in vitro to TMP.
For both mouse and human hepatocytes the lowest dose
elicited a significantly greater increase in UDS activity
compared to control. The higher doses which produced
UDS in rat hepatocytes was cytotoxic to mouse and
human cells. The MTD in these cell cultures produced
only a marginal UDS response
An increase in RDS was reported in rats exposed to UG,
but was not significant due to the variability between
animals. Treatment with TMP caused a significant
increase in the number of cells in S-phase.
UG induced a significant S-phase response in male mice
but not female mice. TMP significantly increased RDS
in both sexes.
Significant (5%) increase in the number of cells with
aberrations at the intermediate dose level, bul nol dose-
related trend was observed
No increase on the number of cells with chromosomal
aberrations was observed.
REFERENCE
Loury et al., 1986
Litton Bionetics, 1977
\p
Ul
-------
TABLE 9-1 CONTINUED
VO
ON
TEST TYPE
Mutation of rat
bone marrow
cells
Dominant lethal
assay
TEST SPECIES
UG administered orally
for 5 consecutive days
Sperm cells of CD-I
mice
CONCENTRATION
500, 150, and 1000
mg/kg
Inhalation exposure
of 0,400 or 1600
ppm; 6hr/d for 8 wks
RESULTS
No significant increase in the number of cells with
chromosome aberrations compared to control value.
Non-significant increases in pre- and post- implantation
loss of embryos as compared to control
REFERENCE
Litton Bionetics, 1977
Dooley et al., 1988
-------
the authors noted that the concentration of mutagens was probably too low to be detected in
the assay.
Unleaded gasoline (PS-6) was also evaluated in a series of microbial bioassays
conducted by Litton Bionetics (1977a). The microbial assays tested for mutagenic activity,
with and without metabolic activation, using bacteria and yeast in both plate and suspension
techniques. The results of the plate assays were negative. In the suspension tests without
activation, an increase in mutation at the highest doses was observed, but was not
statistically significant Statistically significant increases in mutagenicity in one strain (TA
98) were not reproducible. Scattered increases under metabolic activation were not
reproducible and attributable to anomalous reductions in viable cell counts. The adequacy
of these results are difficult to assess because of methodological problems encountered in
the assays, including the inability of unleaded gasoline to dissolve in DMSO and the use of
open plate treatment conditions to test a volatile mixture (U.S EPA, 1988).
9.1.1.2 Mammalian Cells
Dooley et al. (1987) also tested UG, an extraction of UG, and evaporative residue
of UG for increases in the mutant frequency in the mouse lymphoma assay. In this assay,
the lymphoma cells are heterozygous for autosomal mutations at the Thymidine Kinase
locus and are unable to grow in a medium containing bromodeoxyuridine. Cells that have
undergone mutation, however, can grow in this medium. Dooley et al. reported that at
concentrations of UG yielding 10 percent or greater cell survival, no appreciable increase in
mutation frequency with or without activation was observed. UG produced a two-fold or
greater increase in the mutant frequency in the mouse lymphoma assay with and without
metabolic activation. However, this was not considered a mutagenic response because it
occurred at concentrations yielding total growth of less than 10 percent Based on these
results, the authors suggest that UG may contain weakly mutagenic components that are
masked by the toxicity of the total mixture. The non-activated UG extraction and the
activated evaporative residue induced dose-related increases in the mutation frequency in
the mouse lymphoma assay. Although the evaporative residue produced a three-fold
increase in mutagenic activity, it was not considered a mutagenic response because total
growth of this culture was less than 10 percent. The increase appeared to be due to a
reduction in cloning efficiency rather than an increase in the total number of mutant
colonies. The authors concluded, however, that the positive responses elicited from the
two fractions demonstrated the presence of at least two different mutagenic components--
one requiring metabolic activation with a high boiling point and the other a direct-acting
9-7
-------
mutagen-neither of which was present in sufficient quantity in the whole gasoline mixture
to elicit a mutagenic response.
Farrow et al. (1983, as cited in Dooley et al., 1987) reported that UG, both in the
presence and absence of metabolic activation, was mutagenic in the mouse lymphoma
assay. In contrast, Litton Bionetics (1977a) also tested UG in the mouse lymphoma assay
and reported negative results. Two dose levels (0.065 and 0.52 ul/ml) showed an elevated
frequency of mutations above negative controls, but no dose-response was observed at
higher doses. Evaluation of these results by the U.S. EPA (1988) indicated that an
adequate reduction in the suspension growth was not achieved and, therefore, UG was not
adequately tested in this assay.
Richardson et al. (1986) reported no significant increases in the mutation frequency
in human lymphoblasts tested with UG, 2,2,4-trimethylpentane and a volatile fraction of
UG.
9.1.2 Studies of Chromosomal Aberrations
9.1.2.1 Mammalian Cells
Fregda et al. (1979) evaluated chromosome aberrations among 65 males
occupationally exposed to benzene through fuel handling (see Table 9-2). The study
consisted of six groups: two groups of road tank drivers who delivered gasoline, crew
members of coastal tankers, employees at gasoline filling stations, road tank drivers who
delivered milk in the same area, and workers in an industrial gasworks who were exposed
to benzene. Occupational groups were paired for the analysis of chromatid and
chromosome aberration.
Road tank drivers who delivered gasoline (average daily exposure to benzene was
0.4 ppm) had a significantly increased frequency of chromosome and chromatid breaks as
compared with gasoline station attendants (p <0.002) and ship's crew members (p <0.05)
whose average daily exposures to benzene were 0.084 ppm and 6.56 ppm, respectively.
The authors cautioned that while high benzene exposures were encountered in the crew
members, this group was less frequently exposed than the road tank drivers. Comparing
results with a control group from another investigation, the authors found that the
frequency of chromosome and chromatid breaks was lower among controls than in any
group of road tanker drivers or gasworks employees, but of the same magnitude as ship
tanker crews and gasoline station attendants. The authors concluded that the significantly
increased frequency of chromosome aberrations in all road tanker drivers may not have
9-8
-------
TABLE 9-2
GENOTOXIC EFFECTS IN HUMANS FOLLOWING OCCUPATIONAL
EXPOSURE TO GASOLINE
Test material Occupation
Reported results
Reference
Oil and motor
fuels; benzene
exposure
Automobile fuels
exhaust
Road tank drivers who
delivered gasoline, tank
drivers who delivered
milk, coastal tanker crews,
gas station attendants,
industrial workers.
Drivers of diesel engine
trucks, drivers of gasoline
engine trucks, automobile
inspectors.
Significant increase in Fredga et al.,
chromosome and chromatic! 1979.
breaks in road tank drivers
delivering gasoline, but the
frequency was comparable
to that seen in road tank drivers
delivering milk.
Chromosome breaks in non- Fredga et al.,
smoking drivers of diesel 1982.
trucks but not among smokers.
Authors concluded that the higher
frequency among nonsmokers
may be a random event SCE
observed in truck drivers who
smoked.
9-9
-------
been primarily due to benzene exposure since both those delivering gasoline and milk had
the same incidence of chromosome aberrations.
In a more recent study, Frcdga et al. (1982) investigated the incidence of
chromosomal changes in men occupationally exposed to automobile fuels and exhaust
gases. Blood samples were taken from 12 men in each of four groups: drivers of diesel
engine trucks, drivers of gasoline engine trucks, automobile inspectors, and a reference
group. The men in the groups were matched with respect to age, smoking habits, and
length of employment The analysis of chromosomes showed that smokers exposed to
either gasoline or diesel fuel had a higher frequency of sister chromatid exchanges (SCE)
than did the non-smokers (p <0.05). However, the authors were unable to uncover a
strong association between the work environment and frequency of chromosomal changes.
Richardson et al. (1986) also reported no increase in SCE frequency in human
lymphoblasts exposed to UG, 2,2,4-trimethylpentane, and a volatile fraction of UG.
In a study sponsored by Phillips Petroleum (1984), a significant increase in
frequency of SCE in Chinese hamster ovary cells exposed to UG was observed; however,
the increase was less than two-fold above background, not dose-related, and observed only
at the highest dose with metabolic activation.
Litton Bionetics (1977b) reported, in a study to determine chromosome aberrations
in bone marrow cells isolated from rats exposed to UG, a significant but not dose-related
increase in the number of chromosomal aberrations in rats acutely exposed to 0.08 ml/kg.
There was no such increase in rats exposed subchronically to 0.01,0.03, or 0.1 ml/rat UG
for 5 days or after repeat acute exposures. Similarly, oral exposure to 500,750, and 1,000
mg/kg/day did not induce a significant increase in chromosomal aberrations in Sprague-
Dawley rats (Dooley et al.. 1987).
9.1.3 Studies of Chemical Alternations of DNA
9.1.3.1 Mammalian Cells
Loury et al. (1986b) conducted a series of in vivo and in vitro assays to assess the
ability of UG and 2,2,4-trimethylpentane to induce unscheduled and replicative DNA
synthesis (UDS and RDS, respectively) in Fischer 344 rats, B6C3F1 mice, and human
hepatocytes. UDS is a measurement of DNA repair, and therefore, an indicator of
genotoxic activity. RDS is an indirect measure of cell proliferation activity. For the in vivo
assay, UDS and RDS were measured in hepatocytes isolated from rats and mice treated by
gavage with UG and 2,2,4-trimethylpentane. In addition, primary hepatocyte cultures
9-10
-------
were isolated from rats, mice and humans and incubated with UG (0.005 percent to 0.100
percent v/v) and IMP in sealed bottles for the in vitro assay.
In the in vitro UDS assay, a dose-dependent increase in UDS was observed in rat
hepatocytes at two concentrations (0.05 percent to 0.10 percent v/v) of UG. (Only male rat
hepatocytes were used in this assay). Mouse and human hepatocytes had significantly
higher UDS activity at 0.01 percent UG. At higher concentrations (0.03 percent and 0.05
percent), UG was cytotoxic to the mice and human hepatocytes and elicited only a marginal
UDS response. Hepatocytes isolated from rats exposed in vivo to 100,2,000, and 5,000
mg/kg UG and 500 mg/kg IMP showed no increase in DNA repair activity after 2,12, 24,
or 48 hours. The authors reported that the highest UG dose (5,000 mg/kg) was toxic to the
rats. A statistically significant increase in UDS was observed in hepatocytes isolated from
male and female mice 12 hours after treatment with 2,000 mg/kg UG. Cell viability in mice
was lower at the 5,000 mg/kg dose indicating toxicity to the mouse liver at this level.
TMP caused a significant increase in RDS in hepatocytes isolated from rats and
mice 24 hours after treatment. A nonsignificant increase in RDS in hepatocytes from rats
treated with 2000 mg/kg UG was observed 24 and 48 hours after exposure. UG induced a
significant increase in RDS in male mice, but not in female mice hepatocytes. This
response was considerably less than that observed following treatment with TMP.
Loury et al. (1987) also assessed the ability of UG to induce UDS and RDS in
kidney cells of Fischer 344 rats exposed in v/vo or in vitro by gavage (135, 200, and 5,000
mg/kg) and inhalation (0 and 2000 ppm) to UG. The purpose of the study was to
determine whether the induction of kidney tumors by UG is related to genotoxicity or cell
proliferative effects. Following inhalation exposure to 2,000 ppm UG, no increase in UDS
activity was observed. The authors note that repair activity may be occurring but below the
detection limit of the assay. Higher doses (2,000 and 5,000 mg/kg) administered by
gavage also did not induce UDS activity above the detection limit of this assay. No
evidence of DNA repair in primary kidney cells was observed. RDS activity was
significantly increased in male kidney cells at 200 mg/kg after 4 days of treatment
Significant increases in RDS activity were observed in the male rat kidney.
Although the RDS activity increased in female rats exposed to UG, the increase was not
statistically significant.
9-11
-------
9.2 BENZENE
9.2.1 Gene Mutation Studies
Benzene is not mutagenic in bacterial systems, but is a clastogen and causes other
chromosomal changes. In addition, data indicate that benzene must be metabolized before
chromosomal changes are induced. Table 9-3 presents a summary of the genotoxicity
studies of benzene. Selected studies are discussed in further detail below.
9.2.1.1 Bacterial Cells
The only positive results indicating the mutagenic potential of benzene or its
metabolites have been reported in the vascular plant Tradescantia where mutagenic
metabolites were apparently formed (Schairer et al., 1978; Schairer and Sautkulis, 1982),
in yeast (Parry and Eckardt, 1985; Parry, 1985), and in Drosophilia melanogaster (Wurgler
et al., 1985, as cited in Dean, 1985b). McCarroll et al. (198 la, b) obtained evidence of
benzene-induced DNA damage in tests using repair-deficient strains of Escherichia coli and
Bacillus subtilis.
9.2.2 Studies of Chromosomal Aberrations
9.2.2.1 Mammalian Cells
Chromosome aberrations in mammalian cells have been observed in vitro in
Chinese hamster lung fibroblasts (Ishidata and Sofuni, 1985, as cited in Dean, 19855), in
Chinese hamster ovary cells (Danford, 1985; Palitti et al., 1985, as cited in Dean, 1985b),
and in human leukocytes (Koizumi et al., 1974) and lymphocytes (Howard et al., 1985;
Morimoto, 1974).
Dean (1985b) reviewed the available in vivo cytogenic assays conducted with
benzene and concluded that benzene causes cytogenic damage in somatic cell of mice, rats,
and rabbits, but not hamsters. Gad-el Karim et al. (1984) observed that male rodents are
more sensitive to the clastogenic effects of benzene than are females. Dean (1985b)
concluded from the results of chronic inhalation studies that repeated benzene inhalation
exposures in rats (Styles and Richardson, 1984), but not mice (Luke et al., 1985), appear
to induce enzyme systems that reduce the generation of clastogenic metabolites.
Benzene induced significant dose-related increases in the frequency of SCE in
peripheral blood lymphocytes and micronuclei in bone marrow cells of mice and rats
exposed in vivo to 1,3,10, and 30 ppm benzene (Erexson et al., 1985). The induction of
SCE at 1 ppm in rats was considered borderline and statistically significant using the t-test
9-12
-------
TABLE 9-3
GENETIC TOXICOLOGY OF BENZENE*
Test system
Endpoint
Result Reference
Bacterial
Salmonella typhimurium Gene mutation
Bacillus subtilis
Escherichia coli
E. coli
Gene mutation
DNA danage
Gene mutation
Nonmammalian Eukaryotes
Saccharomyces cerevisiae Mitotic gene
conversion
Mitotic crossing over
Gene mutation
Aspergillus nidulans
Tradescanaa
Mitotic segregation
Drisphilamelanogaster Gene mutation
Lyon, 1985*
Cptruvo et al., 1977*
Simmon et al., 1977*
Shahin and Fournier, 1978*
Kaden et al., 1979*
Bartsch et al., 1980*
Florin et al., 1980*
Hoetal., 1981*
Hermann, 1981*
Venit, 1985*
Tanooka, 1977*
+ McCanoll et al., 198la*
+ McCanoll et al., 198Ib*
Rosenkranz and Leiger, 1980*
Parry and Eckardt, 1985
Summarized in Parry, 1985**
Parry and Eckardt, 1985
Egilsson et al., 1979*
Parry, 1985**
Crebellietal., 1986
Schairer et al., 1978*
Schairer and Sautkulis, 1982*
Nylander et al., 1978*
Kale and Baum, 1983*
Wurgleretal., 1985**
Fujikawa et al., 1985**
9-13
-------
TABLE 9-3
(continued)
Test system
Endpoint
Result Reference
Mammalian Cells (in vitro)
Rat liver (RL4) cells
Rat primary hepatocytes
Mouse lymphoma
Mouse fibroblasts
Mouse C3H/10T1/2
Chinese hamster
lung fibroblasts
Chinese hamster
ovary cells
Chinese hamster V79
Syrian hamster
embryo cells
HeLa cells
Human leukocytes
Sister-chromatic!
exchanges
Unscheduled DNA
synthesis
Gene mutation
Cell transformation
Cell transformation
Chromosome
aberrations
Sister-chromatid
exchanges
Gene mutation
Gene mutation
Cell transformation
Unscheduled DNA
synthesis
Chromosome
aberrations
Summarized in Dean, 1985a**
Probst etal., 1981
Williams etal., 1985**
Glauert et al., 1985**
Lebowitz et al., 1979*
Garner, 1985*
McGregor and Ashby, 1985**
Lawrence and McGregor, 1985
Ishidate and Sofuni, 1985**
Palitti et al., 1985**
Danford, 1985**
Dean, 1985a**
NTP, 1986b*
Dean, 1985a**
Garner, 1985**
Gamer, 1985**
Amacher and Zelljadt, 1983
McGregor and Ashby, 1985**
Barrett and Lamb, 1985
Sanner and Rivedal, 1985
Martin and Campbell, 1985**
Barrett, 1985**
Koizumi et al., 1974*
9-14
-------
TABLE 9-3
(continued)
Test system
Endpoint
Result Reference
Mammalian Cells (in vitro) (continued)
Human leukocytes Chromosome
aberrations
Sister-chromatid
exchanges
Human lymphoblasts
Mammals (in vivo)
Mouse
Gene mutation
Micronuclei
Chromosome
aberrations
Sister-chromatid
exchanges
Sperm morphology
+ Morimoto, 1974*
Gerner-Smidt and Friedrich
1978*
+ Howard et al.. 1985**
Gemer-Smidt and Friedrich
1978*
Morimoto and Wolff, 1980*
+ Morimoto, 1983*
+ Morimoto et al., 1983*
+ Ercxson et al., 1985
Gamer, 1985**
+ Diaz et al., 1980*
+ Kite etal., 1980*
+ Meyne and Legator, 1980*
+ Siou etal., 1981*
+ Tunek et al., 1982*
+ Tice etal., 1984**
+ Gad-El Karim et al., 1984**
+ Erexson et al., 1984**
+ Choy et al., 1985
+ Luke et al., 1985**
+ NTP, 1986b*
+ Erexson et al., 1986
+ Meyne and Legator, 1980*
+ Tice et al., 1980*
+ Siou etal., 1981*
+ Tice etal., 1982*
+ Gad-El Karim et al., 1984**
+ Tice etal., 1980*
+ Tice etal., 1982*
+ Erexson et al., 1986
+ Erexson et al., 1984**
+ Topham, 1980*
9-15
-------
TABLE 9-3
(continued)
Test system
Endpoint
Result Reference
Mammals (in vivo) (continued)
Chinese hamster
Rabbit
Human
Micronuclei
Chromosome
aberrations
Sister-chromatid
exchanges
Sperm morphology
Dominant lethal
Micronuclei
Chromosome
aberrations
Chromosome
aberrations
Chromosome
aberrations
Lyon, 1975*
Erexson et al., 1986
Lyon, 1975*
Dean, 1969
Anderson and Richardson, 1981
Styles and Richardson, 1984**
Erexson et al., 1986
Lyon, 1975*
Bio/dynamics Inc., 1980
Siouetal., 1981*
Siouetal., 1981*
Kissling and Speck, 1971*
Fredga et al., 1979**
Watanabe et al., 1980**
Reviewed by IARC, 1982
Van Raalte and Grasso, 1982**
Sarto et al.. 1984**
Clare etal., 1984**
* As cited in NTP, 1986b.
** As cited in Dean, 1985a.
9-16
-------
(P = 0.036) but not the Mann-Whitney U test (P = 0.055). However, the frequency of
SCE was dose-related to the higher doses. Comparison of the dose-response curves for
rats and mice indicate that the shape of the curve for mice and rats is non-linear with the
greatest slope at the lower concentrations. Sister chromadd exchanges have also been
detected in human lymphocytes (Morimoto, 1983; Morimoto et al., 1983).
Toft et al. (1982) evaluated the genotoxic potential of continuous and intermittent
exposure to 21,50, and 95 ppm benzene in bone marrow cells of NMRI mice. The study
measured three bone marrow parameters: number of nucleated cells, number of colony
forming granulopoietic stem cells, and frequency of micronuclei in polychromatic
erthrocytes. The number of nucleated cells and stem cells were significantly reduced at all
exposure doses administered from 4 to 10 days. A statistically significant increase in
micronuclei was observed in mice exposed to 14 ppm from 1 to 8 weeks. Intermittent
exposure to 21 ppm benzene, 5 days/week, 8 hours/day for 2 weeks caused a suppression
in stem cells, and an increase in the frequency of micronuclei, but no change in the number
of nucleated cells. No effects were observed at the 14 ppm exposure for 8 weeks. At
higher concentrations, for short durations, the number of nucleated cells was suppressed,
but not the stem cells. The stem cells appear to be more sensitive to prolonged exposures
to low levels of benzene compared to the nucleated cells. For example, more severe effects
were observed with exposure to 50 ppm, 8 hours/day (4,000 ppm) than with exposure to
201 ppm for 2 hours/day (4,020 ppm).
Toft et al. (1982) also investigated the effects of benzene exposure on the frequency
of micronuclei in mouse polychromatic erythrocytes (PCE). An 8-hour exposure to 95
ppm of benzene increased the frequency of these micronuclei from a control value of about
0.4/500 PCE to 10-15/500 PCE. Mice exposed for 8 hours to 201 ppm had a micronuclei
frequency of 20-25/500 PCE. Subacute continuous exposure to 21,50, and 95 ppm of
benzene increased the number of micronuclei; 35/500 PCE after 2 days of exposure to 95
ppm, 20-30/500 PCE after 5 days of exposure to 50 ppm, and about 5/500 PCE after 4
days of exposure to 21 ppm. Longer exposure durations did not elevate the frequency of
micronuclei in the 21 ppm exposure level. Continuous exposure of 14 ppm benzene
caused a significant elevation in micronuclei (1.75/500 PCE) relative to controls (0.41/500
PCE). Continuous exposure over the next 7 weeks did not elevate this frequency.
Intermittent exposures to benzene at 1 to 10 ppm levels produced no significant changes in
micronuclei frequency.
Benzene induced sperm-head abnormalities in mice (Topham, 1980), but not in rats
(Lyon, 1975). Many studies have reported chromosome aberrations in both peripheral
lymphocytes and in bone marrow cells of workers exposed to benzene at concentrations
9-17
-------
known to be associated with benzene-induced blood dyscrasias (Fielder, 1982, and
references cited therein). These abnormalities often persisted for many years after the
benzene exposure was terminated and the blood picture had returned to normal (Fielder,
1982).
9.2.2.2 Mammalian Cell Transformation
Benzene painted directly onto the ovaries of mice that were subsequently mated has
been associated with tail defects and subcutaneous hemorrhages in the progeny (Sridharan
et al., 1963, as cited in IARC, 1982); these findings were reported to persist through four
generations. Cell transformation has been demonstrated in Syrian hamster embryo cells
(Amacher and Zelljadt, 1983; Barrett and Lamb, 1985; McGregor and Ashby, 1985;
Sanner and Rivedol, 1985), but not in C3H/10T1/2 cells (Lawrence and McGregor, 1985).
Transformation of 3T3 Swiss mouse cells by herpes simplex type 2 virus was induced by
benzene (Johnson, 1982).
9.3 TOLUENE
9.3.1 Gene Mutation Studies
9.3.1.1 Bacterial Cells
Toluene was reported to be negative in a battery of microbial, mammalian cell, and
whole organism test systems. The microbial assays conducted include differential toxicity
testing with wild-type and DNA repair-deficient strains of E. coli and S. typhimurium
(Fluck et al., 1976; Mortelmans and Riccio, 1980), reverse mutation testing with various
strains of S. typhimurium, E. coli WP2, and S. cerevisiae D7 (Litton Bionetics, 1978a;
Mortelmans and Riccio, 1980; Nestmann et al., 1980), and mitotic gene conversion and
crossing-over evaluation in S. cerevisiae D4 and D7 (Litton Bionetics, 1978a; Mortelmans
and Riccio, 1980).
9.3.1.2 Mammalian Cells
Toluene also failed to induce specific locus forward mutation in the L5178Y
Thymidine Kinase mouse lymphoma cell assay (Litton Bionetics, 1978a), was negative in
the micronucleus test in mice (Kirkhart, 1980), and was negative in the mouse dominant
lethal assay (Litton Bionetics, 1981). Sister-chromatid exchange frequencies were not
altered in Chinese hamster ovary cells cultured with toluene in vitro (Evans and Mitchell,
1980).
9-18
-------
9.3.2 Studies of Chromosomal Aberrations
9.3.2.1 Mammalian Cells
In the Russian literature, chromosome aberrations were reported in the bone
marrow cells isolated from rats exposed to toluene subcutaneously (Dobrokhotov, 1972;
Lyapkalo, 1973) and by inhalation (Dobrokhotov and Enikeev, 1975). These findings
were not corroborated, however, in a Litton Bionetics (1978a) study of rats administered
toluene by intraperitoneal injection.
Workers who were chronically exposed to toluene were reported to have an excess
of chromosome aberrations and an increased frequency of sister-chromatid exchanges in
their lymphocytes (Bauchinger et al., 1980; Funes-Craviota et al., 1977). These findings
were not corroborated in studies by Forni et al. (1971) or Maki-Paakknen et al. (1980).
However, it is probable that part of the exposure in the Funes-Craviota et al. (1977) study
was to benzene-contaminated toluene. Sister-chromatid exchange frequencies were
unchanged in human lymphocytes cultured with toluene in vitro (Gemer-Smidt and
Friedrich, 1978).
9.4 XYLENE
9.4.1 Gene Mutation Studies
9.4.1.1 Bacterial Cells
Xylene tested negative in mutagenicity tests in Salmonella typhimurium strains
TA98, TA100, TA1535, and TA1537 with or without metabolic activation (Florin et al.,
1980; Bos et al., 1981; NTP, 1986a). Negative results were also obtained in the rec-assay
tests in Escherichia coli (McCarroll et al., 198 Ib). M- and o-xylene were negative in the D.
melanogaster recessive lethal test, whereas technical xylene was weakly mutagenic.
9.4.2 Studies of Chromosomal Aberrations
9.4.2.1 Mammalian Cells
As Table 9-4 indicates, xylene did not cause chromosome aberrations in rats
(Donner et al., 1980). The numbers of chromosome aberrations were not greater than
control levels in a group of 17 painters exposed to xylene and 10 other solvents (Haglund
et al., 1980). The number of SCE and structural chromosome aberrations in human
9-19
-------
TABLE 9-4
MUTAGENICITY TESTING OF XYLENE
WD
Assay
Reverse
mutation
Reverse
mutation
Reverse
mutation
Reverse
mutation
Reverse
mutation
Reverse
mutation
Reverse
mutation
Indicator/
organism
Salmonella
typhimurium
TA98, TA100,
TA1535.TA1537
S. typhimurium
TA98. TA100,
TA1535
S. typhimurium
TA100.TA1535.
TA97, TA98
S. typhimurium
TA100.TA1535,
TA1537.TA98
S. lyphimurium
TA100.TA1535,
TA100.TA1535
S. lyphimurium
TA100.TA1535,
TA100.TA1535
S. typhimurium
TA100.TA1535.
TA100.TA1535
Compound Application
and/or purity
m-xylene
m-xylene
mixed xylene
(see Section S.I
for composition)
o-xylene
m-xylene
p-xylene
o-xylene
plate
plate
plate
plate
plate
plate
plate
Concentration
or dose
3, 32, 320,
and 3,200 |ig/
plate
3. 32. 320.
and 3,200 Mg/
plate
0. 3, 10. 33
100, and 200
Mg/plate
0. 1.0. 3. 3.
100. and 200
Mg/plate
0.0.3. 1.3.
10. and 33
Mg/plate
Activating Response
system
±S-9 -
±S-9 -
±S-9 -
±S-9 -
±S-9 -
0. 1. 3.3. 10, ±S-9 -
33. 100. and 200
Mg/plate
20. 50. 100.
200. and 500
Mg/plate
±S-9 -
Comment Reference
Toxic to bacteria Florin et al.,1980.
at 3.200 (ig/plate
Toxic to bacteria Florin et al., 1980.
at 3,200 Mg/plate
Slightly toxic at NTP, 1986a.
333 Mg/plate
Slightly toxic at NTP. 1986a.
333 Mg/plate
NC NTP. 1986a.
Slightly toxic at NTP. 1986a.
200 Mg/plate
NC Bos etal., 1981.
-------
NO
Assay
Reverse
mutation
Reverse
mutation
Rec-assay
Recessive
lethal
Chromosome
aberrations
Chromosome
aberrations
Chromosome
aberrations
Indicator/
organism
S. lyphimurium
TA100, TA1535,
TA100.TA1535
S. typhimurium
TA100.TA1535,
TA100.TA1535
Escherichia
coli WP2. WP2
uvr A, CM611,
WP67, WP100.
W3110.p3478
Drosophila
melanogaster
Rat
Human
lymphocytes
in vitro
Group of
painters
Compound Application i
and/or purity <
m-xylene plate
p-xylene plate
technical microtiter
xylene plate and
glass tubes
m-xylene NR
o-xylene
technical xylene
xylene inhalation
mixture
xylene whole blood
cultures
xylene and 10 NR
other solvents
TABLE 9-4
(continued)
Concentration Activating
or dose system
20. 50, 100, ±S-9
200. and 500
Ug/plaie
20. 50. 100. ±S-9
200, and 500
jig/plale
Starting at 1.314 ±S-9
or 1,752 ng/well
with 1 1 twofold
dilutions
NR NA
300 ppm (1.303 NA
mg/m3), 6 hrs/
day. 5 days/week
for 9. 14. and
18 weeks
15.2. 152. 1.520 none
Hg/m3 untreated
25 ppm (109 NA
mg/m3)
Response Comment
NC
NC
NC
NC
weakly 1-5% recessive
mutagenic lethals observed
Number of rats
exposed and marrow
cells analyzed were
not reported
NC
NC
Reference
Bosetal., 1981.
Bos et al.. 1981.
McCarroll et al.,
19815.
Dormer et al.. 1980.
Dormer et al., 1980.
Gemer-Smidt and
Friedrich, 1978.
Haglund et al.. 1980.
NR = Not reported
NA = Not applicable
NC = No comment
-------
lymphocytes treated with xylene in vitro was not significantly different from controls
(Gemer-Smidt and Friedrich, 1978).
9.5 SUMMARY
Unleaded gasoline and benzene, toluene, and xylene have been evaluated for
genotoxic effects in a variety of test systems. Generally, gasoline is not mutagenic in
bacterial systems, while slight but mixed evidence for mutagenic activity is found in
mammalian cells. A study of humans exposed to gasoline found chromosomal aberrations,
but the possible presence of other mutagens appears to have confounded the study.
Genotoxic data for benzene suggest it is clastogenic, with strong evidence for a response in
mammalian cells. Toluene and xylene studies are limited with negative evidence for
mutagenicity and evidence of chromosomal aberrations in a study of workers and rats
exposed to toluene.
9-22
-------
10. CARCINOGENICITY
10.1 EPIDEMIOLOGIC EVIDENCE FOR HUMAN CARCINOGENICITY
There are a large number of chronic studies that evaluate the effects of the inhalation
of gasoline vapors in the gasoline service industry and petroleum-refining industry. This
review briefly summarizes the studies that are relevant to the exposure of gasoline vapors
via inhalation in the gasoline service industry and the qualitative risks of gasoline vapor-
induced carcinogenicity. Detailed descriptions of these studies are provided in a review
submitted by Dynamac Corporation to the U.S. EPA (Dynamac, 1986). A brief overview
of pertinent studies of workers in the petroleum industry is also included in this review.
Since unleaded gasoline was introduced in the mid-1970s, even recent epidemiological
studies are not likely to show an unleaded gasoline effect because of the long latency period
generally associated with cancer. Therefore, this review is not limited to unleaded gasoline
exposure but rather addressed any potential gasoline exposure. Table 10-1 summarizes all
studies reviewed. The most pertinent studies will be further discussed in this health
evaluation.
In general, these studies had many limitations and deficiencies. The major
limitation was the lack of information on gasoline vapor exposure; no quantitative estimates
were reported for any of the studies reviewed. In addition, workers in gasoline service
stations are potentially exposed to other petroleum products (e.g., motor oils, diesel fuel
oils, solvents) as well as automobile and truck engine exhausts. It was not possible to
assess the effects of gasoline exposure independently from these other exposures. Most of
these studies assessed employment in the gasoline industry, rather than exposure to
gasoline per se, for an association with elevated cancer incidence. In general, the studies
did not adjust for confounding variables or conduct latency analyses. Due to the
multiplicity of types of occupational and non-occupational gasoline, oil, and hydrocarbon
exposures in these studies, it is difficult to implicate a single agent or class of compounds.
The studies do suggest, nonetheless, that exposure to one or more of these chemicals or
chemical mixtures is associated with an increased cancer risk.
10.1.1 Epidemiological Studies of Workers in the Gasoline Service
Industry
Ten epidemiological studies were reviewed (eight case-control studies and two
proportionate mortality studies) that evaluated the association between cancer risk and
employment in the gasoline service industry, which includes gasoline service station
owners and attendants, mechanics and garage workers, gasoline and fuel truck drivers, and
10-1
-------
TABLE 10-1
SUMMARY OF EPIDEMIOLOGICAL STUDIES REVIEWED
Reference
Study population/ Comparison population
Type of study
Occupational exposure
9
NJ
Stemhagen et al., 1983
Silverman et al., 1983
Primarily liver cancer cases/ Patients selected
from hospital records.
Lower urinary tract cancer cases/ Sample of
area telephone users and Medicare participants.
Mommsen et al.. 1982, Bladder cancer cases/ Registrants from the
1983; Mommsen and Denmark National Register.
Sell. 1983; Monmmsen and
Aagard, 1983a, b, c, d and 1984
Domiano el al., 1985
Lin and Kessler. 1981
Lin and Kessler, 1979
Milham, 1983
Tabershaw-Cooper,
1974 (unpublished);
197S (unpublished)
Renal cell carcinoma cases/ Patients
selected from hospital admission records.
Pancreatic cancer cases/ Patients selected
from hospital admissions records.
Testicular cancer cases/ Patients selected
from hospital discharge records.
Gasoline station owners and attendants;
gasoline and fuel truck drivers; fuel oil
dealers; auto mechanics and repairmen/
Washington State's mortality experience.
Petroleum refinery hourly workers at
17 U.S. refineries/ U.S. mortality
experience.
Case-control
Case-control
Case-control
Case-control
Case-control
Case-control
PMR
Cohort mortality
Gasoline service station employment.
Gasoline service industry; garage workers
and gas station attendants; trucking transporter
of petroleum products.
Work with oil or gasoline; work with petroleum
or asphalt
Gasoline service station employment.
Employment in gasoline service stations and
garages or dry cleaning business.
Employment in gasoline service stations;
employment in garages.
Usual lifetime occupation as gasoline station
owners or attendants; gasoline or fuel truck
drivers; fuel oil dealers; or auto mechanics
or repairmen.
Employment in one of the refineries for at
least 1 year; "high," "medium," or "low"
hydrocarbon exposure.
-------
TABLE 10-1
(continued)
Reference
Study population/ Comparison population
Type of study
Occupational exposure
Hauls, 1977
Male employees of the Imperial Oil Co./
Canadian modality experience; internal
"nonexposed" group; internal "nonrefinery"
group.
Cohort mortality
Employment by the company for at least 1 year
if an active employee and 5 years if a terminated
employee; daily exposure to crude petroleum,
gas. or breakdown products; daily refinery
site exposure.
Hanis et al., 1979
Theriault and Goulet.
1979.
Thomas et al., 1980
Reeve et al., 1982
Hanis et al., 1982
Rushton and Alderson,
1983
Male employees of the Imperial Oil Co./
Internal "nonexposed" group; internal
"nonrefinery" group.
Male employees at a Canadian oil refinery
plant/ Quebec, Canada, mortality experience.
Decedent males who at their time of death were
active members of the Oil, Chemical, and Atomic
Workers International Union (OCAW) in Texas/
U.S. and Texas mortality experiences.
Decedent males who had been members of
OCAW Local 4-449 in Texas City, TX/
U.S. mortality experience.
Regular plant employees who worked at the
Exxon Baton Route. LA, refinery and
chemical plant/ U.S. mortality experience
Men who had worked at three oil distribution
companies in Great Britain/ Combined mortality
experience of England and Wales.
Cohort mortality
Cohort mortality
PMR
PMR
Cohort mortality
Cohort mortality
Employment by the company for at least 1
year if an active employee and 5 years if a
terminated employee; daily contact with
petroleum or its products; moderately exposed
to petroleum or its products; employment
daily at a refinery site.
Employment at the refinery for 5 years
or more.
Union membership and employment al a
plant where the major operation was
petroleum refining or production of
petroleum products.
OCAW Local 4-449 membership and
employment at one of two petroleum
refineries in Texas City, TX
Employment at the refinery or chemical
plant for at least 1 month.
Employment of at least 1 year at one
of the oil distribution centers.
-------
TABLE 10-1
(continued)
Reference
Study population/ Comparison population
Type of study
Occupational exposure
Morgan and Wong,
1983 (unpublished)
Morgan and Wong.
1984 (unpublished)
Thomas el al., 1984
Wenetal.. 1981.1982.
1983.1984a
Rushton and Alderson.
1981a.b
Schottenfeld et al., 1981
Employees at two Chevron refineries/
U.S. mortality experience
Employees at Mobil Oil Co. refinery in
Beaumont, TX/ U.S. mortality experience.
Decedent cases who were active or retired
members of OC AW and whose cause of death
was brain tumor, stomach cancer, or leukemia/
Decedents selected from records of active and
retired OCAW members, excluding those whose
cause of death was the same as the cases' cause.
Employees at a Gulf Oil refinery/ U.S.
mortality experience
Male workers at any one of eight petroleum
refineries in Great Britain/ Combined mortality
experience of England and Wales for English
and Welsh refineries and the mortality
experience of Scotland for Scottish refineries.
White and Spanish sumamed men from 19 U.S.
petroleum industry companies/ U.S. mortality
experience; SEER program results.
Cohort mortality
Cohort mortality
Nested
case-control
Cohort mortality
Cohort mortality
Cohort mortality
and cohort
morbidity
Employment of at least 1 year at one of the
refineries.
Employment of at least 1 year at the
refinery.
Union membership and employment at one of
the three Beaumont-Port Arthur, TX, area
refineries. Refinery exposures were categorized
as crude treating, coking, grease plant, utilities.
maintenance and labor, receipt and movement,
laboratory, motor transport, and other work
categories.
Employment at the Port Arthur, TX site. The
refinery was engaged in refining crude oil and
manufacturing fuels, oils, lubricants, and
petrochemicals such as benzene, cumene,
eihylene. and cyclohexane.
Employment of at least 1 year at one of
the refineries
Employment by one of the 19 petroleum
industry companies, subsequently categorized
as refinery employment or petrochemical
employment.
-------
TABLE 10-1
(continued)
Reference
Study population/ Comparison population
Type of study
Occupational exposure
Thomas et al., 1982a
Wen et al., 1984c
(summary) Study 5
Wen et al.. 1984b
^ (summary) Study 6
9
Kaplay, 1982 (unpublished);
1985 (unpublished)
McGraw et al., 198S
Hanis el al.. 198Sa
Hanis et al.. 1985b
Divine et al.. 198S
Decedent males who at their lime of death were
members or retired members of the OCAW from
three refineries in Beaumont-Port Arthur, TX,
area/ U.S. and county mortality experience.
Gulf Oil Co. employees and annuitants/ U.S.
mortality experience.
Kidney cancer cases at a Gulf Oil Co.
refinery in Port Arthur. TX/ Noncancer
decedents; mixed controls; i.e., no
exclusion for cancer or vital status.
Petroleum refinery hourly workers at 17 U.S.
refineries/ U.S. mortality experience.
White male employees at Shell Oil Co. Wood
River Refinery/ U.S. mortality experience;
SEER mortality experience.
Regular employees from three Exxon refineries
and chemical plants/ U.S. mortality experience.
Regular employees from three Exxon refineries
and chemical plants with potential exposure to
petroleum, petro-chemicals. and other related
substances/ An internal "unexposed" group.
All employees of Texaco, Inc., who worked at
the refinery, petrochemical, and research
facilities/ U.S. mortality experience.
PMR
PMR
Nested case-
control
Cohort mortality
Cohort mortality
Cohort mortality
Cohort mortality
Cohort mortality
Union membership and employment at one of
the three Beaumont-Port Arthur, TX, area
refineries. The refineries are engaged in refining
crude oil and manufacturing fuels, petroleum
solvents, lubricating oils, petroleum wax,
greases, and petrochemicals.
Employed by Gulf Oil Co. at a U.S. site.
Employment at the Port Arthur, TX, refinery
Employment in one of the refineries for at
least 1 year.
Employment of at least 1 day at the refinery or
the onsite research laboratory.
Employment of at least 1 month at one of (he
refineries.
Employment of at least 1 month al one of (he
refineries and a job title of process operator,
mechanical worker, unskilled laborer, service
worker, laboratory technician, or field
professional.
Employment by Texaco, Inc. al one of their
refinery, petrochemical, or research facilities
for a minimum of 5 years.
-------
TABLE 10-1
(continued)
Reference
Study population/ Comparison population
Type of study
Occupational exposure
Bairon and Divine, 1985
Nelson. 1985
(unpublished)
H- Cole et al.. 1972
9
O\
Howe et al., 1980
Gottlieb and Pickle. 1981
Gottlieb and Carr, 1981
Najem et al., 1982
Decedent brain cancer cases; decedent benign
and unspecified brain tumor cases/ Controls
selected from the original Texaco, Inc.. cohort
Full-time regular employees from 10 U.S.
refineries of Amoco Oil Co./ U.S. mortality
experience.
Bladder cancer cases/ An age- and sex-stratified
sample of the adult population of the area.
Bladder cancer cases/ Neighborhood controls.
Decedent bladder cancer cases/ Decedents from
the same parish of residence, excluding those
whose cause of death was bladder cancer.
Decedents whose cause of death was cancer of
the lung, pancreas, bladder, brain, kidney.
or esophagus, or leukemia/ Decedents from
the same parish or residence as the cases.
excluding those whose cause of death was
the same as the cases' cause.
Bladder cancer cases/ Selected from the same
urology clinic and hospital population as
the cases.
Nested case-
control
Cohort mortality
Case-control
Case-control
Case-control
Case-control
(n = 7)
Case-control
Employment by Texaco, Inc., at one of their
refinery, petrochemical, or research facilities;
research laboratories, or lube oil refining units.
Employment at one of the refineries for at least
6 months; job type (administrative, etc.); contact
with light aromatic hydrocarbons; contact with
heavy oils; contact with refinery products; the
latter three exposures were categorized as none,
occasional, routine, or unknown.
Petroleum product occupations; petroleum
workers excluding machinists and mechanics.
Employment in the petroleum industry.
Usual history of employment in the oil refining
industry; living "near" oil refineries.
Usual history of employme.it in the oil
refining industry; living "near" oil refineries;
work in oil production; employment in the
petroleum industry; workers in oil exploration
and drilling; oil-field workers, welders, operators,
boilermakers, or painters in the petroleum
industry; skilled workers involved in petroleum
pumping and refining; auto repair business.
Employment in the petroleum industry.
-------
TABLE 10-1
(continued)
Reference
Study population/Comparison population
Type of study
Occupational exposure
McLaughlin et al., 1984
McLaughlin et al., 1983
Pickle and Gottlieb. 1980
Wigle, 1977
Gottlieb, 1980
Brandt et al.. 1978
Plotnikov. 1978
Renal cell carcinoma cases/ An age- and sex- Case-control
stratified random sample selected from area
telephone listings.
Renal pelvis cancer cases/ An age- and sex- Case-control
stratified random sample selected from area
telephone listings.
Decedent pancreatic cancer cases/ Decedents Case-control
from the same parish or residence, excluding
those whose cause of death was pancreatic
cancer.
Decendent lung cancer cases/ Decedents from the Case-control
same city, excluding those whose cause of
death was lung cancer.
Decedent lung cancer cases/ Decedents from Case-control
the same parish or residence, excluding those
whose cause of death was lung cancer.
Acute nonlymphocylic leukemia cases/ Case-control
Outpatient department patients treated for
nonmalignant disorders; outpatient department
patients treated for allergic diseases; patients
with chronic myeloid leukemia and chronic
lymphocylic leukemia.
Patients diagnosed with leukemia or Case-control
lymphogranulomatosis/ "Healthy" individuals.
Petroleum, tar, or pitch products; employment
in (he chemical or petroleum industry.
Petroleum, tar, or pitch.
Employment in the oil refining industry;
residence "near" oil refineries.
Employment in the oil refining industry; oil
production occupations; auto repair business;
employment in the oil exploration and drilling
industry; living in "close proximity" to oil
refineries.
Employment in the petroleum industry; oilfield
workers, welders, operators, boilermakers. or
painters in the petroleum industry; employment
as a skilled worker in petroleum pumping or
refining.
Petroleum products. Typical jobs held were
filling station attendant, bus or truck driver,
operator of excavating machines or power
saws, and road hauler.
Employment in the petroleum industry.
-------
TABLE 10-1
(continued)
Reference
Study population/ Comparison population
Type of study
Occupational exposure
Schwartz, 1986
Morgan and Wong. 1985
Decedent white male residents of New
Hampshire.
Decedents from Mobil Oil refinery at
Paulsboro, NJ/ U.S. mortality experience.
PMR
Automobile mechanics and service station
attendants.
Cohort mortality Employment in petroleum refinery industry.
Enterline and Henderson,
1985
Decedents from Mobil Oil refinery at
Torrance, CA/ U.S. mortality and Los
Angeles. CA, mortality experience.
Cohort mortality Employment in petroleum refinery industry.
-------
those who reported working with gasoline. Specific target organs and tissues were
investigated as cancer sites. These included the liver (Stemhagen et al., 1983), urinary tract
(Milham, 1983; Silverman et al., 1983), bladder (Milham, 1983; Mommsen et al., 1982;
Mommsen and Aagard, 1983a, b, 1984; Mommsen and Sell, 1983), kidney (Domiano et
al., 1985; McLaughlin et al., 1985), pancreas (Lin and Kessler, 1981), testes (Lin and
Kessler, 1979), and the hematopoietic tissues (Brandt et al., 1978; Schwartz, 1986). No
cohort mortality studies were identified for workers in the gasoline service industry.
10.1.1.1 Case Control Studies
The case-control studies that found the site-specific incidence of cancer to be
statistically significant (p<0.05) are listed in Table 10-2. Stemhagen et al. (1983) found
that male primary liver cancer cases were more likely to work in gasoline service stations
than were their matched controls. The odds ratio (OR), defined as the odds of having the
disease among the exposed divided by the odds of having the disease among the non-
exposed, was significantly higher among the gasoline exposed workers (OR = 2.99, 95%
confidence interval [CI] 1.20 to 6.88). This difference was evident when analyses were
limited to male cases with hepatocellular carcinoma and their matched controls, the risk
increased (OR = 4.20, 95% CI 1.55 to 11.35). However, in addition to the study
limitations mentioned above, no latency analysis was conducted and the study authors did
not control for the effects of alcohol consumption in the risk analysis reported for liver
cancer. The authors indicated that the number of males with gasoline service station
exposure were too few to analyze risks while simultaneously controlling for the effects of
alcohol consumption (a known risk factor and significant confounder). However, since the
risk for gasoline service station exposure (OR = 2.88, CI 1.20 to 6.88) was similar to the
risk for alcohol consumption (OR = 2.52, CI 0.97 to 6.54) for those in the medium-heavy
category of alcohol consumption (i.e., the category with the greatest risk), it is not likely
that alcohol consumption alone would account for this elevated risk among gasoline station
workers. This study raises the concern that hydrocarbon exposure in service stations,
either alone or in combination with alcohol or pre-existing disease (e.g., cirrhosis), may
elevate the human liver cancer risk.
Mommsen et al. (1982,1983a, b), Mommsen and Aagard (1983a,b, 1984), and
Mommsen and Sell (1983) found a significant (p<0.05) increase in the risk of bladder
cancer (OR = 2.71,95% CI 1.2.1 to 6.10) among men occupationally associated with
gasoline or oil products (U.S. EPA, 1987d). In addition, this trend persisted (OR = 2.32)
after adjusting for confounding variables (e.g., cigarette smoking, low socio-economic
level). However, the data collection and methods of study analysis were flawed and may
10-9
-------
TABLE 10-2
SUMMARY OF SITE-SPECIFIC STATISTICALLY SIGNIFICANT* CANCER FINDINGS FROM
CASE-CONTROL STUDIES THAT EVALUATED EMPLOYMENT IN THE GASOLINE SERVICE
INDUSTRY AS A RISK FACTOR
Type of
Cancer Reference
Primary Slemhagen et al. 1983C
liver
Final study population
(cases/ controls)
265 H78 M. 87 R
530 (356 M. 174 F)
Odds
ratio
(OR)
4.20
Occupational
exposure
Gasoline service
station employment
Comment
(study evaluation)
No latency analysis conducted, confounding
variable (alcohol consumption) not considered
in risk analysis. Study provides limited
evidence of an association. EPA considers
study inadequate.
9
»—*
o
Bladder
Lower
urinary
tract
Mommsen et al., 1982,
1983; Mommsen and Sell,
1983; Mommsen and Aagard,
1983a, b, 1984C
Silverman et al.. 1983C
Pancreas Lin and Kessler, 1981c
212 (165 M. 47 F)
259 (165 M. 94 F)
303 (M)
296 (M)
2.32
1.30
109 f67 M. 42 R
109 (67 M, 42 F)
2.79
Data collection and methods of study analysis
flawed; possible bias. Exposure duration and
latency not considered. Exposures included
gasoline and oil. Study provides insufficient
and possibly invalid data.
Large number of study comparisons (32
categories of industrial employment and
54 occupational categories) many have
elevated risk due to chance; no latency
analysis conducted. Possible bias of
data, weak evidence of association.
Employment in gasoline Occupational categories included gasoline and
service stations, garages dry cleaning - not possible to determine cause
or the dry cleaning for risk. Failure to control for confounding
business (10+ years) variables.
Occupational exposure
to gas and oil
Bus, taxi, and truck
drivers, gasoline
service stations
-------
TABLE 10-2
(continued)
Type of
Cancer
Testes
Reference
Lin and Kessler, 1979C
Final study population
(cases/ controls)
205HW
205 (M)
Odds
ratio
(OR)
Occupational
exposure
Comment
(study evaluation)
Employment in gasoline Only abstract of work published without
service stations or methodology or estimates of risk.
garages
a Significantly different from controls (p <0.05).
b Common deficiencies of all studies include mixed exposures and absence of quantitative measurement of gasoline vapor.
c As a result of study deficiencies, these studies provide insufficient evidence for an association between gasoline exposure
_^ and increased cancer risk.
-------
have created a bias in response. Specifically, the interviewer was not blind to the status of
the subjects as cases, data collection methods were mixed (person-to-person, phone, and
questionnaire), and exposure data were collected in a nonuniform manner (i.e., with
"unstructured questions"). In addition, exposure duration and latency were not considered
in the analysis, and the exposures cited included both gasoline and oil. Consideration of
exposure and latency could have strengthened the evidence for a causal inference, as it
would have provided information on the temporal association between exposure and effect,
and the nature of the dose-response relationship. It is not known to what extent the
methodological flaws may have influenced the results.
Lin and Kessler (1981) reported that men with pancreatic cancer were significantly
more likely to have occupations involving close exposure to gasoline (e.g., service stations
and garages), or to be in the dry cleaning business than were the controls (OR = 2.79).
This risk appeared to be positively associated with increasing exposure (3 to 5 years
exposure, OR = 1.27; 6 to 10 years exposure, OR = 3.80; 10 or more years exposure, OR
= 5.07). However, it was not possible to determine if the observed increased in risk was
due to employment in gasoline occupations, dry cleaning occupations, or some combined
exposure to both. In addition, the authors' analysis failed to control for the effects of
significant confounding factors. On the other hand, except for cigarette smoking (and
possibly diabetes and coffee drinking), potential risk factors have yet to be identified for
pancreatic cancer. Furthermore, since dry cleaning solvents are suspected carcinogens, this
study suggests a correlation between either employment in gasoline stations or the dry
cleaning industry and elevated cancer risk.
Lin and Kessler (1979) also reported that cases of testicular cancer were
significantly more likely to be gasoline station attendants and garage workers than were the
controls. Only an abstract of this work was published, and no estimates of risk were
presented. The lack of any methodologic detail precludes consideration of the data.
Silverman et al. (1983) found that workers in the gasoline service industry had 30
percent greater risk of bladder cancer than those not working in gasoline service (adjusted
OR = 1.3, p value not reported). Truck drivers were found to be at significant risk even
after controlling for age and smoking (adjusted OR = 2.1). This risk increased
significantly (p = 0.004) with increased duration of employment as a trucker. The risk
among truckers who transported petroleum products, after adjusting for age and smoking
(adjusted OR = 3.6), was elevated to more than three and a half times that of non-truckers.
Risks were also elevated (p>0.05) for cab drivers (unadjusted OR = 2.0), bus drivers
(unadjusted OR = 1.5), and delivery men (adjusted OR = 1.8)~all occupations with
potential gasoline exposure. Although actual exposures were not known, the risks seemed
10-12
-------
to be the highest for those with the highest exposure to petroleum (i.e., truckers who
transported petroleum products). However, it was not possible to evaluate whether the
excess risk could be attributed to diesel exhaust, gasoline exhaust, diesel vapors, gasoline
vapors (leaded or unleaded), or some other exposure. There was no quantification of
gasoline vapor exposure.
Latency analyses were not conducted in the Silverman study (1983). Excluded
from the study population of 303 white male cases were 86 cases of lower urinary tract
cancer. The authors presented no assessment of the impact of this potential bias. Since a
total of 32 categories of industrial employment and 54 occupational categories were
studied, some of the observed elevations in risks for bladder cancer cases for certain
occupations may have been due to chance because of the large number of comparisons
made in the study. On the other hand, the consistently elevated response among the
various occupations involving gasoline or gasoline exhaust exposure, as well as some
evidence of a positive dose-response relationship, provides a basis of concern for these
associations.
McLaughlin et al. (1984) studied workers in the gasoline service and petroleum
refining industries and reported that a slight increase in renal cancer was associated with the
occupation of gas station attendant (OR = 1.2,95% CI 0.6 to 2.2). This risk increased
with the number of years of employment, but this trend was not statistically significant
(McLaughlin et al, 1985). The major limitation of this study was the lack of information on
gasoline exposure; this study provides insufficient data on the association of renal cell
carcinoma and gasoline.
The study conducted by Domiano et al. (1985) provided very limited data on the
association between gasoline exposure and renal cell carcinoma. The numbers of exposed
cases were very small (n = 3) and, while risks were elevated among the older smokers for
exposure to gasoline, they were not significantly elevated and may have occurred by
chance. There were no analyses of latency or length of employment. The study controlled
for smoking; however, the smoking category was very broad and was based only on the
smoking distribution of the cases, not on the control subjects.
Brandt et al. (1978) reported that 36 percent (18 of 50) of a group of male patients
with acute non-lymphocytic leukemia had been occupationally exposed to petroleum
products or their combustion residues in such occupations as filling station attendant, bus
or truck driver, or operator of excavating machines or power saws. The authors concluded
that occupational exposure to "motor fuel" containing 6 to 8 percent benzene was found to
be common among working males who developed this form of leukemia. Although this
study provides insufficient data upon which to base conclusions regarding the causal
10-13
-------
association of gasoline vapors and acute non-lymphocytic leukemia, it does suggest that
benzene exposure from motor fuels may be high enough to constitute a significant risk
factor for this cancer.
10.1.1.2 Cohort Studies
Milham (1983) conducted a proportionate mortality ratio (PMR) of white males in
Washington State and found that gas station owners and attendants had a significantly
(p<0.05) greater proportion of deaths due to cancer of the bladder and other urinary organs
than did non-occupationally exposed decedents (see Table 10-3). However, a significant
deficit was observed for cancer of the digestive organs and peritoneum among this same
study population.
Fuel oil and gasoline truck drivers were found to have a significant (p<0.05) excess
of all types of cancer. Fuel oil dealers, auto mechanics, and repairmen were also studied.
Fuel oil dealers were found to have a significant (p<0.05) excess for cancers of the buccal
cavity and pharynx, prostate, brain, and lymphatic and hematopoietic tissues; auto
mechanics were found to have a significant excess mortality for esophageal and respiratory
system cancers, including primary cancer of the bronchus, trachea, and lung.
Even though a significant cancer risk was found in this study, it is considered to be
inadequate for evaluating the gasoline exposure-cancer risk association because of study
deficiencies and limitation. The major limitation of the study is the use of the proportionate
mortality ratio as a method of analysis. The PMR compares the observed number of deaths
in a study group due to a specific cause to those expected based on the proportion of that
cause to all other causes of death in a comparative control population. Proportionate ratios
do not express the risk of members of a population contracting or dying from a disease,
only the proportion of deaths attributable to a specific disease in a given study setting. A
PMR analysis is generally used when the population at risk is not known, and serves to
identify whether or not the mortality distribution is similar between to population groups.
PMRs for two or more causes of death are interdependent rather than independent; the
magnitude of the PMR depends on the number of deaths from other causes besides the
disease under consideration. Hence, a deficit in one cause-specific mortality will inflate
another cause-specific mortality.
A PMR analysis also assumes that the classification and reporting of deaths
(compiling causes of deaths) in study and control populations are complete and
comparable. In the Milham (1983) study, incomplete compiling of data (job category
listing on death certificates) was reported and no attempt was made to control for
confounding factors (e.g., smoking) or confounding exposures (e.g., diesel fuel oil, motor
10-14
-------
TABLE 10-3
SUMMARY OF SITE-SPECIFIC STATISTICALLY SIGNIFICANT
CANCER FINDINGS FROM MILHAM'S 1983 WASHINGTON STATE
PMR STUDY, BY STUDY POPULATION EXPOSED*
Type
of
cancer
Gas station
owner/
attendants
Fuel oil/
gas Truck
drivers
Fuel oil/
dealer/
workers
Auto
mechanics/
repairman
All cancer
Buccal cavity/pharynx
Digestive organs/
peritoneum
Esophagus
Large intestine
Respiratory system
Trachea/bronchus/lung
Genitourinary organs
Prostate
Bladder/urinary organs
Brain
Lymphatic/hematopoietic
tissues:
Lymphatic leukemia
Other lymphomas
a Significant (p <0.05) excesses (+) or deficits (-) are reported.
SOURCE: Milham, 1983.
10-15
-------
oils, gasoline exhausts). As with many other studies, the author did not conduct latency
analyses or analyses by duration of exposure. The findings, however, are consistent with
the Silvennan (1983) study, and suggest that the elevated PMR ratios may be due to an
excess in the proportion of bladder cancers rather than a deficit in the proportion of causes
of death. The design of the PMR study, however, precludes any conclusions from being
drawn on the basis of this study itself.
A second PMR study conducted on male gasoline station attendants and automobile
mechanics in New Hampshire revealed a significant elevation in the incidence of leukemia
among service station attendants (PMR=406, n=4) and automobile mechanics (PMR=165,
n=7) even though deaths were few in number (Schwartz, 1986). However, as in the
Milham (1983) study, these data had many methodological constraints inherent in a PMR
study. In addition, the study cohort was potentially exposed to solvents, lubricating oils
and greases, asbestos, benzene, welding fumes, and car and truck exhaust, in addition to
gasoline vapor. Work history and exposure data were unavailable and latency, follow-up,
and confounding factors were not considered. Although this study is inadequate for
evaluating cancer risk in the gasoline service industry, its findings are consistent with the
possibility that benzene exposures from the gasoline service industry constitute a risk factor
for the induction of leukemia.
10.1.2 Epidemiological Studies of Workers in the Petroleum Refining
Industry and Other Petroleum-Based Industries
Twenty-seven studies were reviewed that evaluated the association between
employment in a petroleum refinery (a work environment with potential gasoline exposure)
and cancer risk. Judged individually, these studies provide inadequate evidence of an
association. However, judged collectively, these studies provide limited evidence of an
association between employment in a petroleum refinery and the risk of stomach cancer,
respiratory system cancer (i.e., lung, pleura, nasal cavity, and sinuses), skin cancer,
particularly melanoma, and cancer of the lymphatic and hematopoietic tissues, specifically
the leukemias, lymphosarcoma, non-Hodgkin's lymphoma, multiple myeloma, and
myelofibrosis. One of the major deficiencies of the epidemiological literature on the effects
of inhalation of gasoline vapors in the petroleum refining industry relates to exposure
assessment Although occupational exposure data bases are now being developed, it is
likely that they will have limited value in assessing long-term retrospective health trends.
Work practices have changed, permitting operation of the modern refinery from a
computerized remote panel; the potential for worker exposure has consequently been
reduced (Dynamac, 1986). Health and safety standards have been passed and
10-16
-------
implemented. In addition, technological developments of the automobile engine and
changes in gasoline composition (e.g., higher aromatic and isoparaffinic hydrocarbon
content in gasoline produced after 1940 and in unleaded gasoline) make an assessment of
exposure estimates for previous studies difficult Job description must serve as an
uncertain substitute for exposure. However, a majority of the refinery studies failed to
report any information on job titles or employee work categories. Furthermore, it is not
known how indicative of exposure a particular job title or work category might be. The
value of existing data nevertheless could be enhanced significantly through improved
exposure estimates for specific job classifications.
Refinery workers are exposed not only to gasoline but also to other petroleum and
petrochemical products manufactured within the refinery. It is not possible to assess the
effect of gasoline exposure independent of other refinery exposures; reviewed studies must
therefore be considered to be confounded by these multiple exposures.
An unknown amount of overlap exists between study populations, especially
among the refinery populations, as multiple studies were conducted at single refinery
locations at approximately the same time. Therefore, results reviewed may not always
represent independent events.
Statistically significant site-specific cancer findings from PMR, cohort, and nested
case-control studies that evaluated employment in the petroleum refinery are indicated in
Tables 10-4,10-5, and 10-6. The reader is referred to the Dynamac (1986) document for
an extensive listing of cohort mortality studies by study population exposed. The results
for all cancer sites combined are equivocal; three studies found statistically significant (p
<0.05) excesses of mortality (see Tables 10-4 and 10-5), four studies found significant
excesses among subcohort populations and significant deficits for all cancer mortality (see
Table 10-5), and nine studies found significant deficits for all cancer mortality (see Table
10-5). Cancer deficits may have been a result of the healthy worker effect (the health status
of industrial workers tends to be superior to that of the general population; therefore, a
confounding bias is produced if the control population for a study of industrial workers is a
cohort taken from the general population), the mixture of non-refinery personnel (e.g.,
clerical and administrative staff) in the study cohort, and the prohibited tobacco use in many
areas of the refinery. These confounding factors may have created a bias that
underestimated total cancer risk.
10.1.2.1 Case Control Studies
Nineteen case-control studies were reviewed that evaluated employment in the
petroleum industry as a cancer risk factor. Two studies provided limited evidence of an
10-17
-------
TABLE 10-4
SUMMARY OF SITE-SPECIFIC STATISTICALLY SIGNIFICANT CANCER FINDINGS
FROM PMR STUDIES THAT EVALUATED EMPLOYMENT IN THE PETROLEUM
REFINERY INDUSTRY AS A RISK FACTOR3
Type of cancer (ICD-8 Code)
Thomas et al.. 1980 Thomas et al.. 1982a Reeve et al.. 1982
o
oo
All cancers (140 to 209)
Digestive organs/peritoneum (ISO to 159)
Stomach (1 SI)
Pancreas (157)
Respiratory system (160 to 163)
Lung/ pleura/ other respiratory (162 to 163)
Skin (172 to 173)
Genitourinary organs (180 to 189)
Prostate (185)
Bladder (188)
Kidney (189)
Brain/other nervous system (191 to 192)
Lymphatic/ hematopoietic tissues (200 to 209)
Hodgkin's disease (201)
Non-Hodgkin's lymphoma (202)
Multiple myeloma (203)
Leukemia (204 to 207)
a Significant (p < 0.05) excesses (+) or deficits (•) are reported.
SOURCE: Dynamac, 1986.
-------
Table 10-5. Summary of Site-Specif Ic Statistically Significant * Cancar Findings Fran Cohort Mortality StudUs That Evaluated Employment
in the Petroleum Refinery Industry as a Risk Factor
Type of Cancer (ICD-8 Code)
All cancer (140-209)
Buccal cavity/pharynx (140-149)
Digestive system (150-159)
Esophagus (150)
Esophagus/stomach (150-151)
Stomach (151)
Other digestive system (152-159)
Intestines/rectum (152-154)
Intestines (152-151)
Large intestine (15))
Tabarshaw-Coopar Asaoc. (1975 Unpublishad) I
-
-
-
Hants (1977) I
+/-
+
+/-
at
"5
*
+
TharlauN IGoulat (1979) |
I
C.
"m
I
-
-
-
-
•
Wan at aL (1984*) I
-
Ruahton & Aldaraon (i 981 a) I
+/-
+
-r
+
Sehottanfalda1aL(l981) |
-
Hanbatal.(l982) |
-
-
Rushlon & Aldaraon (1983) 1
+/-
-
-
Morgan & Wong (1983 Unpub.) |
-
-
-
-
-
Morgan ft Wong (1984 Unpub.) 1
-
+/-
+/-
-
Morgan & Wong (1985 Unpub.) 1
-
Kaplan (1985 Unpublished) 1
-
-
MeGrawataL(1985) I
HanbataL(1985a) j
*/-
-
+/-
Divine at al. (1985) I
-
-
-
-
Nalson (1 985 Unpubltshad) 1
-
-
+
+
Entarllna & Handerson (1985) I
-
-------
(continued)
Type of Cancer (ICD-8 Code)
Large Intestlne/rectw (I5J-I54)
Rectui (154)
Other digestive (155-159)
Liver/gallbladder (155-156)
Pancreas (197)
Respiratory system (I60-I6U
Nasal cavity/sinuses (160)
Lung/pleura/other rasp. (162-16))
Trachea/bronchus/lung (162)
Masothelloma
i
c.
1
w Cooper
I
t-
-
£
»^.
1
-t-
+
+/-
a
C.
•i
?
J!
•f
+
*
P
O)
s
3
•5
§
1
I
•
I
-
-
L (1984*)
I
-
-
-J.
§
c.
1
•a
C
i
*
+
~
If
.j
5
c
!
-
c.
•i
Jl
1
§
c.
•a
c
i
+/-
^
c.
1
•0
1
-
-
-
-
jd
a
0
c.
I
•a
I
-
-
+
*
a
r,
I
•a
I
|
985 Unpub
C.
|
-
-
*
c.
*
•j
1
c.
•
•J
•
Jl
-
-
-
i
•
«
i
•
-
i
1
c.
i
-
-
c.
§
1
•0
i
-------
(continued)
Type of Cancer (ICD-8 Code)
Bona (1 70)
Skin (I72-I7J)
Malanoma (I72)
Can itouri nary organs (180- I 89)
Prostata (185)
Urinary organs (188-189)
Bladdor (IBB)
Brain/other nervous systam (191-192)
Lymphatic/haraatopoiatic tissuas (200-209)
Lyinpho- and ratlculosarcoma (200)
1
5
1
S2
r
I
i
a
«
1
1
1
1
4
1
1
1
i
-
1
c
-
al. (1979)
| Hani, at
P
•a
•§
+
8
O>
C.
'•
1
*
-
-
f
c.
m
1
-S-
c.
•a
i
+
+
5
ifaldatai(1l
]
^
3
g
C.
I
I
-
X
^
3
c.
!
+
-f
3
3
^
C.
1
•f
1
985 Unpublli
C
1
-
at aL (1965)
S
a
-
I
"5
Divlna at
•
-
e-
2
^
§
+
-
£
o>
•4
1
-------
(continued)
Type of Cancer (ICD-8 Code)
Lynphosarcoma (200.1)
Non-Modgkln's lymphold (202)
Multiple myeloma (20))
Leukemia/a leukemia (204-207)
Lymphatic leukemle (204)
Chronic lymphatic Uukcala (704.1)
Nyelofibrosis (209)
Unpublished) 1
S
C
k.
1
c
P
at
C,
m
Hants ct
f
c.
S
&
TherlauK
i
c.
•
c
1
1
i
c.
5
i
+
I
22
s
_c
2
Hants »t i
i
&
1
•B
i
+
-r
-r
•f
•f
^,
&
}
^.
*•
i
c.
I
•0
!
*
^
&
•a
1
f
i
r-
C.
1
-r
I
o»
&
Hants «t a
•
1
f
M
S
i
c.
c
fcHenderse
1
g
to
•p <0.05t •»" Indicates a positive finding (p <0.05) among one or nore cohorts or subcohorts; "—" Indicates e negative finding
(p <0.05) among one or wore cohorts or subcohorts; "»/-" Indicates a positive finding (p <0.05) among one or more cohorts or subcohorts
and a negative finding (p <0.05) among one or more cohorts or subcohorts.
bp . 0.05.
Significant excess of Morbidity rather than mortality.
Source! Oyn
(1966).
-------
TABLE 10-6
SUMMARY OF SITE-SPECIFIC STATISTICALLY SIGNIFICANT
CANCER FINDINGS FROM NESTED CASE-CONTROL STUDIES THAT
EVALUATED DIFFERENT OCCUPATIONAL CATEGORIES IN THE
PETROLEUM REFINERY INDUSTRY AS RISK FACTORS*
Type of cancer
Reference (p values)
Occupational exposure
Stomach
Brain benign
and unspecified
tumors
Thomas et al., 1984
(p <0.
Ban-on and Divine, 1985
(p <0.05)
Employment in the unit
operations of maintenance and
labor, employment in the yard
subcategory of maintenance;
lube oil workers.
Employment in Texaco
research laboratories.
a Excluded from this table are significant findings in which the authors reported that the
measure of central tendency (i.e., mean, median) of the length of employment was longer
for the controls than for the cases.
SOURCE: Dynamac, 1986.
10-23
-------
association between petroleum industry employment and the risk of cancer, the remaining
17 studies provided insufficient data upon which to base conclusions. These studies, like
the gasoline service and refinery studies, are limited by their lack of gasoline exposure
information. No quantitative estimates of exposure were available, and it was not possible
to determine if these workers were actually exposed to gasoline, although such a potential
existed. As in the gasoline and refinery studies, the results reviewed here may be
confounded by the other chemical exposures associated with employment in the petroleum
industry. Table 10-7 presents a summary of the site-specific statistically significant (p
<0.05) cancer findings for these studies.
Howe et al. (1980) reported that bladder cancer cases were more likely to be
employed in the petroleum industry than controls (OR=5.3,95% CI 1.5 to 28.6); this
estimate remained unchanged after controlling for the effects of age and smoking.
However, no latency or dose-response analyses were conducted, and confounding effects
other than smoking were not considered. This study provided limited evidence of an
association between employment in the petroleum industry and the risk of bladder cancer.
McLaughlin et al. (1984) found that renal cell carcinoma cases were more likely
than controls to be exposed to petroleum, tar, or pitch products, even after adjusting for
confounding effects of age and smoking (OR=1.7,95% CI 1.0 to 2.9, p+0.046). This
excess risk appeared to be positively associated with length of exposure (<20 years,
OR=1.1,95% CI 0.05 to 2.5; >20 years, OR=2.6, 95% CI 1.2 to 5.7). However,
McLaughlin et al. (1984) indicated that this significant association for exposure may
involve contact with polycyclic aromatic hydrocarbons (PAHs) as with, for example, coke
oven workers. While this exposure may be significant, it would have been helpful for the
study to have investigated the associations between the exposures and lung or skin cancers
(which have been identified as cancer risks associated with PAHs). A logistic regression
analysis was used to examine the effects of exposure while controlling for the effects of all
significant confounding variables. The logistic regression odds ratio of 1.6 for males was
similar to that of the stratified analysis (age and smoking adjusted OR=1.7); however, the
logistic estimate was not statistically significant. This was to be expected considering the
small number of exposed persons and large number of confounding variables
simultaneously controlled. The major limitation of this study was the lack of information
on gasoline exposure. This study provided limited evidence of an association between
exposure to petroleum, tar, or pitch products and renal cell carcinoma.
10-24
-------
TABLE 10-7
SUMMARY OF SITE-SPECIFIC STATISTICALLY SIGNIFICANT CANCER FINDINGS FROM
CASE-CONTROL STUDIES THAT EVALUATED EMPLOYMENT IN THE PETROLEUM
INDUSTRY AS A RISK FACTOR8
Type of cancer
Reference
Occupational (
Esophagus
Lung
Lung
Bladder
Bladder
Kidney
Renal cell
Leukemia
Acute non-lymphocytic
leukemia
Leukemia/
lymphogranulomaiosis
Gottlieb and Can. 1981b
Gottlieb et al.. 1979bb
Gottlieb, 1980b
Howeetal.. 1980b
Najem et al., 1982b
Gottlieb and Carr. 1981b
McLaughlin et al.. 1984b
Gottlieb and Can, 1981b
Brandt etal.,1978b
Plotnikov. 1978b
Occupations in the petroleum or
petrochemical industry.
Employment in the oil production
industry among those older than
62 years.
Employment in the petroleum
industry; petroleum industry oil-
field workers, welders, operators,
boilermakers, or painters; skilled
workers involved in petroleum
pumping and refining.
Employment in the petroleum
industry.
Employment in the petroleum (fuel)
industry.
Occupations in the petroleum or
petrochemical industry.
Petroleum, tar, or pitch; 20+- years'
exposure to petroleum, tar or pitch.
Occupational in the petroleum or
petrochemical industry.
Petroleum workers. Typical job
titles were filling station attendant.
bus or truck driver, operator of
excavating machines or power
saws, and road hauler.
Employment in the petroleum
industry.
a Significantly different from controls (p <0.05).
b As a result of study deficiencies, these studies provide insufficient evidence for an association between gasoline
exposure and increased cancer risk.
SOURCE: Dynamac. 1986.
10-25
-------
10.1.2.2 Cohort Studies
Hanis et al. (1982) found that a cohort of operators, mechanics, and laborers
showed an elevated standardized mortality ratio (SMR, the ratio of mortality rates,
expressed as a percentage, usually adjusted for age and/or time differences, between the
two groups being compared) for renal cancer, however, this elevation was not found to be
significant Specific gasoline exposure histories were not reported by the study authors,
the follow-up period was short, and the study cohort included personnel from a chemical
plant, confounding the data. Analyses by latency and length of employment were not
performed. No tracing procedure details or source of death certifications were reported.
This study is considered to be inadequate for use as a basis for any conclusions.
Rushton and Alderson (1983) found that when cases were indexed by age, kidney
cancer was more pronounced in an older population of oil distribution center workers;
however, this increase was nonsignificant (Enterline and Viren, 1985). Identification and
follow-up of the study cohort were well conducted; however, limited information was
provided for length of employment, latency, and job description. No information on
gasoline vapor exposure was available.
Enterline and Viren (1985) concluded that the available epidemiological studies,
when considered together, provided some evidence of a small increase in kidney cancer
among older or long-term workers in the petroleum industry. In a subsequent report,
Enterline (1984) used several epidemiological studies of petroleum workers in an attempt to
quantitatively assess the risk of kidney cancer as a result of exposure to gasoline vapors.
In this assessment, exposure levels, taken from data presented by Wen (1984), were
assigned to job categories and were assumed throughout entire employment periods. Five
studies of petroleum refinery workers that presented kidney cancer mortality data by
duration of employment were examined. Studies that showed a deficit in kidney cancer
mortality after extended employment were omitted. Estimated cumulative exposures were
derived from assigned exposures and periods of employment
This assessment has produced extensive criticism from peer review sponsored by
the petroleum industry. Reviewers have been surprised that risk estimates have been
derived for an association that has not been demonstrated to exist, much less demonstrated
to be causal (MacMahon, 1985; Wilson, 1985). Employees in the petroleum refining
industry have not been found to experience a statistically significant increased incidence of
kidney cancer (Cole, 1985; MacMahon, 1985; Wong, 1985); even in the Enterline review,
6 of the 10 studies had an SMR below 100 for kidney cancer, and the remaining four were
only marginally significant. Broad assumptions were made about exposure levels
involving tremendous uncertainties. The Wen (1984) data did not indicate how exposure
10-26
-------
measurements were taken, which job categories were related to measurements, time periods
involved, or whether measurements were representative of facilities surveyed. To apply
Wen's data to a complex variety of exposure situations (some without job tides) was to
introduce tremendous error and oversimplification. The risk estimates derived by Enterline
were based on an assigned average length of employment and were not directly comparable
to those used by U.S. EPA since the estimates resembled a proportion, not a probability
(Wong, 198S). In addition, cohort members examined for the assessment were not
mutually exclusive, and confounding factors (e.g., smoking) were not controlled.
Theriault and Goulet (1979) reported evidence of an association between malignant
neoplasms of the brain and CNS and less than 20 years of employment in the petroleum
industry. Even though the incidence was significantly elevated (SMR=652, n=3),
confounding variables and study deficiencies preclude use of these data. Sixteen percent
of the study population was eliminated from the study due to loss from follow-up (rather
than including these subjects in the study population until point of loss). As a result, the
SMR may have been potentially overestimated. Age- and cause-specific death rates from
1951 were used as the standard population for 1928 to 1962; this use is questionable, since
mortality rates have changed over this period of time. Confounding factors such as
smoking and additional chemical exposures were not controlled. Lastly, gasoline exposure
data were not presented.
Six studies found statistically significant (p <0.05) excesses of skin cancer
mortality among petroleum refinery workers (see Tables 10-4 and 10-5). Three of these
studies provided evidence for melanoma (Nelson, 1985; Reeve et al., 1982; Rushton and
Alderson, 198la). However, there is some overlap of study populations between the
Reeve and Nelson data. In addition, all studies have many limitations and deficiencies. All
studies lacked data on gasoline vapor exposures and did not control for mixed exposures.
Rushton and Anderson (1981) did not provide analyses by age, latency, or length of
employment In addition, the refineries studied were of varying size and may not have
dealt with the same products, and exposures were confounded by adjoining chemical
plants. Mortality results from different refineries indicated many contradictory findings.
Results of the Nelson (1985) study were not consistent among refineries, follow-up was
short, and there was no control of confounding factors. Study population data used by the
Reeve et al. (1982) study were incomplete; no information was reported on job tides or
length of employment. Analysis of latency was not performed, and confounding factors
were not controlled. Individually, these studies provide inadequate evidence of an
association between gasoline exposure and skin cancer, judged collectively, these studies
provide limited evidence.
10-27
-------
Morgan and Wong (1985) conducted a retrospective mortality study of 4,263
refinery workers employed between 1946 and 1979 and found a significant (p <0.05)
excess of death from prostatic cancer among white males with 20 or more years of
employment (SMR= 156.6,28 observed, 17.9 expected). The authors reported that
analyses based on the limited work history data available did not identify the agent or
condition associated with this excess. Job title descriptions were not available for this
study cohort, and individuals were coded by department only. Prostatic cancer was not
found to be significantly increased when the total cohort, including those with less than 20
years latency, was considered. Deaths from all causes combined were significantly lower
(16 to 27 percent) than expected as a result of the "healthy worker effect." Significantly
fewer deaths than expected were also found for nonmalignant diseases of the nervous,
circulatory, respiratory, and digestive systems. Fewer than expected deaths were found for
all cancers combined and for individual cancers of the buccal cavity, liver, skin, bladder,
and kidney. Mortality from brain, rectal, and pancreatic cancers and leukemia was similar
to that expected. Several limitations were inherent in this study. Data on actual gasoline
vapor exposures were not available, exposures were mixed, job title descriptions were
unavailable, unexposed individuals may have been included in the study cohort, diagnostic
conditions were determined from death certificates, and confounding factors such as the
"healthy worker effect" and smoking were not controlled. However, the study authors
recognized many of these limitations and indicated that further study was necessary to
better understand the excess of prostatic cancer.
Various studies suggested an association between prostatic cancer and exposure to
lubricating and dewaxing solvents in the refinery workplace. In 1978, Enterline reported
two deaths (0.08 expected incidence) due to prostatic cancer among a cohort of 305
refinery workers, and in 1984 Wen et al. reported seven deaths (2.94 expected incidence)
due to prostatic cancer among a cohort of 295 refinery workers (Morgan and Wong, 1985).
However, the Enterline study was based on only two deaths and in the Wen study, an
excess of prostatic cancer was found among nonwhite employees only.
Enterline and Henderson (1985) conducted a retrospective mortality study of 1,621
refinery workers employed between 1959 and 1980. Deaths from all causes combined
were significantly lower (23.5 percent) than expected as a result of the "healthy worker
effect." There were no significant excesses observed for any cause of death, including
some cancers previously associated with refinery employment. However, analyses of
death by time since first employment identified a positive trend for prostatic cancer and
hypertension without heart disease. These data may have been related to work exposure at
a previous refinery facility. When workers who had been transferred from this refinery in
10-28
-------
1935 were studied apart from those without this employment history, an excess in
hypertension and cancers of the prostate, central nervous system, and lymphatic and
hematopoietic tissues were found. The study authors considered these analyses to be
unrelated to the primary study. Due to the small cohort size, this study may have been
unable to detect small or moderately sized excesses in specific causes of cancer. In
addition, data on gasoline vapor exposures were not reported, exposures were mixed,
complete records were not available, and the "healthy worker effect" may have biased the
data.
Data from studies on the association between gasoline vapor exposure and cancers
of the digestive system, respiratory system, bladder, and lymphatic and hematopoietic
tissues were limited and ambiguous (Dynamac, 1986). Serious limitations existed for all
refinery studies. Analyses of latency were not conducted, dose-response effects were not
evaluated, confounding factors (e.g., smoking, alcohol consumption, diet) were not
controlled, non-exposed employees were included in the study cohorts, and the length of
cohort follow-up was short. In addition, the number of persons lost to follow-up was
sufficiently large for some of the studies to call into question the author's findings. Other
limitations included a lack of detail on cohort tracing procedures and cause of death coding
procedures employed by some of the researchers. Furthermore, methods employed for
vital status ascertainment were questionable. Data on ethnicity, pertinent in refinery
analyses, were not available for all studies. Finally, the cohort mortality studies are limited
by their inconsistency of results, some of which may be explained by geographic variation
of changes in cancer mortality patterns over time. Some also may be due to differences in
refinery practices or exposure definition.
Despite these study deficiencies, there is limited evidence of an association between
employment in a petroleum refinery and the risk of stomach cancer, respiratory system
cancer (i.e., lung, pleura, nasal cavity, and sinuses), skin cancer, particularly melanoma,
and cancer of the lymphatic and hematopoietic tissues, specifically the leukemias,
lymphosarcoma, non-Hodgkin's lymphoma, multiple myeloma, and myelofibrosis.
In summary, the study conducted by Stemhagen et al. (1983) provided limited
evidence for an association between gasoline service station employment and the risk of
primary liver cancer. The other nine studies, while suggestive, are limited in their abilities
to draw definitive conclusions regarding the potential human carcinogenicity of gasoline.
As mentioned previously, a major limitation of all these studies was the lack of information
on gasoline vapor exposure. In addition, due to the multiplicity of types of occupational
and non-occupational gasoline, oil, and hydrocarbon exposures in these studies, it was
difficult to implicate a single agent or class of compounds. Employment in the gasoline
10-29
-------
industry was assessed as the cause of cancer incidence rather than gasoline vapor exposure.
In general, the studies did not adjust for confounding variables or conduct latency analyses.
10.1.3 Summary of Epidemiological Studies in Progress
The American Petroleum Institute has recently begun three epidemiological studies
that are directly relevant to this review: (1) a historical prospective mortality study of
employees exposed to downstream gasoline in the petroleum industry; (2) a case-control
study of kidney cancer and hydrocarbon exposure among petroleum company workers; and
(3) a prospective study of the mortality of refinery workers. Findings from the
downstream gasoline study will be available in 4 years; case-control study results will be
available in 2 to 3 years. At least 5 years will be required before the prospective study has
sufficient power to detect an approximate doubling of risk for kidney cancer or stomach
cancer.
These studies have methodological limitations. Smoking histories will not be
available for cohort members in the downstream gasoline mortality study. In addition, it
appears than an internal non-exposed group cannot be used as the comparison population
(Wong and Morgan, 1984, unpublished). Thus, these data may be biased by the healthy
worker effect as well as by smoking.
The case-control kidney cancer study only identifies cases of kidney cancer that
resulted in a mortality; it does not identify incidence of cases. As noted by the authors of
this feasibility study, "mortality data might reveal as little as SO percent of the number of
incident cases in some age groups" (ERI, 1985, unpublished). Thus, a bias from
incomplete ascertainment of the cases might exist in this study. Data on the smoking habits
of individual study participants will not be available, although several studies have found a
dose-response relationship between smoking and kidney cancer (U.S. Public Health
Service, 1982; McLaughlin et al., 1984). Therefore, these data may be biased by the
confounding effects of smoking.
A prospective cohort study is being conducted by the Australian Institute of
Petroleum Ltd., in conjunction with the University of Melbourne, Department of
Community Medicine (AIP-UM/DCM, 1983; AIP-UM/DCM, 1985; Christie et al., 1984).
This well-designed study will, at some future time, permit the calculation of risk estimates
while simultaneously controlling for important confounding variables (e.g., smoking,
alcohol). Quantitative estimates of exposure will also be available. However, these results
are several years away.
10-30
-------
A Shell Oil Company study of their Deer Park refinery is also known to be in
progress or to have been very recently completed. Additional details on this study are not
available.
10.1.4 Benzene Carcinogenicitv
Benzene is considered to be a human leukomogen (IARC, 1982; FPC/API, 1986;
NRC, 1986a). Acute myelogenous leukemia is the most common form of adult leukemia
and the form of this disease that has been most frequently related to benzene exposures
(Aksoy, 1980; Aksoy et al., 1976; FPC/API, 1986; Rinsky et al., 1981; Vigliani and Saita,
1964). The variant of acute myelogenous leukemia most frequently reported in association
with benzene is erythroleukemia (Aksoy, 1980; FPC/API, 1986). Acute myelomonocytic
leukemia and acute promylocytic leukemia have been associated with benzene exposures to
a lesser extent, as have multiple myeloma and Hodgkin's disease (Aksoy, 1980; Aksoy et
al., 1974; Bond et al., 1986b; RPC/API, 1986; Rinsky etal., 1987; Decoufle et al., 1983).
Inhalation is the major human exposure route for benzene. In 1972, Aksoy et al.
reported four cases of acute leukemia attributed to chronic benzene inhalation exposures to
atmospheric concentrations of 150 to 200 ppm. Two of the four workers developed
leukemia subsequent to recovery from aplastic anemia (Aksoy, 1972b). Later, Aksoy et al.
(1974) reported 26 cases of acute leukemia or preleukemia among 28,500 workers
chronically exposed to benzene vapors for 1 to 15 years; the incidence of hematological
anomalies was significantly increased (p <0.01) over that of the general population.
Leukemia was found to be temporally related to the onset of benzene use and diminished
following replacement of benzene as a solvent (U.S. EPA, 1978a). The interval between
the onset of the preceding pancytopenic period in the majority of these cases and that of
acute leukemia varied between 6 months and 6 years (Aksoy et al., 1980). Acute
myeloblastic leukemia is the most frequently observed type of leukemia in benzene-exposed
workers followed by erythroleukemia; unexposed workers exhibit chronic myelogenic
leukemia, which is not observed among exposed workers (Aksoy, 1980).
Study deficiencies and methodological limitations restrict interpretation and
usefulness of these data. Inconsistencies in job description, lack of follow-up, lack of age
standardization, and incidence comparison for unknown locations and periods of time were
prevalent in the Aksoy studies (ODW, 1985). In addition, exposures were mixed, and the
amount and duration of exposures were non-specifically estimated.
Vigliani and Saita (1964) and Vigliani and Fomi (1976) reported 11 cases of acute
myeloblastic leukemia and an additional 37 cases of benzene-induced blood dyscrasias in
3,000 leather workers exposed to 25 to 600 ppm benzene. On follow-up, 13 deaths were
10-31
-------
attributed to acute leukemia and 3 to aplastic anemia (Vigliani, 1976; Vigliani and Forni,
1976).
Infante et al., (1977a, b, 1978) reported seven deaths from leukemia (five-fold
increase), of which four were acute myelogenous leukemia (ten-fold increase), in a
retrospective cohort study of 748 rubber plant workers. The expected mortality rate based
on a control cohort of U.S. white males was 1.4; leukemia mortality was significant (p
<0.002). The period between the initial inhalation exposure and death was 2 to 21 years.
Atmospheric exposure concentrations were not precise and were estimated to be below 15
ppm benzene; however, subsequent exposure reports found atmospheric levels to range
from 100 to 1,000 ppm (Tabershaw and Lamm, 1977; Van Raalte and Grasso, 1982).
Rinsky et al. (1981) conducted a 98 percent follow-up of the Infante study cohort
Seven deaths were due to leukemia among 748 workers (SMR=560; p <0.01), and 58
percent of the cohort had been exposed to benzene for less than 1 year. In benzene
workers exposed for 5 years or more, leukemia deaths produced an SMR of 2,100 (p
<0.001). Even though this study has been criticized for a lack of exposure data, multiple
exposures to other chemicals, and flaws in cohort eligibility and completeness, the data do
indicate a decline in the incidence of leukemia that parallels the decline in benzene
exposures at the study locations (Wong et al., 1983; FPC/API, 1986; Snyder et al., 1984;
Van Raalte and Grasso, 1982; Tabershaw and Lamm, 1977).
In 1987, Rinsky et al. re-examined this study cohort after an additional 5 years of
follow-up. Study conclusions were based on imprecise group estimates of exposure.
Total mortality and mortality from all malignant neoplasms combined were not elevated
over expectation; however, a marked increase was found in the number of deaths from
leukemia (9/1,165 observed vs. 2.7 expected) and multiple myeloma (4/1,165 observed vs.
1 expected). The incidence of benzene-related cases was found to increase with increasing
exposure. Latency periods ranged from 5 to 30 years (Rinsky et al., 1987). The incidence
of multiple myeloma was found to occur in individuals with the lowest estimated
exposures; latency periods were found to be greater than 20 years. Rinsky et al. (1987)
hypothesized that whereas the higher benzene exposures produced leukemia, low
cumulative exposures to benzene may produce a malignancy such as multiple myeloma.
The incidence of multiple myeloma among workers occupationally exposed to benzene has
also been reported by Aksoy (1980), Bond et al. (1986b), and Decoufle et al. (1983). The
Rinsky et al. (1987) study will be used for human risk assessment evaluation.
On et al. (1978) studied mortality among 594 workers of a chemical plant who had
been occupationally exposed to benzene. Two deaths attributed to acute myelocytic
leukemia were reported (p <0.047). However, only limited sampling and exposure data
10-32
-------
were reported; the atmospheric concentration of benzene was found to range from 0 to 937
ppm (Van Raalte and Grasso, 1982). The questionable accuracy of the leukemia
diagnoses, multiple chemical exposures, varied cohort work histories, and the limited size
of the study population, as well as the lack of clear dose-response relationship, limit the
usefulness of these data (U.S. EPA, 1987; Fielder, 1982; Van Raalte and Grasso, 1982).
Bond et al. (1986b) conducted a follow-up of the Ott study cohort consisting of 9
additional years and including 362 additional exposed workers. The excess of total
leukemia was found to be nonsignificant; however, that of myelogenous leukemia was
significant (p <0.011). One death from multiple myeloma and four deaths from malignant
melanoma and squamous cell carcinoma (Bond et al., 1986b) were reported. The incidence
of skin cancer was considered to be an isolated occurrence because it had not previously
been reported (Bond et al., 1986). Unquantified exposure data, multiple chemical
exposures, and the limited size of the study population were recognized deficiencies of this
study.
Wong et al. (1983) conducted a study of 8,000 male workers from six chemical
companies (NRG, 1986a). It has not been possible to obtain the original published data;
however, reviews indicate that a statistically significant dose-response relationship existed
between cumulative benzene exposures and leukemia mortality (NRG, 1986). It has been
suggested that the seven leukemia deaths observed in the study do not actually represent an
excess, since a deficit of deaths occurred in the control group artificially inflating the
relative risk (Federal Register, 1987). However, the control group was composed of
groups of workers at the same chemical companies, differing only in not being exposed to
benzene. Therefore, the potential healthy worker effect bias was eliminated.
In addition, Wong et al. concluded that there was a significant association between
occupational exposure to benzene and Ivmphoma and non-Hodgkin's lymphoma (Federal
Register, 1987). As in previous studies, data on duration and extent of benzene exposure
were deficient, and multiple chemical exposures existed (NRC, 1986a).
Thorpe (1974, as cited in ODW, 1985) evaluated leukemia mortality among
petroleum workers who had potential exposure to benzene as a constituent of petroleum
products. Eighteen cases of leukemia were observed over a 10-year period (deaths
expected = 23.2). Uncertain exposure levels, mixed exposures, and questionable leukemia
diagnoses prevent use of these data.
These aforementioned studies are insufficient to provide a clear quantitative
assessment of the relationship between benzene and the risk of developing leukemia. There
is some controversy regarding the minimum level of benzene that may produce adverse
10-33
-------
effects among occupationally exposed workers (FPC/API, 1986; Ott et al., 1978; Bond et
al., 1986b; NRC, 1986a).
10.1.5 Summary of Human Carcinogenicitv Studies
Fifty-six studies were identified that investigated the cancer risks among workers
employed in the petroleum industry. In addition, the available evidence was reviewed
concerning the associations between benzene exposure and carcinogenicity. Benzene is the
only single component of gasoline thus far identified as being causally associated with
cancer in humans or laboratory animals.
Based on an assessment of the available literature, gasoline is presumed to be
carcinogenic to human beings. This finding is based largely on the fact that benzene, a
volatile component of gasoline, is an established human carcinogen. Any exposure to
gasoline would also involve exposure to benzene. In addition, limited animal evidence is
now available that toluene and xylene, which are also gasoline components, are
carcinogenic in rodents. Finally, evidence from epidemiological studies on gasoline
suggests that exposure to this hydrocarbon mixture may itself be carcinogenic to humans,
irrespective of benzene's carcinogenicity.
The epidemiological evidence regarding benzene carcinogenicity is widely accepted.
The evidence regarding the human carcinogenicity of the gasoline mixture, however, is
subject to significant uncertainties. These uncertainties stem from the limited sensitivity of
epidemiological studies to identify carcinogens (particularly weak carcinogens), the
complexity of the chemical exposures associated with the petroleum industry, and the lack
of any clearly defined target organ which gasoline may specifically affect. Moreover, the
fact that many epidemiological investigations have been conducted, each one analyzing
several types of cancer, raises the concern that apparently significant associations may
actually have been random statistical fluctuations. Nonetheless, a more qualitative analysis
of the study findings suggest that clearer positive associations could be established upon
more rigorous analyses. Such associations involve the following sites.
Genito-urinarv Tract Tumors
Four case-control studies (Mommsen et al, 1982; Mommsen and Sell, 1983;
Mommsen and Aagard, 1983a, b, 1984; Silverman et al., 1983) and one cohort study
(Milham, 1983) on gasoline service station workers found increases in bladder cancer
associated with employment. Silverman et al. (1983) also found increased risks associated
with duration of exposure and, possibly, with occupations involving increased exposure to
petroleum. Two case-control studies on petroleum industry workers (Howe et al., 1980;
10-34
-------
Najem et al, 1982) also found significant elevations in bladder cancer associated with their
occupations.
A population based cohort study found a slight increase between occupational
exposure at gasoline service stations and kidney cancer, which increased with employment
duration (McLaughlin, 1985). Neither the increase nor the trend, however, were
statistically significant The same study on petroleum industry workers (McLaughlin,
1985) indicated a statistically significant increase in renal cancer risk associated with
occupational exposure to petroleum, tar, or pitch products (OR = 1.7,1.0-2.9). The risk
increased in significance with duration of employment. An analysis of cohort studies on
petroleum industry workers (Enterline and Viren, 1985) found that a small increase in
kidney cancer may occur among the older or long-term workers. A case-control study on
gasoline service station workers also suggested an increase in renal cancer associated with
occupational exposure among the older workers (Domiano, 1983).
Three cohort studies on petroleum industry workers and one cohort study on fuel
oil dealers found elevations in prostate cancer associated with occupational exposure. One
case-control study found elevated testicular cancer cases among gasoline service station
workers (Lin and Kessler, 1979).
The concern raised by these studies is especially significant in light of the findings
from other epidemiological and lexicological studies which identify the kidney as a target
organ for gasoline toxicity, and from the cancer bioassays which found significant
increases in kidney tumors among male rats exposed to wholly vaporized gasoline.
Liver Tumors
One case-control study (Stemhagen et al., 1983) on gasoline service station
workers found a statistically significant elevation (OR = 2.88) in primary liver cancer
cases. When the analyses were limited to hepatocellular carcinomas, the elevation was
even higher (OR = 4.2). This finding is significant in light of bioassay results that found
an increase in hepatocellular carcinomas in female mice exposure to wholly vaporized
gasoline.
Pulmonary Tumors
Cohort studies on auto mechanics (Milham, 1983) and petroleum industry workers
(Morgan and Wong, 1983; Rushton and Alderson, 1983; Thomas et al., 1980,1982a;
Hanis, 1977; Hanis et al., 1979) provide ambiguous and inconclusive results regarding the
possible association between gasoline exposure and lung cancer. Two case-control studies
on petroleum industry workers (Gottlieb et al., 1979; Gottlieb, 1980) suggest that
10-35
-------
occupational exposure to petroleum is associated with elevated incidences of respiratory
cancers. These case-control studies are severely limited, however, by the fact that they did
not control for cigarette smoking.
Hematopoietic Tumors
One cohort study on gasoline station attendants (Schwartz, 1986), three case-
control studies on petroleum industry workers (Gottlieb and Carr, 1981; Brandt et al.,
1988; Plotnikov, 1978), and six cohort studies on petroleum industry workers (Thomas et
al., 1982; Reeve et al., 1982; Rushton and Alderson, 1981a; Morgan and Wong, 1984;
Rushton and Alderson, 1983; McGraw et al., 1985) found significant associations between
occupational exposure and hematopoietic cancer. These findings are important in light of
the fact that benzene is toxic to blood cells and is a human leukemogen.
Skin
Several cohort studies on petroleum industry workers found associations between
occupational exposure and elevated incidence of skin cancer. These findings are important
in light of the fact that gasoline is toxic to the skin and that individuals may receive a
significant direct exposure to gasoline from use of gasoline-contaminated water supplies.
One problem with the analyses of many epidemiological studies is that comparisons
among studies are very difficult to make. It would require a much more detailed analysis of
the studies, for example, to better understand the potential reasons for different findings.
Differences in design and conduct of the studies could account for different conclusions,
even if the morbidity characteristics of the studies populations were the same. Also, it is
widely recognized that even the best designed and conducted epidemiological studies are
probably unable to acquire all the necessary information for a complete analysis. From a
public health perspective, therefore, a crucial problem in the interpretation of these studies
is whether and to what extent more information is likely to overturn a positive association
between exposure and effect. Such analysis is beyond the scope of this assessment The
probability that deficiencies in a particular study produced a false positive result is reduced,
however, if other studies by other investigators produced similar results. Such
considerations appear particularly applicable in this case. While not conclusive by
themselves, therefore, the findings of these cancer epidemiology studies support the
presumption that exposure to gasoline could be carcinogenic to humans.
10-36
-------
10.1.6 Summary
Assessment of the carcinogenic potential of gasoline and gasoline components rests
on both epidemiology and animal bioassay studies. For epidemiology, insufficient time
has elapsed to evaluate the risk from exposure to unleaded gasoline. Thus the studies on
gasoline workers per se were used. Even in these studies, exposure levels and components
are not well described. Overall, the available human studies suggest that there is an
increase in certain tumor types in occupationally-exposed individuals, while in others there
is a decrease in cancer relative to the overall population. A possible healthy worker effect
may exist Therefore, it appears that there is limited evidence for carcinogenicity at certain
sites from exposure to gasoline.
Tumor sites frequently of concern with regard to benzene exposure are bladder,
kidney, liver, lung, blood forming tissue, and skin. For benzene, the association with
myelomonocytic leukemia and other tumors is supported by the epidemiology studies.
Although the studies all have significant limitations, there is sufficient information for a
quantitative estimate of cancer risk.
10-37
-------
10.2 ANIMAL CARCTNOflENICITY
This section describes the animal carcinogenicity studies of gasoline and three major
components: benzene, toluene, and xylene. The agents are discussed separately because
the studies on the mixture of gasoline are limited to one bioassay on a test sample of
unleaded gasoline (MacFarland et al.. 1984) and the product as sold is highly variable with
respect to hydrocarbon components. Moreover, considerable attention has been given to
the outcome of the MacFarland study in an attempt to show that the carcinogenic
observations in rats and mice are not valid for human risk assessment This proposal has
been extensively reviewed elsewhere and will not be discussed in detail in this chapter.
The following presentation includes an overview of the the problem relative to the animal
carcinogenicity evaluation and a discussion of the relevant carcinogenic studies for gasoline
and the three major components.
10.2.1 Introduction
There are two basic approaches to estimating of the carcinogenic effects of a
mixture such as unleaded gasoline; one measures the effects of the mixture in a long term
bioassay and the other measures the carcinogenic potential of each component of the
mixture. While there are separate carcinogenic studies for the aromatic components of
greatest interest in this review (i.e., benzene, toluene, and xylene), only one long-term
bioassay, sponsored by the American Petroleum Institute, is available for a sample of
unleaded gasoline (MacFarland et al., 1984).
The API study has limitations in terms of dosing protocol and the presence of an
unexpected renal sensitivity in the male rats and hepatic sensitivity of the female mice.
U.S. EPA concluded that the study does provide evidence of carcinogenicity in that it
satisfied the criteria of both LARC and the U.S. EPA (U.S. EPA, 1987d). A skin painting
study did not demonstrate carcinogenic responses.
Some reports have been published to test the proposal that the renal tumors are the
result of an accumulation of unusual urinary protein formed in the liver of male rats which
either initiates the tumorigenic response or may predispose the kidney to the carcinogenic
responses (HEI, 1985,1988). The argument continues that because this protein has not
been shown to be present in humans, gasoline can clearly not be considered carcinogenic in
man. Interestingly, as Table 10-8 indicates, the male rat also distributes the hydrocarbons
selectively (7 times higher) to the kidney compared to other species. This suggests that
differences in distribution in the rat may account for the higher sensitivity. Another
suggestion for the action of the protein is based on the argument that binding of a
component increases the exposure duration in the rat While this supports a quantitative
10-38
-------
TABLE 10-8
DISTRIBUTION OF 2,2,4-TRIMETHYLPENTANE IN
FISCHER 344 RATS
Tissue
Kidney
Fat
Liver
Lung
Heart
Testes
Spleen
Blood
nMol equivalents/
Males
1,1 15*
224
177
23
21
16
10
17
g wet tissue
Females
157
336
193
23
12
~
3
17
a Significantly different from females.
SOURCE: Adapted from Kloss et al., 1984.
10-39
-------
difference in sensitivity, it does not support the suggestion that the positive carcinogenic
response should be ignored.
There are several studies that test the hypothesis that tumors are secondary to renal
toxicity. The proposal is relatively attractive on biochemical grounds and data consistent
with this hypothesis has been developed by several researchers. Although the evidence is
consistent with the accumulation of a specific protein found only in the male rat, occasional
renal tumors are also observed in both female rats and mice.
Another proposal is that the carcinogenic response is due to exposure to a less
volatile fraction of gasoline that will not be released to the air during normal distribution
and use of gasoline. However, the less volatile fraction is present in gasoline contaminated
drinking water.
The primary question is whether studies on the mechanism of nephropathy in male
rats from hydrocarbon exposure should be considered to rule out carcinogenicity. The only
convincing argument as to the non-carcinogenicity of gasoline vapors would be a rat and
mouse bioassay that did not show an increase in tumors in the treated groups. Therefore,
because these data are not available, findings on rat renal tumors are presented in this
chapter with notation of the limitations of the study.
10.2.2 Lifetime Inhalation Bioassav in Rats and Mice
A chronic inhalation study was conducted in rats and mice with typical gasoline
used in the U.S. containing 2 percent benzene (MacFarland et al., 1984). Groups of 100
animals of each species and sex were assigned to each chamber, giving 400 animals per
chamber. Ten randomly selected animals of each species and sex were sacrificed at 3,6,
12, and 18 months. Fischer 344 rats and B6C3F1 mice were exposed 6 hours per day, 5
days per week for 103 to 113 weeks to average vapor concentrations of 67,292, and 2,056
ppm. A significant decrease in body weight gain at the highest exposure level in both sexes
of both species shows that a maximum tolerated dose was exceeded in the study.
Gasoline was delivered from a liquid metering pump to a heated countercurrent
vaporization column which completely volatilized the sample. The specifications of the
unleaded gasoline samples are shown in Table 10-9. Standard measures of body weight,
effects, and biochemical evaluations were performed at appropriate intervals. Gross and
microscopic evaluations were performed on all animals that died during the study. A 40
percent survivability was used to determine termination dates for each group.
Evidence of nephrotoxicity and tumorigenesis was found in the renal tubules of
male rats. Relative kidney weights were altered in the intermediate and high-dose level in
male rats and the high-dose level in female rats. Testes, ovaries, and heart weights were
10-40
-------
TABLE 10-9
CHRONIC GASOLINE INHALATION STUDY
SPECIFICATIONS OF UNLEADED MOTOR GASOLINE
Research octane No.
Motor octane No.
(R + M)/2
Reid vapor pressure, pounds
Distillation, ASTM D-86
IBP
5
10
20
30
40
50
60
70
80
90
95
EP
Recovery
10% evap., °F
50% evap., °F
90% evap., °F
API gravity
Gum, ASTM D381, mg/gal
Sulfur, ppm
Phosphorus, g/gal
Lead, g/gal
Stability, hours
HC analysis, ASTM D1319
Aromatics
Olefms
Saturated
Benzene content
Sample
Used in
Study
92.0
84.1
88.1
9.5
93
105
116
138
164
190
216
238
256
294
340
388
428
97%
112
211
331
60.6
1
97
<0.005
<0.05
24+
26.1vol%
8.4 vol %
65.5 vol %
2.0%
Unleaded
Commercial
Average*
92.1
83.6
87.9
9.9
92
124
220
332
412
59.3
1
27%
7%
66%
1.0%b
a DuPont Road Octane Survey, summer 1976.
b Average benzene content typical of US gasolines.
SOURCE: MacFarland et al., 1984.
10-41
-------
also altered at the high dose level. No changes were reported in mice at any dose in these
organs. In addition, no significant clinical chemistry changes were reported. The kidney
tumor incidence in male rats from unleaded gasoline exposure is presented in Table 10-10.
In addition, one female in the mid dose group had a primary renal tumor (sarcoma).
Pathological changes in mice were reported as not dose related for both neoplastic
and non-neoplastic effects except for an elevated incidence of hepatocellular tumors in the
high dose females mice. The increase in hepatocellular tumors was observed at the high
dose (48 vs. 14 percent in the controls) in female mice that died during the 18 to 24 month
period or at the time of terminal sacrifice (see Table 10-11) was statistically significant.
The liver tumors consisted of hepatocellular adenomas and hepatocellular carcinomas. As
illustrated in Table 10-11, significant (p<0.05) increases in both adenomas and carcinomas
occurred. Several of the hepatocellular carcinomas metastisized to the lung and kidneys.
Renal tumors were also seen in two of the high-dose females.
The authors concluded that the dose-related increased incidence of renal tumors in
male rats could be attributed to exposure to gasoline vapors. Based only on the issue of
liver tumors in mice, it was their opinion that the association of an increase in liver tumors
in high-dose female mice with gasoline vapor exposure was equivocal.
10.2.2.1 Uncertainties in the Carcinogenicitv Studies
There are two major uncertainties in the extrapolation of these results to predict
human risk: (1) the vapor composition in the API study is different than that in the ambient
human environment; and (2) the kidney tumors observed in male rats may result from an
unique mechanism specific to male rats and not to female rats or other species.
10.2.2.2 Fuel Blend Issue
In the API study, a fuel blend, termed PS-6, which was considered representative
of U.S. commercially marketed motor vehicle fuel, was employed. The benzene level was
approximately 2 percent No lead, ethylene dibromide, or ethylene dichloride was added.
In comparison to commercial unleaded gasoline, the test gasoline contained a higher
proportion of benzene and heavy catalytic-cracked naphtha. The major concern is that the
gasoline was entirely vaporized for the animal exposures so that the inhaled mixture was
similar to the liquid gasoline. In usual human exposure situations, the gasoline vapors are
dependent on the differential volatility of the hydrocarbons present in gasoline. The smaller
hydrocarbons and aromatics are more volatile and are present in greater proportions in the
human's ambient environment than in the vapor generated for the animal study. However,
these would be present in contaminated drinking water.
10-42
-------
TABLE 10-10
KIDNEY TUMOR INCIDENCE IN MALE FISCHER 344 RATS FROM
CHRONIC EXPOSURE TO VAPORIZED UNLEADED GASOLINE IN THE
MACFARLAND ET AL. STUDY
Exposure Group (ppm Gasoline Vapor) Time to first
Tumor Type 0 (Control) 67 292 2056 Tumor (Days)
Renal adenoma/ 0/100a 0/100 2/100 1/100 546
adenoma cortex
Renal carcinoma/ 0/100b 1/100° 2/100 o/lOOC-6 692
renal carcinoma
^differentiated
Renal sarcoma'
TOTAL
0/100
0/100b
0/100
1/100
1/100 0/100 748
SnOO^e 7/lOOC-e
a Number with tumors/number examined.
b Statistical analysis for linear trend was significant at the 0.05 level.
c In the MacFarland (1984) Interim Report, renal carcinomas were diagnosed in two low-
dose and five high-dose male rats. Further analysis of the kidney sections, as indicated
in the subsequent Final Report on the study, resulted in the observation of renal
carcinomas in one low-dose and six high-dose male rats.
d The 100 animals in each denominator in this table include 40 animals sacrificed at 3, 6,
12, and 18 months, and decedents and survivors in the remaining 60 animals which were
allowed to survive for the duration of the study. If the 40 interim sacrificed animals are
excluded from each denominator to allow replacement of the 100 total animals with the 60
animals allowed to survive for the duration of the study, the statistically significant
differences shown in this table remain significant at P < 0.05.
e Statistically significant (P < 0.05) increase compared to control group by Fisher's Exact
Test.
f A renal sarcoma was also found in a female rat in the 292-ppm exposure group.
SOURCE: IRDC (1982) as reported in U.S. EPA's staff paper draft of June 1984.
10-43
-------
TABLE 10-11
INCIDENCE OF LIVER TUMORS IN B6C3F1 MICE
EXPOSED TO VAPORIZED GASOLINE
Males
Controls
Low Dose
Mid Dose
High Dose
Females
Controls
Low Dose
Mid Dose
High Dose
Number
Examined
51
42
44
54
57
52
57
56
Hepatocellular
Adenomas
12
4
5
5
1
4
4
8*
Total
Hepatocellular Liver
Carcinomas Tumors
11
14
16
20
7
6
9
20*
23
18
21
25
8
10
13
28*
Liver Tumor
Bearing
Animals3
23 (45%)
15 (36%)
20 (45%)
24 (44%)
8 (14%)
10(19%)
12 (21%)
27 (48%)*
a Some animals had more than one tumor.
* p <0.05.
SOURCE: Adapted from Westpath Laboratories, Inc., 1981.
10-44
-------
10.2.2.3 Renal Toxicitv Issue
A critical evaluation of the kidney effects of unleaded gasoline observed in the
MacFarland study was conducted by Westpath (1980). In exposed rats, there was a
spectrum of histomorphological renal alterations including tubular cell basophilia,
proteinaceous cast formation, chronic interstitial inflammation, renal pelvic mineralization,
nuclear alterations, tubular cell hyperplasia, and primary renal neoplasia. All of these
lesions, except tubular cell hyperplasia and renal neoplasia, were observed prior to 12
months. With one exception, all primary renal neoplasms occurred at or near the final
sacrifice period, which was slightly in excess of 26 months. Both control and treated mice
developed congenital lesions such as hydronephrosis and cortical cysts, and spontaneous
lesions associated with aging. These lesions were found in equal numbers in both control
and treated groups.
Kitchen (1984) described renal neoplasms in more detail. Renal carcinomas
generally consisted of epithelial cells in a tubular or acinar pattern and were located in the
cortex. Renal adenomas included primarily small masses of epithelial cells forming tubular
or papillary structures, and were also located in the cortex. The sarcomas, composed
predominantly of spindle cell type, had a more central pelvic location. Several exposed
male rats over 12 months of age demonstrated karyomegaly and renal tubular epithelial
cells with large nuclei protruding into the tubule lumen. Renal tubular epithelial hyperplasia
was noted in several male rats over 18 months of age. The author suggested that these
alterations may be pre-neoplastic in nature.
In a detailed evaluation of the kidney and liver effects in female mice used in the
MacFarland study, U.S. EPA (1987d) found no histomorphological evidence of chronic
renal toxicity or hepatotoxicity in kidney and liver sections of high-dose (2,056 ppm)
female mice as compared with control mice.
In the MacFarland study, the gasoline was entirely vaporized for the animal
exposure so that the vapors inhaled were essentially identical to gasoline in the liquid
phase. Such a vapor mixture would not be found in ambient situations due to the
differential volatility of the hydrocarbon compounds present in gasoline. In fact, the
branched-chain hydrocarbons that induce nephrotoxiciry, particularly 2,2,4 -
trimethylpentane, have very low vapor pressures and are present in much lower quantities
in the vapor emissions at the service pump or when volatilizing from other sources of
gasoline, than they are in liquid gasoline.
The mechanism of hydrocarbon-mediated nephrotoxiciry, particularly as it may
relate to the appropriateness of the male rat as a surrogate model for humans, has been
10-45
-------
extensively studied. It appears from recent research (Charbonneau, et al., 1978b; Kloss et
al., 1984; Lock et al., 1987b) that the hyper-plastic and neoplastic kidney lesions observed
in male rats after chronic exposure to hydrocarbons are the result of hyaline droplet
accumulation within the proximal tubule of the male rat kidney. Such an accumulation
produces cell damage or death, which in turn stimulates cell proliferation and ultimately a
neoplastic response. It is postulated that the hyaline droplet accumulation is the result of a
particular protein, alpha-2-microglobulin, synthesized in the liver of the male rat, which
becomes covalently bound to metabolites of 2,2,4-trimethylpentane, secreted out of the
liver as such, filtered by the glomerulus of the kidney, and reabsorbed in the proximal
tubule cells. However, this altered protein, unlike other proteins, cannot be degraded into
its constituent amino acids by the lysosomal apparatus of the kidney cells, but instead,
becomes deposited in the kidney as hyaline droplet accumulations. Production of alpha-2-
microglobulin is very low in female rats and absent in the mouse. Since alpha-2-
microglobulin has not been detected in humans, the relevance of the induction of
nephropathy and renal neoplasia in the male rat to human risk from gasoline vapor
exposure is highly uncertain in the opinions of these researchers.
While the foregoing argument is attractive from the biochemical standpoint, its
relevance to the issue of renal tumors in rats rests on the point that the accumulation of
hyaline droplets in the kidney demonstrated in subchronic studies is the cause of renal
tumors at multiple sites found in a single chronic study. While this may be the case, an
equally plausible and acceptable explanation is that differences in the pharmacokinetics of
the agent may explain the susceptibility of rat kidney to both the nephrotoxic and the
carcinogenic effects. This is consistent with the presence of an occasional tumor in the
mice and female rats which, as extensive work has shown, do not accumulate hyaline
droplets, and do not accumulate TMP preferentially as does the male rat It is concluded,
therefore, that there is insufficient basis to reject the observation that the sample of gasoline
tested in the MacFarland study is a carcinogen.
The mouse strain used in the chronic inhalation study of unleaded gasoline is highly
susceptible to liver tumors and, according to Trump et al. (1984), may not appropriately
represent the sensitivity of the general human population. Since a significant increase in
tumors was seen only in high-dose females (a 14 percent background incidence of liver
tumors in the female controls), and no effects were observed in male mice, the results are
judged as equivocal in nature. Other than the classical argument that is presented when
liver tumors are observed in mice, no support was found to suggest that the response to
gasoline is unique to mice. Since mouse liver tumors have been used to estimate
carcinogenic potency with other hydrocarbons, it is believed to be equally appropriate to
10-46
-------
use them in the case of gasoline.
10.2.3 Skin Paintinp Study
In the Eppley Institute for Research in Cancer (1983) study, carcinogenic potential
of gasoline was investigated by skin painting SO Swiss mice with unleaded gasoline 3
times/week (0.05 ml/application) until death or sacrifice after 2 years. Two positive control
groups of SO male mice each received skin applications of 0.05 or 0.15 percent
benzo(a)pyrene on the same schedule. Another group of 122 male mice served as untreated
controls. After 2 years of treatment, there was no apparent carcinogenic response; lung
adenomas, liver hemangiomas, and malignant lymphomas were the predominant tumors in
both treated and untreated control mice with incidences in gasoline-treated mice comparable
or lower to those of the controls. Hyperkeratosis, actoabscess, and ulceration of the skin
were significantly increased (p<0.01) in the treated groups as compared with the untreated
controls. The incidences of dermal squamous cell carcinoma in the two positive control
groups were 92 and 96 percent, respectively, substantiating that a sensitive test system was
employed for dermal carcinogenicity.
10.2.4 Carcinogenicitv Studies for Benzene. Toluene, and Xvlene
Benzene, toluene, and xylene are frequently found in drinking water supplies as a
result of gasoline spills or leaking underground storage tanks. It is theoretically possible to
predict the risk of a mixture based on the carcinogenicity of the major components. In this
case, the carcinogenic effects of benzene would dominate this consideration. Furthermore,
the need to base the carcinogenic risk on the major carcinogenic component of gasoline is
strengthened by the questions raised relative to the MacFarland study and the fact that the
amount of benzene used in gasoline varies from less than 1 percent to as high as 5 percent.
The following sections summarize data relative to the animal carcinogenicity of each
major component. This discussion is not exhaustive. Since the actions of benzene on the
blood forming organs is well founded, it is curious that there were no effects reported in
the MacFarland study. A protective effect of gasoline hydrocarbons against benzene is one
possible, but unlikely, explanation. Although available data indicate that the actions of
toluene and xylene as carcinogens are believed to be minimal at best, they are included in
this section.
10.2.4.1 Benzene
Available studies on the carcinogenicity of benzene for the period 1932 to 1981
have been summarized in IARC (1982). Val Raalte and Grasso (1982) reviewed the
10-47
-------
available data through mid-1982. NTP (1986b) and NRC (1986a) have summarized some
of the more recent carcinogenicity data on benzene. Studies reviewed include both positive
and negative findings.
Hematopoietic system neoplasms, mainly leukemias or lymphomas, have been
associated with exposure of rodents to benzene via the oral route (Maltoni et al., 1985;
NTP, 1986b) or via inhalation (Cronkite, 1986; Cronkite et al., 1984,1985; Maltoni et al.,
1985; Snyderetal., 1980).
Other neoplastic lesions associated with oral or inhalation exposure of rodents to
benzene include carcinomas of the mammary gland, Zymbal gland, skin, oral cavity, nasal
cavity, lungs, and preputial gland; adenomas of the harderian gland and lung; papillomas of
the skin and oral cavity; and tumors of the forestomach, liver, lungs, and ovaries
(Cronkite, 1986; Cronkite et al., 1984, 1985; Maltoni et al., 1985; NTP 1986b; Stoneret
al., 1986; Snyder et al., 1980).
IARC (1982) cited six investigations in which there was no indication that topical
application of benzene induced skin tumors. In the NTP (1986b) study, Fischer 344 rats
and B6C3F1 mice of both sexes were administered 0,50,100, or 200 mg/kg benzene
(males) or 0,25,50, or 100 mg/kg benzene (females) 5 days/week for 103 weeks. Mice
exhibited compound-related neoplastic effects in the hematopoietic system (e.g., malignant
lymphomas). Zymbal gland carcinomas were seen in both species. Squamous cell
papillomas or carcinomas of the skin oral cavity (rats) and forestomach (mice) were seen,
and mice exhibited a significant increase in the incidence of pheochromocytomas.
Furthermore, neoplastic effects in the lung, liver, harderian gland, preputial gland, ovary,
and mammary gland were observed in mice.
Cronkite et al. (1984,1985) reported that female mice exposed via inhalation to 300
ppm benzene for 6 hours/day, 5 days/week, for 16 weeks were found to have increased
incidences of leukemia (8/88 vs. 20/89). Furthermore, the incidence of Zymbal gland
tumors and ovarian tumors was increased (1/88 vs. 16/89 and 0/88 vs. 8/89, respectively).
These authors speculate that shorter exposures of study animals to benzene may result in
higher incidences of neoplasia than long-term continuous exposure because the latter may
suppress the incidence of lymphoma or shorten the lifespan so that lymphomas cannot be
observed.
10.2.4.2 Toluene
A carcinogen screening assay by Maltoni and his colleagues (Maltoni et al., 1985)
indicated a positive association between toluene exposure and increased malignant tumor
incidence in Sprague-Dawley rats. Increases were observed both in the numbers of
10-48
-------
malignant tumors per animal and in the number of tumor bearing animals for both sexes
when toluene was administered by gavage at 500 mg/kg for 141 weeks.
The findings of Maltoni et al. contrast with the results from earlier investigations.
The Chemical Industry Institute of Toxicology (CIIT, 1980) conducted studies on Fischer
344 rats. Exposure to 113, 375, or 1,125 mg/m3 of toluene did not produce any increased
incidence of neoplastic or degenerative changes. The lack of increased neoplastic changes,
however, is subject to two important qualifications. The highest test dose (1,125 mg/m^)
may not have been the maximum tolerated dose. Also, the spontaneous incidence of
mononuclear cell leukemia in Fischer 344 male rats is high (16 percent) and thus may
render this strain inappropriate for study of a chemical that might be myelotoxic (U.S.
EPA, 1983b). Studies have indicated that toluene is not carcinogenic when applied
topically to the shaved skin of mice (Pohl and Fouts, 1973; Lijinsky and Garcia, 1972;
Coombs et al., 1973; Doak et al., 1976), and that it does not promote the development of
epidermal rumors following initiation with DMBA (Frei and Kingsley, 1968; Frei and
Stephens, 1968). Other skin painting experiments did not indicate any increase in tumors
in three strains of mice exposed twice weekly to 0.05 to 0.1 ml of toluene for 1 year (Doak
et al., 1976). Also, toluene did not result in any promoting effect when applied topically to
Swiss mice for 20 weeks following initiation with 7,12-dimethylbenzanthracene (Frei and
Stephens, 1968; Frei and Kingsley, 1968).
10.2.4.3 Xvlene
Maltoni et al. (1985) reported a series of studies primarily on the carcinogenicity of
xylene in rats. In these studies, a group of 40 male and 40 female Sprague-Dawley rats
were administered xylene (500 mg/kg) by gavage in olive oil, 4 to 5 days/week for 104
weeks. Control animals (50 per sex) received olive oil on the same schedule. After the
dosing period, the rats were observed until death (maximum of 141 weeks). Malignant
tumors were detected in 14/40 and 22/40 males and females, respectively, from the xylene-
treated group as compared to 10/50 male and 11/50 female controls. Tumor sites and
survival data were not presented; thus, insufficient data were available upon which to base
an evaluation of the findings or an assessment of the carcinogenic potential of xylene.
The NTP (1986a) has completed a peer review of a 2-year chronic gavage study of
mixed xylenes in Fischer 344 rats and B6C3F1 mice. The xylene used in this study
contained 60.2 percent m-xylene, 13.6 percent p-xylene, 9.1 percent o-xylene, 17.0
percent ethylbenzene, less than 0.3 percent other volatile impurities, and less than 5 ppm
benzene. Xylene was administered in corn oil, by gavage, to groups of 50 male and 50
female rats at levels of 0, 250, or 500 mg/kg, 5 days/week for 103 weeks. Groups of 50
10-49
-------
mice of each sex were administered doses of 0,500, or 1,000 mg/kg on the same schedule.
Control animals received corn oil. There was a dose-related decrease in survival in male
rats, but many of the deaths were gavage-related. No significant differences in survival
were observed for female rats, or for male and female mice. No increased incidences of
tumors were found among treated female rats or treated female mice when compared to
controls. Interstitial cell tumors of the testes occurred in 43/50 control, 38/50 low-dose,
and 41/49 high-dose male rats. Although the tumor incidences in the low- and high-dose
groups were not significantly different compared with controls by the Fisher Exact test,
significant differences for the high-dose group were obtained by life table test (p<0.001)
and incidental tumor tests (p=0.027), which adjust for survival. Testicular tumors were
observed in 13/13 high-dose males that died during weeks 62 to 92, compared with 4/9
vehicle control males that died during the same period. No other differences were found
for other time intervals; thus, the effect was probably not compound-related. Although the
statistically significant tumor incidence suggest tumorigenesis when time-to-tumor data are
evaluated, the NTP (1986a) concluded that under the conditions of this bioassay, there was
no evidence of carcinogenicity of mixed xylenes in rats or mice.
Two dermal carcinogenicity studies of xylene are reported in the open literature.
Berenblum (1941) reported that xylene was not carcinogenic in skin painting studies in
mice. Pound (1970) conducted initiation-promotion studies in mice with xylene as the
initiator and cotton oil as the promoter. Pound (1970) concluded that xylene did not act as
an initiator. Since skin painting studies such as these are not considered to be adequate
tests for systemic carcinogenicity, further details of the study design are not included.
10.3 QUALITATIVE ASSESSMENT OF CANCER RISKS OF GASOLINE
It is possible to evaluate the evidence of and potency of the carcinogenicity of
substances such as gasoline to humans based upon studies involving non-human species
by assessing non-quantitative indicators of carcinogenic potential and potency. Such
qualitative indicators of possible cancer risk to humans are exemplified by the following.
• corroboration
• multiple routes of exposure
• multiple species
• carcinogenicity to mammals
• causation of malignant tumors
• diversity of tumor types
• diversity of primary tumor locations
• early tumor onset
10-50
-------
• evidence of genotoxicity
• structure of activity relationships
• quantitative risk assessment
Each of these qualitative indicators of cancer potency is evaluated with regard to
gasoline and/or constituents thereof in the paragraphs that follow.
10.3.1 Corroboration
The fact that findings of carcinogenicity have been reported by more than one
investigator, in multiple studies, reduces the possible role of experimental error, and
increases our ability to rely upon the evidence. This criterion has clearly been met in the
case of gasoline and, most notably, with respect to benzene.
10.3.2 Multiple Routes of Exposure
If carcinogenic responses are elicited by a substance administered via multiple
exposure routes, such as oral in food, oral in drinking water, gavage (stomach tube),
dermal contact, and inhalation, this suggests that it possesses relatively great versatility as a
carcinogenic agent when compared with a substance that is carcinogenic by only a single
route of administration. Gasoline itself was shown to be carcinogenic via inhalation,
although not via skin painting. Benzene has been shown to be carcinogenic via the oral and
inhalation routes in rodents.
10.3.3 Multiple Species
The fact that at least two species (such as mice and rats) have exhibited carcinogenic
responses following exposure to a substance indicates that the effect is not species-specific,
with obvious implications for humans. In the case of gasoline, evidence of carcinogenicity
has been derived from cancer bioassays involving rodents, as well as epidemiological
studies involving individuals occupationally exposed to gasoline or its constituents
separately, especially benzene. Indeed, benzene is regarded as a known human
carcinogen.
10.3.4 Carcinogenicitv to Mammals
The fact that species exhibiting carcinogenic responses include mammals imparts
special significance for humans. It is reasonable to assume that mammals share more in
common with one another, including sensitivity to toxic substances, than do groups of less
phylogenetically (or taxonomically) related species. In addition to the known role of
benzene as a causative agent in human leukemia, gasoline has been shown to be
10-51
-------
carcinogenic to mice and rats (see section 10.2).
10.3.5 Causation of Malignant Tumors
The fact that some tumors were malignant rather than benign constitutes an added
basis for concern, although few if any tumorigenic substances are know to cause only
benign or only malignant tumors. Gasoline has clearly been shown to induce malignant
tumors, such as metastasizing hepatocellular carcinomas in female mice. Benzene
causation of acute myelogenous leukemia in humans further exemplifies this qualitative
indicator of carcinogenic potential.
10.3.6 Diversity of Tumor Types
The fact that several types of tumors were exhibited in some organs constitutes an
indicator of carcinogenic potency, and increases the likelihood of carcinogenic risk to
humans. Gasoline has been associated with causation of hepatocellular adenomas and
carcinomas, as well as renal adenomas, carcinomas, and sarcomas, along with a variety of
apparently preneoplastic changes (see section 10.2). Similar evidence of tumor diversity
has been evinced in the case of benzene. Administration of benzene to animals has been
associated with leukemias, lymphomas, and carcinomas of the mammary gland, Zymbal
gland, preputial gland, skin, oral cavity, and lungs, as well as with a variety of adenomas
and papillomas.
10.3.7 Diversity of Primary Tumor Locations
The fact that multiple organs were affected enhances the impression of carcinogenic
potency; if the target organs affected in rodents are also present in humans, this precludes
assertions that only target organs unique to rodents-such as the forestomach-are affected.
Clearly, gasoline and its constituent, benzene, have been associated with a diversity of
primary tumor locations, although metastasis and formation of secondary tumors has also
been documented (see section 10.2).
10.3.8 Multiplicity of Tumors
The fact that more than one tumor per individual was exhibited indicates a greater
carcinogenic potency than would be indicated by a substance producing the same elevation
of tumor incidence, but with fewer tumors per individual. This qualitative indicator is
applicable only when care has been taken to assure that the tumors reported were of
primary rather than secondary origin: they were each caused independently by the chemical
rather than by metastasis or pre-existing tumors in the same individual. Although some
10-52
-------
bioassay animals exposed to gasoline vapors exhibited multiple tumors, a systematic test of
whether and to what degree the incidence of multiple tumors per individual differs between
treated and control groups.
10.3.9 Earlv Tumor Onset
The fact that treated animals exhibited not only an elevated incidence, but an earlier
onset of tumors compared with untreated controls indicates potency of a test substance.
This qualitative indicator cannot be evaluated based upon the data reported earlier in this
chapter.
10.3.10 Evidence of Oenotoxicitv
The fact that a substance has been associated with mutagenic effects suggests that
the mechanism of cancer causation is genotoxicity rather than epigenetic effects (effects
exerted upon targets other than genes). Genotoxic animal carcinogens may be of greater
concern than epigenetic carcinogens with respect to potential human cancer risk. Although
mutagenicity testing of gasoline has yielded equivocal results, dose-related elevations in
mutation frequency have been elicited in at least three tests involving mouse lymphoma
cells in culture with and/or without metabolic activation (see chapter 9). Studies involving
exposure of bacteria to benzene have been generally negative for gene mutation, although
chromosome abnormalities have been induced. These results suggest that causation of
cancers by benzene and by gasoline as a mixture may be mediated by non-genotoxic
mechanism(s) of action.
10.3.11 Structure Activity Relationships
The fact that a substance having a chemical structure similar to known carcinogens
suggests possible carcinogenic potential of the substance. For example, at least 37, or
nearly all hydrazine analogs tested, have proved to be tumor producers in such organs as
intestines, blood vessels, lungs, livers, kidneys, breasts, and the central and peripheral
nervous systems. This observation justifies particular a priori concern about other
members of this chemical class, even though data about those other members may be
otherwise limited. The question of whether compounds structurally similar to indicator
substances addressed in the present assessment has not been evaluated.
10.4 CONCLUSIONS
It is a finding of this assessment that the animal bioassays provide a sufficient basis
for presuming gasoline to be a probable human carcinogen. While concerns have been
10-53
-------
raised regarding the relevance of the gasoline bioassays to human health risk assessment,
they rest largely on speculative arguments regarding mechanisms of carcinogenicity in
rodents. As such, several scientific issues must be addressed if these concerns are to
become relevant to risk assessment. Some of the more prominent scientific issues are listed
below.
1. Questions have been raised regarding the relevance of the cancer bioassays because the
rodents were exposed to wholly vaporized gasoline. While it is possible that the volatile
portion of the gasoline mixture may not have been responsible for the increased tumor
incidence observed in these studies, an adequate bioassay on this fraction of this mixture
has yet to be conducted.
2. The mechanisms of renal carcinogenicity are still not yet completely understood,
including those which postulate a link between nephropathy and neoplasia. Specifically,
while nephropathy is common in old rats, kidney tumors are rare in unexposed animals.
3. 2,2,4-trimethylpentane is preferentially concentrated in the kidney. Yet, no data were
available concerning the proportion of TMP in the kidney that is associated with the alpha-
2-microglobulin and the proportion concentrated there by other means. If the kidney
tumors were caused by TMP, the weaker response in the female rats could simply be due to
the fact that they did not receive as much of the compound in their kidneys.
4. In the case of hydrocarbon-induced nephropathy, not all cells which contained hyaline
droplets stained positive for alpha-2-microglobulin. Thus, it is possible that other proteins,
not necessarily unique to the male rat, could result in hyaline droplet formation and
subsequent nephrotoxicity. Certain human population subgroups who have high
concentrations of low molecular weight proteins (such as lysozymes or Bence-Jones
protein), may also be at risk if kidney tumors and protein-induced nephropathy are linked.
5. Kidney tumors were observed in the high-dosed female mice and in a mid-dosed female
rat in addition to the male rats. Because these tumors are generally rare in control animals,
their appearance in the tested animals indicates that the renal carcinogenicity of gasoline
may not be sex- and strain-specific. In addition, the appearance of the kidney tumors in the
female rodents was not accompanied by severe nephropathy, thus further discounting the
role of nephropathy in kidney tumorigenesis.
10-54
-------
6. The exact sites of tumorigenesis is uncertain.
7. If nephropathy is shown to be a causative mechanism in the development of renal
tumors, it may still be relevant to humans. Patients with chronic renal diseases appear to be
at increased risk for the development of kidney adenomas and carcinomas. This is
consistent with the alpha-2-microglobulin hypothesis, which ultimately states that kidney
tumors result from non-specific, cytotoxic mechanisms (e.g., lysosomal enzymes released
from dying cells or increased genetic mutations resulting from increased cell turnover).
Gasoline exposure may augment the nephrotoxicity in these patients, and, consequently,
their risk of cancer.
8. Few studies have been done using agents that induce both nephropathy and kidney
tumors. Also, in all the chronic studies only a single strain of mice and a single strain of
rats were used. Thus, none of the chemicals known to induce both nephropathy and
kidney tumors have been tested in more than a very limited number of species or assays.
9. Human exposure to gasoline occurs through a wide variety of pathways, such as
inhalation (vapors, aerosols, contaminated dusts, deliberate intoxication), soil ingestion,
ingestion of contaminated food, dermal exposure, or ingestion of contaminated drinking
water. Thus, if it is found that only certain fractions of the gasoline mixture are
carcinogenic, this finding may apply to only a specific subset of exposure situations.
10. The animal bioassay findings of dose-related increases in female mouse liver tumors
supports the presumption that gasoline is a carcinogen.
10.5 SUMMARY
One adequate carcinogen bioassay of gasoline vapors has been conducted with
gasoline vapors under the sponsorship of the American Petroleum Institute. In that study,
statistically significant increases in kidney tumors were observed in male Fischer 344 rats
and hepatocellular tumors in female B6C3F1 mice. Major uncertainties are: (1) the vapor
composition in this study was different from the ambient human environment (i.e., the
gasoline was entirely vaporized for the animal exposure, whereas in real-life situations,
gasoline vapors are dependent on the differential volatility of the hydrocarbons present in
gasoline), and (2) the kidney tumors observed in male rats may result from a unique
mechanism specific to male rats and not to female rats or other species.
Hematopoietic system neoplasms, mainly leukemias or lymphomas, have been
10-55
-------
associated with exposure of rodents to benzene via the oral route or inhalation. Other
neoplastic lesions associated with oral or inhalation exposure of rodents to benzene include
carcinomas of the mammary gland, Zymbal gland, skin, oral cavity, nasal cavity, lungs,
and preputial gland; adenomas of the harderian gland and lungs; papillomas of the skin and
oral cavity; and tumors of the forestomach, liver, lungs, and ovaries.
Available data are inadequate to determine the carcinogenicity of toluene or xylene.
10-56
-------
11. RISK ASSESSMENT
11.1 DETERMINATION OF EQUIVALENT DOSES
When estimating health risks from chemical exposure, possible pharmacokinetic
differences associated with extrapolations between species, dose levels, and routes of
exposure must be analyzed. The primary purpose of this analysis is to ensure that
pharmacokinetic differences between the study population and the reference human
population do not lead to a significant underestimation of the potential health risk. Once
this condition is satisfied, pharmacokinetic studies may then be used in efforts to define the
dose-response relationships between chemical exposure and health effect more precisely
than could be done by using only the administered dose or exposure concentrations. Much
of this secondary analysis can be quite refined, however, and is beyond the scope of this
assessment.
Initial factors to consider when correlating pharmacokinetics with toxicology are
how much of the administered dose is absorbed by each exposure route, and how the
absorbed dose varies among species and between exposure levels. Data sufficient to
characterize the variables which influence the degree of absorption are usually lacking.
Therefore, certain assumptions must be employed to ensure that pharmacokinetics are being
properly considered in the estimation of an upper bound on potential health risks. Thus, it
is important that the absorbed dose is not overestimated in the study population or
underestimated in the reference population, as these errors would both serve to
underestimate the risk.
The major pharmacokinetic issues relevant to the quantitative risk assessment of
gasoline and its components concern uptake via the pulmonary and dermal routes and the
tissue clearance of the absorbed compounds. Uptake of gasoline via the oral route is
generally considered to be complete in both humans and laboratory animals. Dermal uptake
from dilute aqueous solutions per unit surface area is assumed to be equivalent across
species and exposure levels.
Because of their relatively high blood/air and tissue/blood partition coefficients,
initial uptake of many gasoline components from the pulmonary region proceed rapidly and
can be estimated from appropriate blood/gas partition coefficients. Assuming that the air
concentration is stable, once blood levels have equilibrated with tissue levels, pulmonary
uptake should equal tissue clearance (Andersen, 1983). For inhalation and all other
exposure routes, flow limited clearance of absorbed gasoline components is assumed for all
relevant experimental and actual exposure situations. Given this assumption, it can be
further inferred that uptake follows a linear relationship until metabolic saturation is
11-1
-------
reached. These assumptions are supported by the work of Andersen (1983), who found
that clearance of many low molecular weight hydrocarbons is flow limited at low
concentrations. Therefore, interspecies differences in delivered dose can be estimated on
the basis of relative uptake and perfusion relationships.
Thus, with the assumption of flow limited clearance, metabolic clearance of
gasoline vapors and its aromatic constituents should be related to blood flow to the
metabolizing organs. Because metabolic parameters at these low (e.g., sub-ppm) exposure
levels have not been adequately characterized, this assessment considers that all tissues may
possess significant metabolizing ability.
If the assumption is made that metabolism at higher concentrations can be estimated
from the liver clearance, then delivered dose differences can be considered provided
saturation has not been reached. Human and animal studies of xylene and toluene
metabolism (Ogata et al., 1970) indicate that metabolic rates are linear up to exposure
concentrations as high as 750 mg/m^ (about 200 ppm). Animal studies (Deichmann, 1963)
indicate that a linear response occurs at benzene concentrations up to approximately 100
ppm. Thus, for most lexicological studies on sensitive health endpoints, flow-limited
clearance can be assumed for these three compounds.
Equations have been developed to estimate pulmonary uptake of vapors under
steady-state conditions (Andersen, 1983). These equations which appear below, are based
on the premise that elimination occurs through metabolism and that losses due to excretion
of the un-metabolized compound (such as in the urine or through the skin) do not
significantly alter the calculations.
"ami = Valv (cexp - Calv) = Q(cart - «ven) = Q Cart ERSystemic (1 )
where:
= the uptake from the atmosphere;
= the minute alveolar ventilation;
cexp = exposure concentration;
Calv = alveolar (exhaled air) concentration;
Cart = arterial blood concentration;
cven = venous blood concentration;
Q = cardiac output;
ERsystemic = systemic extraction ratio.
11-2
-------
The expected blood/gas concentration ratio at steady state (IIss) can be derived from
an algebraic manipulation of the above equations.
nss = cart / cexp = X blood/gas (2)
1 + (Q/ Valv) X blood/gas ERSystemic
where:
X blood/gas is the blood/gas partition coefficient
The following assumptions are made to simplify the calculations. First, it is
assumed that metabolism at low doses is essentially 100% (systemic extraction being equal
to total tissue perfusion). Second, a minute alveolar ventilation rate of 10 liters/minute
(corresponding to an inhalation rate of 21.6 m^/day) is assumed. This alveolar ventilation
is about twice the rate of an ordinary person at rest and about half the rate of a person
engaging in light exercise (50 w) (Astrand, 1983). A cardiac output of 7 liters/minute is
also assumed. Seven liters/minute is half-way between the cardiac output at rest - 5
Uminute- and during light exercise - 7 to 12 L/min (Astrand, 1983). These last two
assumptions result in an estimated Q/ Valv ratio of about 0.8. Consequently, Equation 2
can be simplified to the following forms for the low dose (Equation 3) situation.
nss = A. blood/gas (3)
1 + 0.8 X blood/gas
For a compound with a low blood/gas partition coefficient, such as ethane (X
blood/gas = 0.1), the riss would be near its blood/gas partition coefficient (e.g., 0.1),
regardless of the exposure concentration. (Although ethane is not present in gasoline, it is
structurally similar to a major class of gasoline components, the short-chained aliphatics.)
For a compound with a high blood/gas partition coefficient, such as xylene (X blood/gas =
38), however, the Ilss would be closer to unity (X blood/gas/ 0.8 X blood/gas) at low
exposure. Assuming that these two chemicals represent the partition coefficient bounds on
biologically important gasoline components, then a range of pulmonary uptake estimates
for gasoline can be derived. An assumption of complete metabolism at low doses has little
impact on pulmonary uptake when flow limited clearance is assumed.
11-3
-------
Ethane:
- Low dose assumption (i.e., complete metabolism)
= cart/cexp = X. blood/gas _ = 0.1 =0.1
1 + 0.8 A blood/gas 1 + 0.08
For an exposure concentration of 1 ug/L, the Can = 0. 1 x 1 ug/L = 0. 1 ug/L
uatm = 7 L/min x 0.1 ug/L x 1.0 = 0.7 ug
% uptake = 0.7 ug/ ( 9 L/min x 1 ug/L) = 0.07 =
Xylene:
- Low dose assumption (i.e., complete metabolism)
IIss = can/cexp = X blood/gas _ 38 =1.2
1 + 0.8 X blood/gas 1 + 30.4
For an exposure concentration of 1 ug/L, die Cart = 1.2 x 1 ug/L =1.2 ug/L
uatm = 7 L/min x 1.2 ug/L x 1.0 = 8.4 ug
% uptake = 8.4 ug/ ( 9 L/min x 1 ug/L) = 0.93 = 93%
These examples illustrate the likelihood that the fraction of gasoline absorbed via the
pulmonary route is very different from the inhaled fraction. Particularly, the aromatics
appear to be preferentially absorbed relative to the other hydrocarbons in the mixture.
Characterization of the absorbed gasoline vapor fraction is an important research need, as it
would provide a greater pharmacological understanding of gasoline toxicity.
Particularly, limited data exist concerning several key areas of gasoline vapor
pharmacokinetics: (1) an inadequate database on the partition coefficients for die several
structurally diverse categories of compounds in gasoline; (2) lack of information
concerning the interactions of compounds within the mixture which could influence
pulmonary absorption; (3) lack of information concerning the effects of metabolic
saturation on pulmonary absorption; and (4) lack of sufficient identification and
characterization of all the compounds within the gasoline mixture which have the potential
to be biologically important.
For the most pan, these uncertainties cannot be addressed quantitatively. Rather,
general assumptions are made based on the known pharmacokinetic characteristics of
specific gasoline components. Assuming that metabolism is flow-limited for benzene,
11-4
-------
toluene, and xylene for all relevant exposure situations, consideration of pulmonary uptake
and organ perfusion parameters leads to a conclusion of a rough dose equivalence among
species on the basis of exposure concentration. This conclusion is an outcome of two
countervailing properties: inhalation rates are faster in rodents relative to human beings, but
so are the perfusion rates. These rates are similar in magnitude. Thus, while rodents
would inhale more compound than human beings during any given period of time, they
would also eliminate it more quickly due to a faster rate or perfusion and metabolism. The
similarity of inhalation rates and perfusion rates as functions of body weight are illustrated
in equations 5 (Fiserova-Bergerova and Hughes, 1983) and 6 (Guyton, 1947).
Liver perfusion (liters/minute) = 0.066 (BW)0-74 (5)
Respiratory rate (cc/minute) = 2.1 (BW)°-75 (6)
This equivalence is not exact, however, because of a higher ventilation rate in
human beings engaged in light activity throughout the day. Consideration of increased
ventilation would increase the inhaled dose by about a factor of 1.5 - 2 relative to what
should be predicted from Guyton's equations. This comparison is described below.
Respiratory rate (humans) = 2.1 x 70.0000-75 cc/min
= 9,000 cc/min
= 9L/min
Alveolar Ventilation (humans) = 6 L/min
By contrast, the inhalation rates used in this assessment are assumed to be the following:
Ventilation Rates
Workers (Occupational studies) L/min nvVhr
Total Ventilation 20 1.2
Alveolar Ventilation 13.3 0.8
Adults (General Population)
Total Ventilation 15 0.9
Alveolar Ventilation 10 0.6
11-5
-------
Infams(10kg;4m3/day)
Total Ventilation
Alveolar Ventilation
3
1.7
0.17
0.1
Animal body weights were not specified in many of the animal studies reviewed in
this document. Default body weights are thus assumed for the purposes of calculating
inhaled doses on a mg/kg/day basis. These default weights are: monkeys - 5 kilograms;
rabbits - 4 kilograms; rats - 250 grams; and mice - 25 grams. These estimates are based on
visual examination of a graph which describes the range in experimental body weights, as
provided by Dourson and Stara (1983). The inhalation rates for these animals are
presented below.
Respiratory rate (monkeys) = 2.1 x 50000-75 cc/min
= 1,200 cc/min
= 1.2L/min
Respiratory rate (rabbits)
Respiratory rate (rats)
Respiratory rate (mice)
: 2. 1x40000-75 cc/min
= 1,000 cc/min
: 1 L/min
: 2.1x2500-75 cc/min
• 100 cc/min
0.1 L/min
: 2.1x250-75 cc/min
: 20 cc/min
•• 0.02 L/min
Although these inhalation rates are based in general assumptions concerning body
weights, actual deviations from these body weight estimates are not likely to change the
inhalation rate estimates significantly. Also, because these inhalation rates are only
approximations, no adjustments are made for alveolar ventilation. This may introduce
slight errors into the analysis; however, not adjusting for alveolar ventilation in animals
reflects the approximate nature of these estimates and serves to simplify the risk estimation
process.
11-6
-------
Thus, for systemic effects, if the inhaled dose from a rodent study is standardized
to a mg/kg/day dose, estimation of the equivalent human dose will require a scaling to
account for slower perfusion, and hence, metabolism and elimination. If direct effects on
the respiratory system are considered, interspecies adjustments can be made solely on the
basis of the inhaled mg/kg/day dose. Scaling based in metabolic difference is achieved by
applying the following relationship.
DA x BWA/BWA0.74 = DH x BWH/BWH0-74
where:
DA = animal dose (mg/kg/day)
BWA = animal body weight
BWA0-"74 = animal liver perfusion
DH = human dose (mg/kg/day)
= human body weight (kg)
4 = human liver perfusion
Similarly, comparisons involving drinking water studies are based on drinking
water concentrations, as water consumption rates among species also vary in relation to
surface area. Thus, for these studies, a mg/kg/day dose is scaled to reflect the faster rate of
metabolism and elimination of ingested compounds in laboratory animals.
As with the inhalation rate estimates, these scaling factors represent general
approximations. They are, nonetheless, unlikely to fluctuate significantly with variations in
body weight.
A conservative assumption employed in this assessment is that all absorbed
gasoline and gasoline components are metabolized. It is further assumed that all gasoline
which reaches the alveoli is absorbed, although this latter assumption is likely to be very
conservative for compounds with low blood/gas partition coefficients. Thus, estimates of
the human systemic delivered dose from any given air concentration can be made by
multiplying the concentration by the alveolar ventilation rate and the duration of exposure.
For respiratory effects, the dose can be estimated by multiplying the total ventilation rate by
the exposure duration.
11-7
-------
TABLE 11-1
00
SUMMARY OF SELECTED ANIMAL LOAELS AND NOAELS DERIVED
FROM GASOLINE HEALTH EFFECTS STUDIES
Test Route of
Effect Level (mg/kg/day)
NOAEL1 LOAEL2
Gasoline
Ventilation Changes
(Leaded Gasoline)
Kidney Toxicity
(Unleaded Gasoline)
Kidney Toxicity
(Full Range Alkylate)
Kidney Toxicity
(Unleaded Gasoline)
Hematotoxicity
(Leaded Gasoline)
Hematotoxicity
(Unleaded Gasoline)
Reproductive Effects
(Gasoline)
Organism
Monkey
Rats
Rats
Rats
Rats
Rats
Rats
Exposure^
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
j - j fl —
Subchronic
(13 weeks)
Subchronic
(21 days)
Subchronic
(21 days)
Chronic
(2 years)
Subchronic
(13 weeks)
Subchronic
(13 weeks)
Subchronic
(30-34 days)
Animal dose HED4
mg/kg/d
16 16
-
1.7 0.3
-
32 6.4
117 23
-
Animal dose HED Reference
mg/kg/d
1
17 2.6 2
2
21 43
1
1
28 64
-------
TABLE 11-1
(continued)
Effect Level (mg/kg/day)
Endpoint
Benzene
Hematotoxicity
(Cytopenia)
Hematotoxicity
(Cytopenia)
Hematotoxicity
(Spleen Changes)
Clastogenicity
Neurotoxicity
(Reflex Activity)
Fetotoxicity
(Low birth weight)
Fetotoxicity
(Hematotoxicity)
Developmental Effects
(Hematotoxicity)
Toluene
Neurological
Effects
Test
Organism
Mice
Mice
Mice
Rats
Rats
Rats
Mice
Mice
Humans
Route of
Exposure^
Inhalation
Oral
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
Study Type
Subchronic
(32 days)
Subchronic
(28 days)
Subacute
(1 week)
Acute
(8 hours)
Subchronic
(5.5 months)
Subacute
(9 days)
Subacute
(9 days)
Subacute
(9 days)
Acute
(6 hours)
NOAEL1 LOAEL2
Animal dose HED4 Animal dose HED Reference
mg/kg/d mg/kg/d
8.6 0.8 5
8.0 0.8 6
0.9 0.09 7
0.1 0.02 - - 8
1.2 0.2 - - 9
5 1 - - 10
4 0.4 11
4 0.4 - - 11
16 16 - 12
-------
Endpoint
Test Route of
TABLE 11-1
(continued)
Study Type
Effect Level (mg/kg/day)
NOAEL1 LOAEL2
ft
Neurobehavioral
(Open field behavior)
Neurobehavioral
(Rotorod peif.)
Neurobehavioral
(Wheel turning)
Hematotoxicity
Pulmonary
(Respiratory Infections)
Fetotoxicity
(low birth weight)
Xvlene
Reproductive
Organism
Mice
Mice
Mice
Mice
Mice
Mice
Rats
Exposure^
Oral
Oral
Inhalation
Inhalation
Inhalation
Inhalation
Inhalation
* * *
Subchronic
(2 months)
Subchronic
(2 months)
Subchronic
(20 days)
Subchronic
(20 days)
Subchronic
(20 days)
Subacute
(7 days)
Subacute
(7 days)
Animal dose HED4 Animal dose HED
mg/kg/d mg/kg/d
14.4 1.4 -
3 0.3
1 0.1
11 1.1
0.5 0.5
580 58
86 17
Reference
13
13
14
14
15
16
17
-------
1 NOAEL: No Observed Adverse Effect Level
2 LOAEL: Lowest Observed Adverse Effect Level
3 HED: Human Equivalent Dose (mg/kg/d), determined for systemic effects by adjusting for interspecies
differences in metabolism. See text for details.
4 See Appendix D for specific details on exposure protocols.
References:
1. Kuna and Ulrich, 1984.
2. Haider et al., 1984.
3. MacFarland et al., 1984.
4. Lipovskii, 1978.
5. Baarson et al., 1984.
6. Hseih et al., 1988.
7. Green etal., 1981.
8. Erexson et al., 1986.
9. Novikov, 1956.
10. Kuna and Kapp, 1981.
11. Keller etal., 1986.
12.Andersen et al., 1983.
13. Kostas and Hotchins, 1981.
14. Horiguchi et al., 1977.
15. Aranyi et al., 1985.
16. Hudak and Ungvary, 1978.
17. Ungvary et al.. 1980.
-------
11.2 DERIVATION OF QUANTITATIVE RISK ASSESSMENT CRITERIA
The derivation of health criteria for gasoline and its constituents benzene, toluene
and xylene first involves the identification of critical studies with adequate quantitative
dose-response information. These health studies were identified in the toxicity, genetic
toxicity, cancer, and reproductive/development effects chapters. Several health endpoints
were identified for various exposure durations. Once these studies are identified,
determinations are made of those studies which represent the most sensitive responses for
the health endpoint which have been investigated. Then, the studies most appropriate for
risk assessment, as determined from considerations of study design and lexicological
relevance, are selected for quantitative risk extrapolation. Health criteria based on these
most sensitive responses should consequently provide protection from other, less sensitive
health responses.
Appendix D presents the critical studies identified for the health effects which have
been reviewed in the previous chapters. No observed adverse effect levels and lowest
observed adverse effect levels are determined. They are presented in a standard mg/kg/day,
human equivalent dose form to allow for comparisons among studies. Gasoline is
evaluated first followed by benzene, toluene, and xylene.
Table 11-1 summarizes the most sensitive health effects endpoints for non-cancer
effects based on the lowest NOAELs and LOAELs derived from the animal studies of non-
carcinogenic effects of gasoline, benzene, toluene and xylene. Carcinogenic risk
assessment criteria are based on the cancer potency estimates developed by the U.S. EPA.
A more detailed discussion of these factors and associated risk levels is presented in section
11.2.2.
11.2.1 Derivation of Quantitative Health Criteria
Health criteria presented in this section are expressed in terms of mg/kg/day doses
and equivalent air concentrations. The focus of these criteria is on inhalation exposures
because: (1) development of air criteria is a primary objective of this assessment, (2) except
for accidental poisonings, ingestion of gasoline-contaminated groundwater, and dermal
contact with contaminated water, inhalation is the primary means of human exposure to
gasoline, and (3) most studies are inhalation studies and thus, limited data exist from which
to derive route-specific health criteria. For situations in which multi-route exposure
pathways are being considered (such as gasoline contamination of residential water
supplies), these criteria (expressed as mg/kg/day exposures) are generally considered to be
applicable for non-inhalation routes. Exceptions to this equivalency assumption occur
when the direct respiratory system effects are being considered.
11-12
-------
Several LOAELs and NOAELs were identified in the previous section. For
threshold effects, the application of uncertainty factors is usually required to estimate
ambient criteria which are protective of public health. In this assessment, these criteria are
denoted as "reference criteria." While there is some variability concerning their application,
depending on the quality and amount of supporting information, these factors are generally
applied as follows.
Extrapolation Factor
LOAEL to NEL (No Effect Level) 5
Animal to Sensitive Human Populations 100
Study Human Population to Sensitive
Human Populations 10
Subchronic to Chronic Exposure 10
Two points specific to this risk assessment should be emphasized. Generally, the
100-fold uncertainty factor that is used when no adverse effect levels are estimated from
animal studies incorporates some consideration for surface area differences. Because
surface area differences have already been addressed in this assessment, it is possible that
this factor may be overly conservative. On the other hand, it is also possible that a ten-fold
uncertainty factor to account for sensitive human populations may not be conservative
enough in some cases. Studies have found ranges in inter-individual variability in response
to toxic substances as high as two orders of magnitude (Calabrese, 1985; Harris et al.,
1987). Moreover, in the absence of comparative human data, there is inherently more
uncertainty when a human response level is predicted from animal data. Thus, the 100-fold
factor is retained in this assessment Should a case be made from the available information,
the resulting reference doses can be adjusted accordingly.
Also, a five-fold factor was selected to estimate a no effect level from a lowest
observed adverse effect level. A five-fold factor was selected because it generally
encompasses the range of LOAEL/NOAEL ratios that have been calculated from surveys of
the existing literature (see for example, Dourson and Stara, 1983). It also seems
appropriate to use this factor when there is adequate information on the dose-response
relationships (Anderson, 1988), as is the case with the effects evaluated in this assessment.
As with the use of the animal to human extrapolation factor, arguments may be made that
could justify a higher (perhaps as much as 10) or lower (perhaps as low as 1 or 2)
extrapolation factor. Consequently, the reference doses may again be modified if more
refined attention is given to this area of uncertainty.
11-13
-------
In this document, the effect levels are determined for the exposure periods used in
the reference study. Thus, no subchronic or chronic factor is employed. Averaging times
may range from minutes or hours (for lethality, CNS effects, or sensory irritation) to 24-
hour exposures (for subacute and subchronic effects) to longer intervals (weeks, months,
years) for chronic effects such as cancer. The exposure concentration is determined by
multiplying the dose (mg/kg/day) by either 10 kg (to reflect a body weight for a one year
old infant) or 70 kg (to reflect the adult body weight), and divided by either 21.6 m^/day to
reflect a total minute ventilation and by 14.4 m^/day (to reflect the minute alveolar
ventilation of 10 L/min) for systemic effects.
The lexicological uncertainty factors may range from as much as 500 (extrapolating
from an animal LOAEL to a no effect level for sensitive human populations) to 10
(extrapolating from a NOAEL in a human study population to a no effect level for sensitive
human populations).
The application of these factors to the dose level is intended to provide a consistent
and conservative derivation of health criteria. More rigorous evaluation of the health
endpoints may result in the application of lower uncertainty factors. Selection of
uncertainty factors used in this assessment has, however, considered the extent to which
the available data could modify the magnitude of the assumed uncertainty. For example, a
factor of 10 is often used to estimate a no effect level from a lowest observed adverse effect
level. Comparisons of LQAELs and NOAELs, however, indicate that this value may be
overly conservative, especially when adequate dose-response information is available
(Anderson, 1988). Indeed, most LOAEL/ NOAEL ratios are in the range of 1 to 5. The
State of Maine employs a factor of 5 when estimating no effect levels from LOAELs in
studies with adequate dose-response information (Anderson, 1988). Because most effects
identified in this assessment have adequate dose-response data, a factor of 5 is used in the
following risk estimation process.
Also, since the human equivalent dose derivation attempts to account for some of
the potential interspecies differences between human beings and laboratory animals, the
interspecies uncertainty factor of 100 may also warrant further evaluation. A specific
example in this regard is the extrapolation of an effect which occurs in a particularly
sensitive species. The air criteria based on male rat kidney tumor lesions, for example,
may also be overly conservative in that this sex/species combination may already be
indicative of a sensitive human population. While such concerns should be addressed for
certain criteria, it is not known whether criteria based in other, less well characterized
endpoints, represent equivalent degrees of conservatism.
11-14
-------
Because it is beyond the scope of this assessment to evaluate these possibilities for
each specific health endpoint, it has limited its approach to a conservative estimate of
potential public health risk. Thus, the conclusion can be made that these health criteria
reflect upper bounds on potential health risks, and that actual risks are likely to be lower
than these criteria indicate. Uncertainties in the assessment, however, prevent more precise
estimates of what those actual risks might be.
Application of uncertainty factors is not generally recommended to derive health
criteria for carcinogenic effects. Instead, because chemical carcinogenesis is assumed to be
a non-threshold response, any exposure is associated with some potential risk. An upper
bound on this potential risk is derived by calculating an upper bound slope on the dose-
response data for an animal bioassay, or a maximum likelihood estimate from human
epidemiological data, assuming low dose linearity in both cases. These slopes, which
represent cancer potency estimates, have been developed by the U.S. Environmental
Protection Agency's Carcinogen Assessment Group. They are presented below for
gasoline and benzene. This assessment has not found evidence that is sufficient to discount
the assumptions of non-threshold and low dose linearity for these carcinogens.
Substance Cancer Potency Estimate (per mg/kg/day)
Benzene 0.026
Gasoline 0.0035
The corresponding reference air criteria for the most sensitive health effects
endpoints for non-cancer effects (see Table 11-1) for gasoline, benzene, toluene, and
xylene are presented in Table 11-2. As mentioned above, the uncertainty factor approach
was not used to derive reference air criteria for carcinogenic effects of gasoline and
benzene. Instead, the results of a linearized multi-stage model were used to estimate air
concentrations which correspond to the upper bound lifetime cancer risks of 10'6,10'5 and
10-4. These data are presented in Table 11-3.
Review of Table 11-2 shows that sensitive toxicity endpoints for each substance
(gasoline, benzene, toluene, and xylene) are associated with a fairly definable dose range.
For gasoline, kidney toxicity is associated with human equivalent doses in the 2 to 4
mg/kg/day range. For benzene, hematoxicity occurs in the dose range of 0.1 to 1
mg/kg/day. For toluene, thresholds for sensitive neurobehavioral, hematological, and
immunological effects occur in the dose range of 0.5 to 1.5 mg/kg/day. Of the studies
which showed responses in these dose ranges, the following endpoints were selected: (1)
gasoline, the kidney toxicity associated with the 2.6 mg/kg/day dose (Haider et al., 1984),
11-15
-------
TABLE 11-2
SUMMARY OF REFERENCE AIR CRITERIA FOR GASOLINE AND GASOLINE COMPONENTS
Substance
Endpoint
Type
Human
Eauivalent Dosel
Uncertainty Reference Reference Air Criteria (ug/m-
Factor Dose^ Adults Infants
Gasoline
Acute Exposures
None
Subacute Exposures
None
Subchronic Exposures
Gasoline - L
Gasoline - U
Gasoline - FRA
Gasoline - L
Gasoline - NR
Ventilation
Kidney Effects
Kidney Effects
Hematotoxicity
Reproductive
NOAEL
LOAEL
NOAEL
NOAEL
LOAEL
16 mg/kg/day
2.6 mg/kg/day
0.3 mg/kg/day
6.4 mg/kg/day
6 mg/kg/day
100
500
100
100
500
0.1 6 mg/kg/day
0.0052 mg/kg/day
0.003 mg/kg/day
0.064 mg/kg/day
. 0.012 mg/kg/day
520
26
15
310
60
400
20
11
240
.
Effects
Chronic Exposures
Gasoline - U Kidney Effects LOAEL
Benzene
Acute Exposures
4 mg/kg/day
500
0.008 mg/kg/day 40
30
Clastogenicity NOAEL
Increase in SCE
0.02 mg/kg/day
100
0.0002 mg/kg/day 1.0
0.7
-------
TABLE 11-2
(continued)
Substance
Endpoint Type
Human
Eauivalent Dose^
Uncertainty Reference Reference Air Criteria (ug/m:
Factor Dose^ Adults Infants
Benzene
Subacute Exposures
Spleen changes LOAEL
Fetotoxicity LOAEL
(Hematotoxicity)
Developmental NOAEL
(Hematotoxicity)
0.09 mg/kg/day
0.4 mg/kg/day
0.4 mg/kg/day
500 0.000 18 mg/kg/day 0.9
500 0.0008 mg/kg/day 3.9
100 0.004 mg/kg/day 19
0.7
-
-
Subchronic Exposures
Cytopenia LOAEL
Cytopenia LOAEL
Neurotoxicity NOAEL
(Reflex Activity)
0.9 mg/kg/day
0.8 mg/kg/day
0.2 mg/kg/day
500 0.00 18 mg/kg/day 8.8
500 0.001 6 mg/kg/day 7.8
100 0.002 mg/kg/day 10
6.6
5.9
7.4
Chronic Exposures
None
Toluene
Acute Exposures
Neurological NOAEL
Effects
16 mg/kg/day
10 1.6 mg/kg/day 7800
5900
-------
TABLE 11-2
(continued)
oo
Substance Endpoint Type Human Uncertainty Reference Reference Air Criteria (ug/nv
Equivalent Dose 1 Factor Dose^ Adults Infants
Toluene
Subacute Exposures
Fetotoxicity LOAEL 58mg/kg/day 500 0.12mg/kg/day 580
(Low birth weight)
Subchronic Exposures
Neurotoxicity
Open field beh. NOAEL 1.4mg/kg/day 100 0.014 mg/kg/d ay 68
Respiratory Inf NOAEL 0.5mg/kg/day 100 0.005 mg/kg/day 24
Hematoxicity LOAEL 1.1 mg/kg/day 500 0.0022 mg/kg/day 11
Chronic Exposures
None
Xvlene
Acute Exposures
None
Subacute Exposures
Reproductive tox LOAEL 17 mg/kg/day 500 0.034 mg/kg/day 165
Subchronic Exposures
None
Chronic Exposures
-
52
19
8
-
None
-------
L - leaded gasoline
UL - unleaded gasoline
FRA - Full range alkylate fraction
NR - composition not reported
SCE - Sister chromatid exchange
' Human Equivalent Dose determined for systemic effects by adjusting for interspecies differences in matabolism. See text for details.
2 Reference Dose is equal to the human equivalent dose/uncertainty factor.
3 Reference air criteria is the reference dose (in mg/kg/day) expressed in ug/nv* using the following factors:
70 kg adult inhaling 14.4 m3/d or 10 kg infant inhaling 2.7 m3/d.
-------
TABLE 11-3
SUMMARY OF ESTIMATED AIR CONCENTRATIONS CORRESPONDING
TO LIFETIME CANCER RISKS FOR GASOLINE1
Substance Endpoint
Cancer Potency Value2 Risk Level Human Equivalent Dose
1 U.S. EPA Carcinogenic Potency Factors.
2 Cancer potency value is the 95% upper bound slope of the dose-response curve (mg/kg/d)'1.
Unleaded
Gasoline
Unleaded
Gasoline
Benzene
Kidney tumors in
male rats
Hepatocellular
carcinoma/adenoma
in female mice
Cases of
leukemias
in humans
3.5 x 10-3
2.1 x ID'3
2.6 x ID'2
lO-6
10-5
10'4
lO-6
10-5
10-4
10-6
10-5
lO'4
2.9 x lO'4
2.9 x lO-3
2.9 x lO-2
4.8 x lO'4
4.8 x 10-3
4.8 x ID'2
3.8 x 10-5
3.8 x lO'4
3.8 x ID'3
1.4
14
140
2.3
23
230
1.8
18
180
11-20
-------
on the basis of it being the most sensitive response and a lexicologically significant effect
for humans, (2) benzene, the developmental hematotoxic effect associated with the 0.4
mg/kg/day dose (Keller et al., 1986), based mainly on its lexicological significance and the
known hematotoxic potential of this compound, as well as its being a relatively sensitive
indicator of benzene's biological activity, (3) toluene, open field neurobehavioral no effect
level of 1.4 mg/kg/day (Kostas and Hutchins, 1985), based on the appropriateness of the
neurobehavioral test, the sensitivity of the test, and from findings of toluene-induced
neurobehavioral effects in other animal studies and in epidemiological studies, and (4)
xylene, reproductive/fetotoxic effects associated with a dose of 17 mg/kg/day (Ungvary,
1980), based on the lexicological significance and sensitivity of the effect. The non-cancer
reference doses for gasoline and benzene, toluene and xylene are summarized in Table 11-
4. Reference oral doses have also been calculated for the sensitive health endpoints based
on ingestion of 2 liters of water per day.
11-21
-------
TABLE 11-4
NON-CANCER REFERENCE DOSES FOR GASOLINE AND SELECTED
INDICATOR CONSTITUENTS
substance
(toxic
endpoint)
Reference
air levels
(ug/m3)
adult infant
Reference
dose*
(mg/kg/d)
Exposure
Interval
Reference
oral dose**
(mg/L)
gasoline
(kidney effects)
15
11
0.003
Subchronic 0.10
benzene 19
(developmental effects)
0.004
Subacute
0.10
toluene
(neurotoxicity)
68 52 0.0014
Subchronic 0.05
xylene
(reproductive effects)
165
0.034
Subacute
1.2
* based upon assumed weights and pulmonary ventilation rates (when applicable) as
follows:
mouse: 0.025 kg, 0.029 cu M/day
monkey: 5 kg, 1.7 cu M/day
rat: 0.25 kg, 0.14 cu M/day
human: 70 kg, 21.6 cu M/day (pulmonary)
14.4 cu M/day (systemic)
infant: 10 kg; 4.0 cu M/day (pulmonary)
2.7 cu M/day (systemic)
**
Oral references doses based on inhalation doses and consumption of 2 L water/day.
11-22
-------
11.3 RISK CHARACTERIZATION
11.3.1 Quantitative and Qualitative Assessment of Carcinogenic Risk
Potential incremental lifetime (70-year) cancer risks associated with exposure to
unleaded gasoline and selected constituents are set forth in Table 11-5. As the table
indicates, risks may significantly exceed the one-per-million risk benchmark under all
exposure scenarios. Nearly all of the tabulated cancer risks would be universally regarded
as unacceptable by regulatory agencies in the U.S. and other countries. It is, therefore,
appropriate to carefully consider the uncertainties associated with the estimates, as assessed
later in this chapter.
The cancer potency of gasoline and benzene may be placed in the context of other
substances whose carcinogenic potencies have been determined by U.S. EPA Carcinogen
Assessment Group (CAG). Such a comparison is made in Table 11-6, which sets forth
relative carcinogenic potencies of 59 chemicals evaluated by CAG. Table 11-6 also
presents classifications of each substance with respect to the quantity and quality of
evidence available for estimating its carcinogenic potency. The classification codes are
defined in Table 11-7. A comparison of the cancer risk associated with unleaded gasoline
vapor (0.0035 per mg/kg/d) with other cancer risks set forth in Table 11-6 reveals that only
one of the 59 tabulated chemicals exhibits a lower cancer risk: 3-chloropropene (synonym,
allyl chloride), 0.00047 per mg/kg/d. Consistent with this comparison, the cancer risk
associated with benzene, which may constitute about 1.8 percent of the volume of unleaded
gasoline, was assigned a risk value of 0.026, which is higher than the gasoline value by a
factor of 8.3.
It is appropriate to inquire about the source of the difference between the potency
estimates applicable to benzene vs. gasoline. If benzene constitutes the only carcinogen
present in gasoline, in which its concentration is about 1.8 percent, then gasoline would be
expected to exhibit a potency of 1.8 percent of that of benzene, or 0.000522 per mg/kg/d.
In fact, its estimated potency of 0.0035 is 6.7 times higher than the expected value. The
difference may be attributable to the presence of higher benzene concentrations in gasoline
vapor, to imprecision of the potency estimates, to the presence of additional carcinogens
which account for a predominant fraction of the total cancer potency of unleaded gasoline,
or to some combination of these.
It is important to illustrate the fact that a relatively low cancer potency need not
correlate with a relatively low cancer risk. Health risks arise from the presence of two
necessary conditions: potency and exposure. A very high potency, such as that of the
chlorinated dioxins, may pose only a low risk if exposure potential is low. Conversely, if
11-23
-------
TABLE 11-5
KJ
POTENTIAL CANCER RISKS ASSOCIATED WITH EXPOSURE TO
GASOLINE AND SELECTED INDICATOR CONSTITUENTS
Exposure Scenario Estimated exposure3 (mg/kg/day)
mean maximum
scenario 1: self-service customer
gasoline
benzene
toluene
xylenes
scenario 2: gi
gasoline
benzene
toluene
xylenes
9.4
7.3
5.7
2.2
X
X
X
X
as station attendant
1.8
2.1
3.8
1.5
X
X
X
at gas station
10-3
10-5
10-5
10-5
exposed
10-2
10-2
ID'2
1.0 x
7.2 x
4.9 x
2.6 x
Cancer Potency D
(oer me/ke/d)
Estimated lifetime cancer risk0
mean maximum
exposed via inhalation 1*3
10- !
10-4
10-4
10-4
0.0035
0.026
_
-
3.3 x
1.9 x
—
-
ID'5
10'6
3.5 x ID'4
1.9x lO'5
.
-
via inhalation1'-*
.
1.4 x
.
-
10-1
0.0035
0.026
_
.
6.3 x
5.5 x
m
-
10-3
10-4
_
3.6 x 10-3
_
.
scenario 3: resident living downwind of gas station exposed via inhalation1^
gasoline 3.1 x 10'3 1.6 xlO'2 0.0035
benzene 2.6 xlO'5 1.1 x 10'4 0.026
toluene 6.2 x 10'5 2.9 x 10'4
xylenes 2.7 xlO'5 UxlO'4
scenario 4: resident inhaling vapors from nearby leaking underground storage tank1*4
gasoline - - 0.0035
benzene 3.6X10'1 1.9 0.026
toluene 6.2 xlO'1 5.9
xylenes 4.2X10'1 3.6
scenario 5: resident exposed to gasoline via ingestion of contaminated well water2'4
gasoline 1.7 xlO'1 2.9 0.0035
benzene 1.4x10-2 7.0 xlO'2 0.026
toluene 8.0 x 10"3 5.0 x 10'2
xylenes 8.6 x 10'3 4.0 x 10'2
1.1 x 10'5 5.6 xlO'5
6.8 x lO'7 2.9 x 10'6
9.4 x JO'3 4.9 x ID'2
6.0xlO-4 l.OxlO-2
3.6 x ID'4 1.8 x ID'3
-------
TABLE 11-5
(CONTINUED)
Exposure Scenario Estimated exposure Cancer Potency & estimated lifetime cancer risk
mean maximum (per mp/kg/d) mean maximum
scenario 6: resident exposed via inhalation and dermal contact during showering*'*^
gasoline l.TxlO'1 3.4X10"1 0.0035 6.0 xlO'4 1.1 x 10'3
benzene 1.4xlO'2 2.8 x 10"2 0.026 3.6 xlO'4 7.3x10"*
toluene 8.0 x 10'3 1.6 xlO'2
xylenes 8.6 x 10'3 1.7 xlO'2
a refer to Chapter 5
b US EPA Cancer Potency Values
~ c estimated lifetime (70 years) cancer risk = (estimated exposure dose) x (assumed cancer risk)
K> ' assumes inhalation of 14.4 cu M/d. 24 h/d
**" 2 assumes ingesu'on of 2 L water/day
3 based upon arithmetic means of monitoring studies described in "Exposure Assessment"
4 based upon limited case-study information. Estimated risks for any given site need to be determined on a site-specific basis.
5 assumes mean values equal mean drinking water exposures, and upper limits equal twice drinking water maxima
-------
TABLE 11-6
RELATIVE CARCINOGENIC POTENCIES OF CHEMICALS EVALUATED
AS KNOWN ANIMAL CARCINOGENS, OR SUSPECTED OR KNOWN
HUMAN CARCINOGENS1
Substance
Metals
arsenic
beryllium (oxide)
beryllium (sulfate)
cadmium
chromium (CrVI)
nickel (refinery dust)
Exposure U.S. EPA Potency
route carcinogenicity index3
category2
oral
inhalation
inhalation
oral
inhalation
inhalation
inhalation
nickel (subsulfide) inhalation
Substituted alkane and alkene hydrocarbons
1,3-butadiene
methylene chloride
chloroform
carbon tetrachloride
1 ,2-dichloroe thane
1 , 1 ,2-trichloroethane
1 , 1 ,2,2-tetrachloroethane
hexachloroediane
acrylontrile
monochloroethylene
(vinyl chloride)
1 , 1 -dichloroethylene
trichloroethylene
tetrachloroethylene
ethylene dibromide
inhalation
inhalation
oral
oral
oral
oral
oral
oral
occupational
oral
inhalation
oral
oral
oral
A
A
B2
B2
Bl
A
A
A
B2
B2
B2
B2
B2
C
C
C
Bl
A
C
B2
B2
B2
7x102
2xl03
2X102
3x105
TxlO2
4xl03
2xl02
4X102
IxlO2
1x10°
IxlO1
2xlOJ
9x10°
8x10°
3x10°
3x100
IxlO1
IxlO2
IxlO2
IxlO2
8x10°
8x103
Slope
(risk per
mg/kg/d)4
15
50
7.0
3,000
6.1
41
0.84
1.7
1.8
0.014
0.081
0.130
0.091
0.0573
0.20
0.0142
0.24
1.16
2.3
0.011
0.051
41
11-26
-------
TABLE 11-6
(continued)
Substance
Exposure
route
category2
U.S. EPA
carcinogenicity
Potency
index^
Slope
(risk per
mg/kg/d)4
Substituted alkane and alkene hydrocarbons (continued)
ethylene oxide
3-chloropropene
(allyl chloride)
hexachlorobutadiene
unleaded gasoline
Aldehydes, ketones. esters.
formaldehyde
acetaldehyde
bis(chloromethyl)ether
bis(2-chloroethyl)ether
inhalation
oral
oral
oral
and ethers
inhalation
oral
inhalation
oral
Bl
B2
C
B2
A
B2
A
B2
2x101
4 x ID'2
2X101
4x10-!
1x10°
3 x 10'1
IxlO6
2xl02
0.35
0.00047
0.0775
0.0035
0.0455
0.0077
9,300
1.14
Aromatic and substituted aromatic hydrocarbons
benzene
hexachlorobenzene
2,4,6- trichlorophenol
benzidine
3,3-dichlorobenzidene
2,4-dinitrotoluene
diphenylhydrazine
epichlorohydrin
Nitrosamines
N-nitroso-dimethylamine
N-nitroso-diethylamine
N-nitroso-dibutylamine
N-nitroso-pyrrolidine
inhalation
oral
oral
occupational
oral
oral
oral
oral
oral
oral
oral
oral
A
B2
B2
A
B2
B2
B2
B2
B2
B2
B2
B2
2x10°
5xl02
4x10°
4X104
4xl02
6X101
IxlO2
9 x 10'1
2x103
4xl03
9xl02
2X102
0.029
1.67
0.0199
234
1.69
0.31
0.77
0.0099
25.9
43.5
5.43
2.13
11-27
-------
TABLE 11-6
(continued)
Substance
Nitrosames (continued)
N-nitroso-N-methylurea
N-nitroso-N-ethylurea
N-nitrosodiphenylamine
Cvclic pesticides and related
cc-hexachlorocyclohexane
B-hexachlorocyclohexane
y-hexachlorocyclohexane
(lindane)
aldrin
dieldrin
chlordane
heptachlor
heptachlor epoxide
toxphene
DDT
PCB
benzo(a)pyrene
coke oven emission
tetrachlorodibenzo-p-
dioxin (TCDD)
hexachlorodibenzo-p-
dioxin
Exposure
route
category2
oral
oral
oral
compounds
oral
oral
oral
oral
oral
oral
oral
oral
oral
oral
oral
oral
occupational
oral
oral
U.S. EPA
carcinogenicity
B2
B2
B2
B2
C
B2-C
B2
B2
B2
B2
B2
B2
B2
B2
B2
A
B2
B2
Potency
index3
SxlO4
4xl03
IxlOO
SxlO2
4X102
3xl02
6xl03
SxlO3
SxlO2
2xl03
4xl03
SxlO2
IxlO2
2xl03
SxlO3
inapplicable
SxlO7
2xl06
Slope
(risk per
mg/kg/d)4
302.6
32.9
0.00492
2.7
1.5
1.1
16
20
1.3
9.1
1.13
1.13
0.34
4.34
11.5
2.16
156,000
6,200
1 Adapted from U.S. EPA. 1987. Health assessment document for beryllium.
Washington, DC: U.S. Environmental Protection Agency.
2 See Table 11-7 for explanation of these standardized U.S. EPA codes.
3 Potency index is the slope rounded off and multiplied by the molecular weight.
4 Incremental lifetime cancer risk is calculated as the (95-percent upper-bound) slope of the
dose-response curve ([mg/kg/d]'1) multplied by the exposure dose (mg/kg/d).
11-28
-------
TABLE 11-7
QUALITATIVE CARCINOGENICITY CATEGORIES ESTABLISHED BY
THE U.S. ENVIRONMENTAL PROTECTION AGENCY*.*
Human evidence animal evidence fbioassays)
(epidemiology) sufficient limited inadequate no data no evidence
sufficient
limited
inadequate
no data
no evidence
A
Bl
B2
B2
B2
A
Bl
C
C
C
A
Bl
D
D
D
A
Bl
D
D
D
A
Bl
D
E
E
Adapted from U.S. EPA, 1987a.
^carcinogenicity categories:
A. human carcinogen
B. probably human carcinogen
C. possible human carcinogen
D. not classifiable as to human carcinogenicity
E. evidence of non-carcinogenicity for humans
11-29
-------
a substance exhibits a low cancer potency, it may nonetheless pose a high cancer risk if
potential exposure is high. In the case of gasoline, potential exposure, as quantified earlier
(see chapter 5), may be very high under certain exposure conditions.
The issue of risk acceptability encompasses two issues which are distinctly different
from risk comparisons. First, risks may be deemed acceptable if they are acceptably low,
as determined by societal values and ideally reflected in formulations of regulatory agency
policies, positions, and decisions. Second, risks-even unusually high risks-may be
deemed acceptable at least temporarily if they have been minimized relative to all feasible
alternatives. In such cases, agencies may appropriately attempt to drive technologies
toward improved performance with respect to health risk issues. In the case of risks
associated with gasoline an example of this principle is the reduction of the lead content of
in gasoline.
11.3.1.1 Cancer Risk Research Needs
The principal area in which further research is needed is that of cancer
epidemiology. Epidemiology is exceedingly weak at present because of the recent
introduction and dominance of unleaded automotive fuels. Although a significant number
of epidemiological investigations involving leaded gasoline have been conducted (see
chapter 10.1), investigations involving unleaded gasoline are incomplete with regard to
long-term follow-up of cohorts. Moreover, a significant if not a predominant fraction of
exposure of early cohorts to gasoline must have involved leaded gasoline. In studies
revealing (or eventually revealing) elevations of cancer mortality among cohorts exposed to
unleaded gasoline, confounding exposure to leaded fuels would probably preclude
attribution of the elevated mortality to unleaded gasoline. A primary value of conducting
further epidemiological studies is to refine quantitative estimates of cancer potency to
humans, an objective which can be accomplished even by epidemiological studies yielding
negative results, if they are well-designed. Such studies, if sufficiently sensitive, can
impose upper limits upon cancer risk to humans exposed under various relevant scenarios
of importance for public health policy formulation.
In view of intrinsic weaknesses of epidemiological studies, such as those described
above and others, it is appropriate to address the issue of whether and to what degree it
may be appropriate to also rely upon cancer bioassays involving non-human mammals.
Studies involving rodents should be accorded only low priority, given the strength of
evidence already available supporting the conclusion drawn in this assessment that gasoline
and at least one of its volumetrically important constituents (benzene) are known or
suspected human carcinogens. Indeed, given the strength of available evidence, studies
11-30
-------
designed to further test the hypothesis of human carcinogenic risk by any procedures other
than definitive epidemiological ones should be excluded from the arsenal of state agencies,
and relegated to voluntary activities to be undertaken by interested private parties for
academic or other reasons.
11.3.2 Quantitative and Qualitative Assessment of Non-Cancer Risks
The principal issue in quantifying non-cancer risks is to determine whether and to
what degree toxic doses exceed exposure doses under each of the scenarios of interest.
Consequently, in characterizing risks which have been quantified, the principal qualitative
issue of concern is to determine whether and to what degree the data forming the basis for
risk quantification are adequate. If they are not, it is likely that future research, if
conducted, will reveal lower thresholds of toxic action in connection with substances of
potential concern than the thresholds identified in the present assessment
Data pertaining to the lexicological properties of gasoline and/or of its constituents--
benzene, toluene, and xylene-appear to be adequate with respect to some but not all of the
threshold lexicological endpoints of interest Numerous studies have made it possible to
elucidate some acute, subacute, and subchronic lexicological endpoints in detail, with
notable exceptions, for example, in the areas of reproductive and developmental
toxicology.
A second observation of importance for characterizing non-cancer risks is that
effective dose levels would normally be expected to decline from acute to subchronic to
chronic regimens, inasmuch as the exposure duration lengthens. Review of Table 11-2
suggests that further testing or further data analysis might reveal lower effective dose levels
within some exposure categories. Indeed, in most cases the lowest lexicologically effective
chronic levels were higher than reported acute or subchronic levels, increasing the
importance of the short-term studies, even in the present context, which focuses primarily
upon long-term risks.
11.3.2.1 Non-Cancer Risk Comparisons
The large number of lexicological endpoints associated with non-cancer risks
precludes simple and objective comparison with familiar risks. However, it is possible to
compare the reference doses (summarized in Table 11-4) with estimated exposure doses to
determine whether a margin of safety exists between them. This has been done in Table
11-8, in which estimated doses under each exposure scenario are compared with the
reference doses for inhalation or oral exposure. Margins of safety associated with each
exposure scenario were calculated as the quotient of the reference dose/exposure dose.
11-31
-------
TABLE 11-8
POTENTIAL NON-CANCER RISKS ASSOCIATED WITH EXPOSURE TO
GASOLINE AND SELECTED INDICATOR CONSTITUENTS
Exposure Scenario
Estimated exposure3 (mg/kg/day)
Reference dose (RfD)b
margin of safely (RfD/exp. dose)
scenario 1:
gasoline
benzene
toluene
xylenes
scenario 2:
gasoline
benzene
toluene
xylenes
scenario 3:
gasoline
benzene
toluene
xylenes
scenario 4:
gasoline
benzene
toluene
xylenes
scenario 5:
gasoline
benzene
toluene
xylenes
self-service customer at gas station exposed
9.4xlO-3 1.0 x 10'1
7.3 x ID'5 7.2 x ID'4
5.7 x 10'5 4.9 x 10'4
2.2 x 10'5 2.6 x 10'4
gas station attendant exposed via inhalation*
1.8
2.1 x 1C'2 1.4 x 10- '
3.8 x ID'2
1.5 x ID'2
via inhalation'*3
0.003
0.004
0.0014
0.034
,3
0.003
0.004
0.0014
0.034
0.32
55
25
1545
0.002
0.19
0.034
2
0.03
5
3
131
.
0.03
-
resident living downwind of gas station exposed via inhalation1'3
3.1 x lO-3 1.6 x ID'2
2.6xlO-5 1.1 xlO'4
6.2 x ID'5 2.9 x lO'4
2.7 x 10'5 1.3 x 10'4
resident inhaling vapors from nearby leaking
.
3.6X10'1 1.9
6.2x10-' 5.9
4.2X10'1 3.6
resident exposed to gasoline via ingestion of
1.7X10'1 2.9
1.4x10-2 7.0x10-2
8.0 x ID'3 5.0 x ID'2
8.6 x ID'3 4.0 x ID'2
0.003
0.004
0.0014
0.034
underground storage tank*'4
0.003
0.004
0.0014
0.034
contaminated well waler^'4
0.003
0.004
0.0014
0.034
0.97
154
23
1260
_
0.01
0.002
0.08
0.018
0.29
0.18
4
0.19
36
5
262
.
0.002
0.0002
0.009
0.001
0.06
0.03
0.85
-------
TABLE 11-8
(continued)
Exposure Scenario Estimated exposure3 (mg/kg/day) Reference dose (RfD)b margin of safely (RfD/exp. dose)
mean maximum fmg/kg/d) mean maximum
scenario 6: resident exposed via inhalation and dermal contact during showering*»*>5
gasoline 1.7 xlO'1 3.4 xlO'1 0.003 0.018 0.009
benzene 1.4 xlO'2 2.8 xlO'2 0.004 0.29 0.14
toluene 8.0 x 10'3 1.6 xlO'2 0.0014 0.18 0.09
xylenes 8.6 x 10'3 1.7 x 10'2 0.034 4 2
a refer to Chapter 5
b refer to Table 11 -4
£ 1 assumes inhalation of 14.4 cu M/d, 24 h/d
w 2 assumes ingestion of 2 L water/day
3 based upon arithmetic means of monitoring studies described in "Exposure Assessment"
4 based upon limited case-study information. Estimated risks for any given site need to be determined on a site-specific basis.
' assumes mean values equal mean drinking water exposures, and upper limits equal twice drinking water maxima
-------
When this quotient was more than or equal to unity, a margin of safety is indicated,
whereas lower values denotes the absence of a margin of safety. Exposures may exceed
the reference doses, as indicated in Table 11-8 by tabulated factors less than unity occurring
under all six exposure scenarios.
11.3.2.2 Non-Cancer Risk Research Needs
As suggested by Table 11-2, data relating to acute, subchronic, and chronic
exposure regimens are incomplete. However, threshold doses likely to elicit non-cancer
toxic effects appear to be generally higher than doses (of benzene or gasoline) likely to
significantly elevate cancer incidence. Hence, further research into these numerous non-
cancer toxic effects, though undoubtedly of academic interest, presently appears to merit
low priority as a means of resolving the issue of public health policy formulation with
regard to gasoline exposure control and risk management One exception exists in the area
of genotoxicity, owing to its relevance to cancer causation. State agencies contemplating
research into these other areas may wish to defer to interested academic or other parties.
11.4 UNCERTAINTIES
Introduction
The state of the an of toxicology is presently insufficiently advanced to permit
complete and accurate prediction of potential risks posed by human exposure to a particular
toxic substances. Contributing to this insufficiency are several specific areas of
uncertainty. These include uncertainty about the identities of each substance to which
exposure would be of potential concern (a qualitative insufficiency), and about the absolute
and relative abundances of each substance in gasoline (a quantitative concern). Moreover,
epidemiological data pertaining to both carcinogens and non-carcinogens are typically
uncertain as to exposure durations and levels. Likewise, data pertaining to both
carcinogens and non-carcinogens derived from animal studies, though relatively more
complete, are subject to uncertainty as to their applicability (quantitatively and qualitatively)
to humans. Further, humans may vary in their sensitivity to toxic substances in gasoline
and, finally, the health effects exerted by the toxic substances may differ when multiple
substances are present These issues are discussed sequentially in the subsections that
follow.
11-34
-------
Uncertainties Associated with Exclusion of Substances of Potential Concern in Gasoline
It is possible that gasoline contains substances of potential concern which are
unidentified and omitted from consideration in the present assessment. These substances
may pose significant risks to public health that have been overlooked herein. Such
unconsidered risks may arise from either the invariant components of gasoline other than
benzene, toluene, and xylenes; or with the variability of its composition, depending upon
producer, climatic zone, and season. For example, a hypothesis has been developed which
postulates that the male rat kidney response is associated with exposure with
trimethylpentane. The importance of TMP as a contributor to overall cancer risks posed to
humans by potential exposure to gasoline has not been made. This may be resolved by
bioassay testing of gasoline with and without trimethylpentane.
Uncertainties Associated with Exposure Intensities. Durations, and Pathways
Inaccuracies in assumptions about intensities and durations t
It is possible that this risk assessment may have adopted inaccurate assumptions
about exposure of reference individuals to substances of potential concern in gasoline. Six
exposure scenarios were considered, of which three related to gasoline service station
emissions and three to leaking underground storage tanks. In the latter category, exposure
intensities may vary continuously, from as little as a single molecule of an indicator
substance to as much as the taste threshold in drinking water or the odor threshold in air.
Above these levels, protracted exposure is unlikely to occur, inasmuch as individuals
experiencing such exposures would presumably take appropriate actions to quickly
terminate them. It is important to emphasize that the significance of this uncertainty is not
to invalidate or weaken the assessment as a basis for policy formulation, but only to qualify
its conclusions about risk to reflect the deviations of actual gasoline exposure cases from
the exposure scenarios analyzed in this assessment
Redundancy of multiple pathway exposure assessment.
The CAG cancer potency factor applicable to benzene was based upon
epidemiological data originating from occupational settings. Typically, in such studies,
exposure of employees is estimated based upon air levels measured or inferred at the work
place. However, employee exposures do not only occur via inhalation of substances
suspended in ambient air. Employee exposure occurs via multiple pathways, just as it does
in residential exposures to gasoline in groundwater and air. Thus, many if not most CAG
unit risk values intrinsically consider alternative modes of exposure.
11-35
-------
Uncertainties Associated with Potential Non-Cancer Risks
Non-carcinogens
One way of accommodating inevitable uncertainty in risk assessment is to compare
expected exposure or dose levels with dosages known to be lexicologically safe. This has
been done with respect to non-carcinogenic substances of potential concern by comparing
reference doses with exposure doses (EDs) of the substances, conservatively assuming full
absorption.
Carcinogens
Usually the most critical toxic effect of carcinogens is considered to be
carcinogenesis. This is because carcinogenesis is assumed to be associated with a finite
probability of occurring at any dose above zero, whereas most if not all other toxic effects
are associated with threshold doses below which they have zero probability of occurring.
However, carcinogens may also exert non-carcinogenic toxic effects, as has been
extensively documented earlier with gasoline and benzene. Uncertainties associated with
these non-carcinogenic effects are qualitatively the same as those described with regard to
non-carcinogens.
Uncertainties Associated with Potential Cancer Risks
Significant uncertainties are associated with quantifying cancer risks to humans
based upon data from microbial bioassays, animal bioassays, and epidemiological studies.
These uncertainties necessitate adoption of conservative assumptions, so that errors, when
made, are made on the side of caution where protection of public health is concerned. One
conservative assumption relates to the cancer potency value itself, specifically it is assumed
in this risk assessment that, at low, environmentally realistic doses, each carcinogen
exhibits a potency equal to the 95-percent upper bound value indicated by its dose-response
curve in the low-dose range. This curve is obtained by use of the conservative, non-
threshold linear multistage (LMS) model, which tends to overestimate cancer potency in the
low-dose range.1
1 The LMS model has, however, recently been shown to underestimate cancer
risk in some cases (Bailar, et al., 1989).
11-36
-------
Uncertainties Associated with the Significance to Humans of Rodent Liver Tumors
Some evidence of carcinogenicity of gasoline constituents in animals consists of
data on induction of liver tumors in rodents. In particular, the chronic benzene inhalation
study involving mice and rats conducted by MacFarland et al. (1984) revealed statistically
significant elevations in the incidences of hepatocellular adenomas and carcinomas in
female mice. The significance of such data in the context of human risk has been
controversial for reasons relating to the observation that some laboratory strains of rodents
tend to exhibit relatively high spontaneous (background) incidences of liver tumors (see,
for example, Trump et al., 1984). Thus, breeding may have produced rodent strains that
have already incorporated significant precancerous changes that, in other organisms, would
require exposure to a carcinogen. Consequently, a variety of non-specific stresses (e.g.,
cytotoxicity, cell proliferation) could augment a tumorigenic response in these sensitive
animal strains, but not in other strains which lack these precancerous changes (Fox and
Watanabe, 1985). Should a suspected non-genotoxic carcinogen only be associated with
non-specific effects, the significance of the response to human health may be questioned;
the nature of the dose-response relationship (particularly at doses well below the
experimental range) may be different than that of other carcinogens. Differences may arise
given the presumption that a threshold would have to be reached before the chemical could
contribute to the development of the tumor, and then, only in certain pre-disposed
individuals.
These concerns must be counterbalanced by the low sensitivities of the animal
bioassays, and by the basic uncertainties regarding the role of chemically-induced
carcinogenesis in the human population. A positive bioassay response provides evidence
that a chemical contributes to tumorigenicity by participation in one or more critical stages
of disease development Identification of one possible mechanism by which chemical
exposure may influence tumor development (e.g., through increased cell proliferation) does
not summarily disqualify other mechanisms from consideration. Moreover, negative
findings relative to a particular aspect of carcinogenicity may be of limited value unless it
can be demonstrated that the tests employed are relevant to the chemical's mechanism of
action.
The interpretation of such correlative evidence is especially relevant with regard to
genetic toxicity. Lack of activity in a battery of genetic toxicity tests does not necessarily
mean that the chemical acts through an epigenetic mechanism, as the chemical could alter
DNA in a way that the tests do not measure. The findings of Reynolds et al. (1987)
support this concern. These investigators found that two compounds (furan and furfural)
which are negative in Salmonella assays and positive in the B6C3F1 mouse liver tumor
11-37
-------
assay did, in fact, give rise to a different spectrum of activating mutations when compared
with tumors in the untreated animals. Other carcinogens not found to be mutagenic in the
Ames assay (e.g., formaldehyde, ethylenethiourea, methylene chloride) have, nonetheless,
produced genomic rearrangements when assayed in other test systems (Schiestl, 1989).
While the dose-response characteristics of these various DNA lesions may vary, these
assays do provide evidence that the carcinogenic activity of chemicals heretofore recognized
as non-mutagenic (such as gasoline) may involve a combination of genetic and non-genetic
events.
Uncertainties Associated with Svnerpism and Interference
Predictions of cancer and non-cancer risks posed by gasoline are subject to
uncertainties associated with potential interactions among gasoline constituents to which
individuals are simultaneously or sequentially exposed. The number of potential
interactions among mixed substances increases exponentially with the number of
substances in the mixture.
The term 'synergism' applies to the general phenomenon of interactive effects that
enhance lexicological potency, whereas the term 'interference1 applies specifically to
interactions in which multiple agents partially or completely cancel lexicological potency.
Although rationales supporting each view have been formulated, so little is known that it is
not possible to reliably predict whether lexicological enhancement or interference-or
neither of these--is likely to predominate in a complex mixture. Currently, the U.S. EPA
policy pertaining to risk assessment involving complex mixtures consists of adopting the
neutral assumption of additivity of effect, that is, the net absence of either lexicological
enhancement or interference (U.S. EPA, 1987a). Given the limitations of the present state
of the an in toxicology, we must continue to accept uncertainties associated with assessing
risks posed by complex mixtures. These and other uncertainties illustrate the prudence of
maintaining a commitment to worst-case risk assessment.
Uncertainties Associated with Sensitive Subpopulations
Toxic substances are known to discriminate against some unusually sensitive, or
hypersensitive, individuals who may be exposed to them. This discrimination may be
based upon genetic predisposition, or upon the physiological or even the nutritional state of
an individual at the time of exposure. The presence of unusually sensitive individuals
among the potentially exposed population, and their degree of variation in sensitivity
compared with the general population, both constitute uncertainties associated with the
present risk assessment. Unusually sensitive individuals could include pregnant women,
11-38
-------
the very young, the very old, or the infirm, as well as individuals who may suffer from
chronic respiratory, immunological, or other predisposing illnesses. For example,
individuals suffering from autoimmune diseases such as lupus erythematosis may be
hypersensitive. That is, their immunological systems may react to significantly lower
concentrations of chemical insults than do the immunological systems of other members of
the general public. In addition, individuals suffering from chronic respiratory illnesses,
including those with allergy, asthma, or emphysema, may exhibit unusually high
sensitivities to chemicals present in gasoline.
With specific reference to benzene exposures, limited data are available regarding
host factor differences which modify the toxic response. There is some indication that
benzene toxicity may decrease with age (Doskin, 1971), a finding that is supported by
animal studies which investigated the effects of benzene exposure on the developing
immune system. Other epidemiological studies, however, do not suppon this finding at
least with regard to adolescent and adult populations (Aoyama, 1980; Goldstein, 1977). In
general, however, benzene exposures are most likely to present a concern for the very
young and the very old, whose immune systems are not functioning optimally. Animal
studies indicate that females may be more sensitive than males to the cytopenic effects of
benzene (Deichmann et al., 1963), a finding that has limited suppon from epidemiological
studies (Goldstein, 1977). Pregnant women, whose hematopoietic systems are naturally
under stress may be a particular risk with regard to benzene exposure. On the other hand,
cytological studies indicate that males may be more sensitive than females to the clastogenic
effects of benzene (Salamanca-Gomez et al., 1989)
There is also evidence of a possible hormonal influence on benzene toxicity
(Goldstein, 1977). The effects of benzene exposure may also be more severe among
individuals with some underlying physiological impairment. Thalassemia, and its
associated bone marrow hyperactiviry, may render patients more susceptible to the harmful
effects of benzene (Goldstein, 1977). Malnutrition, by weakening an individual's immune
system, may potentiate benzene toxicity. In addition, compounds which alter benzene
metabolism (such as alcohol and other drugs, or other industrial chemicals) may affect
benzene toxicity by modifying blood levels of toxic benzene metabolites (Goldstein, 1977).
Benzene risks may also be enhanced through co-exposure to other myelotoxic agents (such
as radiation, metals, halogenated hydrocarbons, and pesticides) (Goldstein, 1977; Luster et
al., 1987).
11-39
-------
APPENDIX A
SOLUBILITIES OF THE PRINCIPLE COMPONENTS OF GASOLINE
A-l
-------
A-2
-------
SQIUBUITIIS Of GASOLIHC COHPONlNTS
Component
StnUht -Chain Paraffins
propane
n-butane
n-pentane
n-hOKane
n-haptane
n-oclano
n-decano
n-nonane
n-undecane
n-dodecane
Boiling Point
foraulo Boiling Order1 °f
CiN, 1 -41.71 62
C«H|| 6 11.10 61
CtMti 16 96.91 10
47
6S
19
19
CtHM >9 ISS.71 9.
11
16
9.
CiHi. 9S 209.11 1.
1.
4.
1.
C(H|a ISO ISO. 20 0.
0.
1.
0.
CieHu 110 14S.42 0.
0.
C»Hto I" 101.44 0.
Solubility*
.4 t
•«• *
•* *
• 6 *
•' <
.* I
.SO
S t
4 ,
• i t
47 ,
91 t
» S
19 ,
24 ,
66 ,
•S >
1* >
411
022
020
122
1.1 9/10*0 H,0 *
1.6 9/IO*a MjO *
1.0 9/10*9 NtO *
O.S 9/10*9 M*0 *
1.1 9/10*9 N«0 at 0°C *
0.6 a/IO*| H,0 *
» 0.0 g/101 NaO *
1 . 1 e/'l04a, NtO *
0.1 9/10*9 HaO *
0.1 9/IO*| HjO at 0°C 4
0.10 9/10*9 HjO *
.10 g/IO* N,0 *
.11 g/IO* H)0 *
.10 g/IO* N,0 at 0°C *
.04 g/IO* H,0 *
.06 g/IO* NiO *
.01 g/IO* NjO 4
.OS |/IO* NjO Ot 0°C *
t 0.012 0/10*9 HjO k
9/IO*i NjO »
g i 0.00} g/to'g H,0 *
x 0.007 g/IO'a H,0 *
CnHj, 164 104.60 Not found
C)aHa( 176 421.10 0.
0.
0.
0017
0029
In
000
, 0.0006 9/10*9 MjO '
t 0.000& |/IO*g M»0
sea water. '
t 0.001 9/10*9 HaO *
Boiling Order It o listing of tho hydrocarbon* In the order of boiling point.
Solubllltlci are at 2S«C unUss othorMlso stated. Also for dilute solutions 9/10*9
HcAultHt (I966t.
folak and lu (1971).
Price (1976).
KriyianoMika and Saellga M97B).
Btkcr
rranht
Suiion inn Calder 11971).
-------
smuauiiirs or GASOUNC COHPONINIS
Component
BTJKtic«Lf*rirtiiu
tiobutono 1 1 -Ml hyl propone)
I.I-dlealhjrlproptne (neopenUne)
tiopenlono (I-Mlhylbutine)
a.)-dtMlhrtbuUne
>.)-dlMlMbulono
i.l.l-trlMthylbultno
I-Mthylpentine (liohextnc)
1-Mlhylpenteno
I.t-dlMthylpenttno
1.4-dtMltiylBfnUno
Belllno Point
roraulo Balllno Order1 °» Solubility1
C«H|. 1 10.19 40.9 t 2.1 0/10'c H,0 '
C*Nii • 49.10 11. t , 1.0 o/lo"o H,0 *
C|Hu II 01. II 47. ( 1. 0/10*0 H|0
4». , 0. o/IO*o H,0
40. , 1. t/IO*o H,0
71. t 1. ft/10** H|0 t 0°C 4
C»HM ii in. si 10. t i. o/io'o N|0
11. t 0. 0/10*0 HiO
II. i 0. 0/10*0 h,0
19. ,0. 0/10*0 H»0 «i 0°C *
C«N|« 19 lit. 10 II. , 0.4 fl/lo'a HiO *
II. t °-*' 0/10*0 N|0 It 0°C 4
19. , O.I 0/10*0 H,0 *
C|M|» II III. S9 4.10 « 0.01 o/IO*0 H|0 *
C«NM 10 140.4* II. 0 , 0.9 o/lo'o H,0 *
It. 7 , O.I o/IO*0 H|0 '
11.4* « 0.11 0/10*0 N]0 It 0°C 4
11.0 i O.I i/lo'o M|0 '
C«HM 11 I4S. 91 II. 0 t 0.4 0/10*0 MiO *
17.9 . 0.1 o/IO*0 M(0 *
II. t , 0.4 0/10*0 H,0 ft 0°C * *
II. 1 i 0.4 0/10*0 H,0 *
C»H|« 4ft 174. SS 4.40 , 0.11 i/lo'g NjO *
C|H|« SO 17ft. 90 4.06 , 0.19 0/10*0 HjO *
4.40 , 0.09 o/IO*0 N|0 *
4.41 , O.OS 0/10*0 HjO §
6. SO , 0.11 fl/IO*o MjO Ot 0°C *
Botllno Qrdtr U • lining of lh« hydroorbons In order of bolltno point.
Solubtllllct »r« ftt Il'C witl«st otlwrwlt* ft(»t«4. Alto for dilute loluttons o/IO*o
HcAullfre (19441.
Polth and lu (19711.
Prtc« (I96t|.
-------
soiuauims or GASOLINE CONHUCNTS
Component
•ruiched tirarf Ins (Conttnuedt
I.l-dlMthylpentane
2.1-dlMthylpentene
S-ethylpent»ne
2.1.4-trlMthylpcnUn* (Isooctene)
2.I.l-trlMthy1pcnt»ne
2.1.4-trlMthylpenl«ne
2.1.1-trtMthylpentane
2-Mthyl-l-elhylpentene
l-Mthy1-l-ethylpent*ne
2.2-dlMthyl-l-olhylpenUne
2.4-tflMthyl-l-ethylpenUno
t-Mthy1he>une (IsonepUne)
M»1MMM.
1.4-dtMlhylhCKftne
2.1-dlMthylhexene
foraule
C|N,t
CiN.t
C»H,§
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
CiH,.
C.H,.
C.H it
Boiling Order1
62
70
01
01
100
IIS
116
120
127
162
171
60
71
124
119
Boiling Point
106.92
191.61
200.26
210.61
229.71
216.24
2U.S7
240.17
244.07
272.09
270.04
194.09
197.11
241.90
240.09
Solubility*
S.94 t 0.01 g/IO'g HjO *
S.2S t 0.02 g/IO'g HaO •
1.22 * 0.4S g/IO'g H,0 *
2.44 « 0.12 g/IO'g HjO *
2. OS a 0.00 g/IO*g M,0 •
2.46 x 0.10 g/IO*g H|0 »l 0°C '
1.14 * 0.02 g/IO*g HiO '
2.99 ,0.16 g/IO'g HjO *
2.10 i 0.09 .g/IO'g HjO '
2.14 « 0.09 g/IOftg H,0 (t 0°C '
1.16 i 0.01 g/lO*g HaO *
2.S9 » 0.16 g/IO'g H|0 *
I.4S t 0.20 g/IO*g HjO *
I.4S t 0.20 g/IO'g HjO 4
1.42 i 0.20 g/IO'g HjO 4
1.42 * 0.20 g/IO'g HjO 4
2.S4 i 0.02 g/IO'g HiO »
4.9s t o.oe g/io'g Hao *
S.24 , 0.09 g/l6*g MjO «l 0°C *
2.64 t 0.00 g/IO*g HaO *
I.4S t 0.20 g/!04g H,0 4
I.4S t 0.20 g/IO*g*HiO 4
Boiling Order Is • listing or the hydrocarbons In the order of balling point.
Solubilities «re *l 2S*C unless olnerotse steted. Also for dilute solutions g/IO'g «g/l.
•rice 11976).
fstlMted solubility using nowigreph; Kabedl end Denner (1979).
HcAultUe (I9bb|.
f>al»k «nd lu (1971).
-------
SOLUBiums or CASOLINC COHTONINIS
Component
B£AACtad_fArif£lfU (Continued)
1.1-dlMthylhciitne
1.4-dtMlhylhentne
I.I-dlMthy1he»ana
i.t-dt.*tMh«»M
1-tthylheMne
1.1.1-trlMthylheMtne
1.1.1-trUclhylhentne
1.1.4-trlMlhy1h«Htn«
I.4.«-trlMlhy1heMno
t.l.f-trlMthylhiMna
I.I.t-lrlMthylheiine
]-Mthyl-4-elhyllteieno
I-Mthyl-I-elhylheiane
1.1.1.1-totraMlliylheHane
|-o«thylheptina
4-Mtnylheatine
1,1-dtMlhylheatkne
foraula
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C,Ht.
C.M,.
C.H..
C|Hj,
C.M,.
C.M,.
C..H,,
C.H,.
C.M,.
C.M,.
Boiling Point
Boiling Order1 °f
111 lll.M
107 111.97
104 110.19
101 114.11
114 141.14
114 140.41
140
141 1S9.77
IS4 147.17
110 1SS.1S
111 179.01
111
177
111
111 141.74
111 141.00
171
Solubility1
I.4S , 0.10 9/10*8 N,0 >
I.4S t ••!• 0/l**0 H,0 *
I.4S , 0.10 g/IO(g N,0 *
I.4& t 0.10 f/IO*g HjO *
I.4S , 0.10 g/IO*g HjO '
1.41 , 0.10 g/IO*g H,0 '
1.41 t 0.10 g/10% HjO *
1.41 , 0.10 g/IO*g HjO *
1.41 , 0.10 g/10'0 H|0 '
I.IS t 0.00 g/10'0 NjO 4
O.S4 , 0.01 g/l»*g H|0 *
0.19 , O.OI a/IO*g HjO at 0°C '
1.11 t 0.10 g/io'g N|0 *
0.11 . O.Ot i/IO*g H,0 '
0.11 « O.Of g/10'g H,0 *
Not found
O.IS i 0.11 g/IO*g H,0 *
O.Oi , 0.11 g/IO*g H,0 *
a. ii t o.oi g/io'g Hto *
Order Is • listing of hydrocarbons In Ihi ordir of boiling point.
SolubtlUUs »ro «t 1»*C unless olh*rM«so stated. Also for dilute solutions g/IO(a
istlMled solubility using nawi«r*ph; Kabedt end Oitmer |I9>9|.
HcAutlffo (1946).
folth «nd lu (19)1).
-------
sOLuauincs or GASOLINE COMMMCNIS
Component
FonuU Boiling Order1
Boiling Point
Solubility*
|r inched Paraffins (Continued)
2.1-dlMthylhcptane
2.A-dtattthylheplane
2,4-dtMthylheptane
1.4-dlMthy Inept an*
I
1
2
•ij I
I
I
I
1
2
1
.»-dt«ethylheplene
.i-dlMlhylhepUno
,2.4-trtMthylheptane
.2,S-trtMthylhcplane
.I.i-trle*tnylnepUne
.S.I-lrtMthylheptan*
,4.4-trlMtnylneptane
.l.i-trlMthylhcptine
.•.t-trlMthylheptane
.1.4-trlMthylhoptano
1.4.4-trlMlhylheptane
1.4.S-trlMthylhcptane
4-otnylneptane
1-ethylheptanc
C.H,. 160
C.K!. IAS
C(Hi0 \m\
C.H,. tOI
C.H,. 100
C.H,, U9
CieH|j 192
Cio"n "*
CioH» IK
CioNai I01
Cio"n '02
Cie^Ji 206
C|oH]t 207
CieHjj 217
CioNn 210
C.oH.t 219
C.H,. 104
C.H,« 100
0.
27S.10 0.
0
,12
,12
.12
0.12
0.12
0
0
0
100.07 0
0
0
0
0
0
0
0
1
209.4 1
.12
.79
.79
.79
.79
.79
.79
.79
.79
.79
.79
.42
.42
s 0.
* o.
• o.
* o.
* o.
* o
* •
a 0
* o
• o
t 0
* o
* o
t 0
t 0
i 0
i 0
t 0
OS
OS
.OS
.OS
OS
.OS
.11
.11
.11
.11
.11
.11
.11
.11
.11
.11
.02
.02
g/lo*g
o/io*0
9/10*9
9/10 9
9/10*0
9/IO*g
0/10*0
0/10*9
9/10*9
0/10*0
g/IO*g
9/10*9
9/10*0
9/10*9
9/10*9
9/10*9
9/10*9
9/10*9
H,0
H>0
HtO
N,0
NjO
H,0
H,0
H,0
HjO
H,0
M,0
MjO
H,0
HtO
HaO
H,0
NjO
'«jO
*
>
'
1
1
I
1
»
»
I
I
1
1
I
*
1
*
1 Oolitic Order Is • IUHn9 o' hydrocarbon! In the order of boiling point.
* Solubilities «r* tt 2S«C unless other-lie sttted. Also for dilute solutions g/IO g eta/1,
* Cstletetcd solubility using noMgreph; Kebedt and Dinner (1979).
-------
SOIIWIUIIIS Or GASOUNC COHTONfurS
Coapoflcnl
BClOCllKl-firiff llU IConllnucdl
I.I-dtatthylhopUno
t-Mthylh«pUno
t,4-dtMthy1oct*no
4-othylocUno
«-Mthylocl»no
3-Mthylocttno
l-Mthyloclino
I-Mthy1non«no
)-Mthylnonino
>
oo 4-Mthy1nonin«
l-Mlhylnontno
Cyclopontino
cyclohiMtno
Mthylcyclopcnlino
cyclohopUno
,.«,,,
C,H,.
C.H..
C|«N||
C|«H||
C.H..
C,H,.
C,M,.
C|*Hj|
CigHji
C|(H||
C|H,.
C|N,,
C.H.,
C|N|4
Bolllna Ordir1
IS!
124
211
214
IIS
114
II*
224
226
222
221
21
41
"
I.S
Boiling Point
*r Solubility*
1.42
0.791
0.04
0.011
210.14 O.IIS
219.01 1.42
2*1. SI 1.42
112.41 0.011
114. • 0.022
110.1 1.022
119.2 0.022
110.41 IS4 |
140.0
111.11 SS t
44. S
Sl.S
141.14 41 .
41.1
24S.I1 10 t
• 0.01 8/10*8 H,0 *
t 0.020 8/10*8 N|0 *
i 0.01 8/10*8 M,0 *
t 0.001 8/10*8 H,0 *
1 0.011 8/10*8 H,0 *
* 0.02 8/10*8 H,0 *
* 0.02 8/10*8 H,0 >
t 0.01 8/10*0 Had *
1 0.01 0/10*8 M,0 *
t 0.11 0/11*0 H,0 *
, 0.01 0/10*0 NjO *
* 8/10*0 HiO *
t 1.0 0/10*0 H|0 *
2.1 0/10*0 M,0 *
, 0.0 0/10*0 H,0 *
, 1.1 0/10*0 M,0 4
1.4 b/IO*b H,0 *
, 1.0 8/11*0 H|0 *
1.0 0/10*0 N«0 '
Ordtr Is o listing of tho hydrocarbons In ordor of boiling point.
Solubllttlts «ro a 2S*C unloss olhtrwlso sUUd. Also for dllulo solutions 8/10*8
CstlMtod solubility uslna nonagraph; Kobad! »nd 0«nntr M»l»|.
Prlco |l«16|.
HcAullff* IH64».
H*Ck*r •••« Slilu
•a/I.
-------
SOLUIILITIES Of CASOUM COHPOHWS
Component
Poraule
Order*
Solubility*
(ContinuedI
lhylcrclohexine
C,HM
101
Ill.it
M.O , I.I g/IO€B N|0 *
16.0 t 0.2 0/10*0 H,0 «
olhylcyclopenUne
n-pr opy 1 1 yc 1 openl MM
ethylcyclohtKtne
tiopropylcyctopentene
Ifobutytcycloptnlene
Isopropylcyclohtitne
propyUyclohtMtne
tulylcyc1openl»ne
liobutylcyctoheitne
l-butylcyclonoNUi*
loC-butylcyctohtMent
butylcyclohtiitni
pentylcyclopenltno
pentyUydohtKtne
C.MM IOS
C(N|i 166
C|M|| 167
»«H|» m
c*riit w
C.rt,i 200
C,d,, 210
C.H.. 211
C | o)Hjg 21)
C io)H|0i 114
C i0H|dj 2*K
'•iojH|(> 241
C|fH|t 249
C i |Hj) 274
210.24
267.71
269.21
2S9.SS
190.11
110.17
114.10
111.00
140.121
140.116
U4 . 740
1S7.70
24S 0/10*0 H,0 • 7I°C '
l». 0 t 1.7 0/10*0 H,0 *
7.04 « 0.10 0/10*0 H,0 *
1.19 t 0.46 0/10*0 HiO '
1.19 t 0.46 0/10*0 NjO *
0.6ft « 0.09 0/10*0 HiO *
0.6S i 0.09 0/10*0 HtO '
0.61 t 0.09 0/10*0 H|0 '
0.61 t 0.09 0/10*0 HjO '
0.10 t 0.01 0/10*0 N)0 *
•.10 t 0-01 0/10*0 H,0 *
0.10 , 0.01 0/10*0 HjO *
0.10 « 0.01 0/10*0 H»0 *
O.lli , 0.011 fl/io't M,0 *
Hal round
1 •ailing Ordtr Is • llstlni of hydrocarbons In tht order of bollIn* point.
1 Solubilities »ro tt H*C unless othtrMli* aUied. Also for dilute solution* t/lo*g
* HcAulUre (I9&6).
4 frlco |I976|.
1 GulOvl 11944).
* ItttMted tolubtllty using now>gr«ph; Kabidl »nd Dinner |I979).
•0/1.
-------
soiuaitnifs or CAMLIMC COMPONCNTS
Component
l-butene
1 -pantone
1-hoMno
t-hopttno
l-octona
1 -nonene
l-daceno
Olcflns (oih«r 'h«n i-oi«finsi
, trans-I-butene
o
cls-2-butcne
Irons -2-penlene
Cls-l-p«nUna
trans -l-heicno
cts-7-hCHcne
cls-1-hoMne
trans-1-tienene
clt-I-heplent
trans -l-heplene
foraula
C,H.
C»H,a
C«N,|
C,NM
C.M..
C,*,.
C..H,.
C«H.
C,H.
C|N,.
CiN.a
CtHii
C.H.,
C.HM
C»H,,
C|H io)
C |H |«)
Boiling Order1
1
11
U
70
111
197
III
,
•
17
II
17
41
14
IS
97
91
•olllii8 'otnt
•f
10.71
IS. 94
140.17
200. S4
ISO. 10
m.ii
11. SI
11.70
97.44
91. SO
114.19
ISa.OO
ISI.it
1S1.76
209.14
100.11
Solubility1
III « 10 8/10*8 H,0 '
140 , 7 8/10*8 H,0 *
SO t I.I 8/10*8 HjO *
14.1 , 2.0 8/10*8 H|0 *
2.7 . O.I 8/10*8 MjO *
O.il , 0.09 8/10*8 H,0 *
0.10 , 0.01 8/10*8 H|0 *
410 8/10*8 H|0 at 20°C '
410 8/10*8 H|0 at IO°C '
101 f 1 8/10*8 H|0 '
201 i I 8/10*8 HjO *
SS.9 » 7.1 8/10*8 N,0 *
SS.9 ( 7.1 g/IO*a N|0 4
SS.9 t I.I 8/10*8 H,O *
SS.9 t 1.1 8/10*8 N,0 *
IS i 1.4 8/10*8 H|0 *
IS a 1.4 8/10*8 HjO *
1 tolling Or4«r tf a Hit Ing of hydrocarbon) In tht order of bolllna point.
* SolwbllUlts ir« at 2S*C unless othtrwlto SUltd. Also for dilute solutions |/IO*8 «8/l.
1 McAullffl (I9U).
4 fltlMted solubility using noaograph; Kabadl and Dinner 11979).
* tandolt-Bornslaln (19611; solubility as I-tutenc.
-------
soLuaiiniis OF GASOLINE COMPONENTS
Component
Formila
Boiling Order1
Boiling
Point
Solubility1
OliMfli father than 1-Qleflnsl IConllnued)
cts-1-heptene
trans-1-heptene
cls-2-octene
trtns-2-octcne
tr*ns-4-octenc
cts-2-decene
trens-2-decene
Branched OleMni
2-«ethyl-l-butcne
tsobutene (2 -Methyl propcnt)
1-Mthyl-l-butene
2-«ethyl-2-butene
2-euthyl-l-pentent
4-Mthyl-l-pcntene
I.l-dlMlhyl-l-butene
I.l-dtMthyl-l-butcne
I.l-dlMthyl-2-butene
l.l.l-trlewthyl-l-butene
C|H,4
C,MM
C.N,.
C|H)(
C.M,.
C toHjo
C loHjo
CiM,.
C,H.
C|N|0
C|H|.
C.M,,
C.M,,
C.H.i
C«Hii
C.Hu
C>H,«
OS
04
IS2
149
110
219
217
14
4
10
19
11
24
20
26
47
40
204.
204.
2SO.
2S7.
2S2.
00
19
60
101
141
120
106
112
161
172
IS
21
IS
0
OS
.09
.SO
.11
.42
.10
.96
.2S
.11
.77
.20
IS. 2
IS. 2
t 2.
s 2.
1.4 t O.S
S.4 :
1.4 :
0.09
0.09
ISS
261
110
ISS
>• t
40 t
Not
Not
Not
Not
1 O.S
t 0-S
t 0.
t 0.
s "
* 21
s "
t 22
I.I
2.6
1 g/IO*g HjO >
1 g/IO(g M,0 *
0/10*0
g/10«g
0/10*0
NiO '
H,0 »
H,0 »
01 g/IO* H,0 '
01 g/IO
o/fo*.
g/IO*g
0/10*0
g/IO*g
0/<0*B
g/IO*g
* H,0 •
H,0 '
HjO 4
N,0 *
H.O »
H,0 4
M,0«
Found
Found
Found
Found
1 Boiling Order Is • lifting of hyrdocirbont In the order of boiling point.
1 Solubilities «re at 2S»C unless otherwise steted. Also For dilute solutions g/io g «g/l.
1 fstlMtcd solubility using nonogrtph; Kabcdl end Oanncr (1979).
4 HcAullffe (1966).
-------
SOLueiuTICS or GASOUNC COHPONIMTS
Component
•ranched Oleflns (Continued)
l-Mthyl-l-pgnteng
•.-•ethyl-clS-2-pentene
iHMthyl -trins-2-pintene
l-Mlhyl-cls-t-pcntong
)-«ethyl-trins-l-pantene
4.4-dtMthyl-l-pentene
I.I-dlMthyl - 1 -pentene
«.4-dlMthyl-clt-2-pcntene
«.« -dimethyl -trans -l-penleng
).4-dt«Mthy1-ctS-2-penlcne
}.4-dtMtltyl-trans-2-pantcne
t.t-dtMthyl-l pcntene
2.l-dlMthyl-2-pentene
2 .4-dlMthyl - 1 -pintcni
I.«-dlacthyl-I-pcnlcni
I-«lhy1-l-panltn«
l-athyl-l-pinltni
l-«lhy1-!-ptnt«n«
rorvula
C«H,,
C.H,,
C.H,,
C.H,.
C.H,,
CiH,«
C.H.,
C|H,«
C|H,«
C|H,«
C|H|«
C>N,«
C|N,«
CiHi,
CiM,.
C|N,«
C,H,,
CiN,,
•outfit Point
•olttni Order* °»
21 129. St
2S III. SO
}• III. SO
)• ISI.I7
42 !»•.?»
40 U2.SI
41
II
41
72
7*
S*
»» 297.12
II
S«
79
17
•0
Solubility1
S6 , • t/10'i HjO
S6 t 0 |/IO% H,0
56 « 0 8/IO'f NjO
S6 « • t/IO*« NjO
S4 * 0 i/IO's N«0
Not round
Not round
Not round
Not round
Not round
Not round
Not round
Not round
Not round
Not round
14 , 2 i/IO*« H,0
14 , 2 •/!••» H|0
14 , J t/«0*« H,0
1
I
•
>
•
1
1
I
1 Boiling Order Is • listing or hydrocarbons In the order or boiling point.
' Solubilities ere «t IS*C unless otherwise stetcd. Also for dilute solutions g/lo'g «g/l.
* IstlMted solubility using nomogreph; Kibedl and Otnner 11919).
-------
SOIUBILITKS OF GASOLINE COHPONCNTS
Component
•rtnthed Qleflna (Continued)
1,4,4-trlMthyl-l-pentene
2.4.4-trlMthyl-I-pentene
2 -•» thyl - 1 - heMcne
>-«tttiy)-l -he»ene
B-aethyl-l-hcKcne
fHMtnyl-2-neMene
t-Mlnyl-trtns-1-hciiene
iHMthyl-cts-l-hexene
l-Mthyl-cts-l-heaene
l--dlMthyl-lr»ns 1 iic-«ne
foreul*
C.H.,
CeKie
C,HM
C,HM
C»H,«
CiH,4
CtH,,
C.HM
C,MM
C,H,«
CfMu
C,M*4
CiH,«
C |H|%
C|H| j)
C |H|4
C.H,,
C.HI.
Boiling Point
Boiling Order1 *f
94 214. S9
100 120.14
IS
SI
61
•6
60
ss
•1
II
91
•2
6S
64
66
69.
110
19
Solubility1
Not Found
Hot Found
14 i 2 g/10'ff MjO '
14 i 2 g/IOCff NjO *
14 i 2 g/l«*g NjO '
14 t I g/IO*g H,0 *
14 * 2 g/IO*g HtO *
14 , 2 g/IO'g M,0 '
14 t 2 g/IO*g HjO '
14 t 2 g/IO'g N,0 *
14 , 2 g/IO*« H,0 *
14 * 2 g/10*0 H,0 *
14 t 2 g/IO*g M,0 *
14 , 2 9/IO*« HjO *
14 « 2 •/!«•• NiO *
14 , 2 g/IO*g H,0 *
Not Found
Hot Found
1 Boiling Order It » listing of tiydroc«rbc4ii In the order of tolling point.
' Solutlllllo are »l 2S*C unless otherwise steled. Also for dilute solutions g/IO(g
* CstlMted solublllly using no«ograph; Kebedl and Dinner (H7S|.
-------
soLUBiums or GASOUNC COHPONCMIS
CMP onto t
CltlBBlKlhU
Cyclopentene
cyclohgHtng
I-•ethytcyclogenlene
l-«gthylcyclopentene
lHMthy1cyclottencne
l-0thy1cyclopenten«
4-Mthy1cyc1oh«Mne
•o(lloa Order*
Boiling folnt
C,H,
C.H,.
C.H,.
C|N,,
CiH,,
CiN,,
4?
14
II
III
91
104
III. 14
114.91
Solubility1
Branched P1ef|n«. ICont Inucdl
l.l-dlMthyl-I-hgMnt
1,1-dtMltiyl-lrins-l-hgNcno
t-othyl-1-neiana
l-othyl-l-heiene
t-Mthyl - 1 -neptgng
l-Mthyl*l-lMptgng
I-Mthyl-I-htptono
I-ettlhyl-l-octono
t-Mthyt-I-oclgng
1 -«ethyl -l-nonene
C.H,.
C.H,,
C.H,,
C.H,.
C.H,,
C.H,,
C.H,,
C.H,.
C.H,.
CI.HM
l«l
91
111
101
111
114
Ml
191
199
11 J
III. 19 Not found
Not round
1.1 * 0.« 0/10*0 H,0 »
1.1 * 0.4 0/10*0 H,0 •
H4.7 1.1 , 0.4 0/10*0 M,0 »
1.1 * 0.4 0/10*0 N|0 »
I.I j 0.4 0/10*1 HiO *
0.4 , O.I 0/10*0 N,0 *
0.4 i O.I 0/10*0 HjO *
0.09 , 0.01 0/10* MtO *
SIS t 10 0/10*0 HjO *
III t 10 0/10*0 MjO *
111 t 10 g/IO*g H.O »
111 i 10 0/10*0 M,0 *
SI , 1 0/10*0 H,0 «
44 t 9 0/10*0 NjO *
44 t 9 0/10*0 NjO *
1 Boiling Order Is • listing of hydrocarbons In the order of boiling point.
' Solubilities are at li*C unless otherwise staled. Also for dilute solutions 0/!0*g •*/!,
' IstlMled solubility using luwograph;. Kabadl and 0«nner II»)9I.
4 McAulirfe 119141.
-------
SOLU8ILI1HS Of GASOLINE COMPONENTS
Component
Foraulu
Bollln« Order
Boiling Point
Solubility1
Conluaited fllene*
l.l-butidlene
l-*ethy1-l,l-buU0 *
IS7.0 t 1.0 9/10*9 HtO *
200 9/10*0 H|0 '
.0 j 1.6 1/10*0 HfO *
, J 0/10*0 H,0 *
t 1 9/10*0 HjO It 0°C *
.0 t ?.« 9/I01* HjO *
A '
146
16?
196
114
196 9/10*0 M,0
B«llln0 Order ts • Itstlng of hydrocarbon* In the order of boiling point.
SolubUUIet «rc «t 2S*C unltts olh«r«ls« itkted. Also for dilute solution* 9/10*9
HcAultffe (1Mb}.
PoUh »nJ lo 119)1).
H*K (I960).
Price II9H).
Bolioii *«il CUusscn I HID-
H»ck«y tnd Shlu II9T&).
Sullen *»J Cildcr
•9/1.
-------
SOLUOIllllfS Of CASOLINC COMPONENTS
Component
SubaUlUlcd-ArflOftLUi (Continued)
o-xylene
alhylbcniene
l.>.4-trUiclhylbcnfcn« (pscudocuaene)
Isopropylbiniena (cuaene)
n-propylbeniene
l-Mlhyl-l-athylbcnieng
1 -aethyl - 4 •athylbenienc
l-Mihyl-l-athylbenicne
I.M-trlMCthylbeniena (•esttylene) .
t-butylb«nicne
•oiling Order Is o listing of hydrocarbons
Solubilities ara at 2S*C unless otherwise
HcAullffc (1966).
Sullon and Caldcr 1 19)1).
Price 11916).
Polih and lu (1971).
,.«.,.
C.H,.
C.H,.
C.H,,
C.H,,
C,H,,
C.H.,
C.H.,
C,H,,
C(H,|
°
In the order
slated. Also
•oiling Point
Boiling Order1 *f
190 291.94
170 277.11
220 1)6.01
201 IOC. 11
209 1IO.S9
214 122. IS
2IS 121. SI
220 129.20
111 121.49
22S 116.41
of boiling point.
for dilute solutions g/lo*g »g/l
Solubility*
I7S t 0 1/10*9 H.O *
170. S t 7.S g/io'g H,0 4
167.0 , 4.0 |/I049 H,0 *
211 , 4 9/10*9 H,0 *
147 , 2 9/10*9 HjO at 0°C *
IS7 t • 0/IO'g H.O '
161.2 t 0.9 i/io'g N,0 4
197 t 1 9/10'a M,0 at 0°C '
177 s 1 9/IO*| H.O *
111.0 , 1.4 g/io'g HjQ *
200 g/lo'g N,0 '
160 g/IO*g H,0 *
S7 t 9/10% H.O *
S9.0 0.0 9/10*9 H|0 *
SI. 9 1.2 0/10*0 H,0 *•
SO t 0/10'a H.O *
61.1 0.0 g/io'g HjO *
40.1 1.2 0/10*0 HjO *
71 9/10*0 HjO *
SS 9/10*9 H,0 *
40.0 ,1.6 g/IO'g H,0 *
40.0 , S,6 9/10*0 H,0 *
40.0 , S.6 9/IO*g H,0 *
41.7 t 0.1 0/10*0 H|0 *
91 g/IO*9 H,0 *
19. S T 0.1 g/IO*9 H,0 *
1* 9/10*9 MjO *
-------
SOLUBIIUICS OF GASQlIMC CQMTOHCNTS
Component
Boiling Point
ror«uU Balling Order1 °r
Solubility*
SlitallLuUd-JBUlBOLUi (Continued)
Isobutylbentene
loc-butylbeniene
l-Mlhyl-1-lsopropylbeniene
l-«ethyl-4-lsopropylbcnicne
1.2,1-trlMlhylbenien*
l-«ethyl-2-tiopropylbeniene (o-cyaene)
1.1-dlclhylbemene
2.1-dlnydrotndcnc (Indent
1-Methyl-1-n-propylbenicne
n-butylbeniene
l-Mtthyl-4-n-propylbeniene
l.t-dlethylbeniene
1.4-dleth/lbenjene
t-Melhyl-2-n-propylbeniene
I.1-dtBKthyl-S-ethylbtntene
C.H.,
C.H,.
l.4-dlMthyl-2-elhy1beniene
iHMthy linden
CioH,«
210
211
216
240
241
244
241
246
247
IS I
211
2S1
294
2SS
2S6
117
142.97
1O.9S
147.09
HB.9S
112.67
161.19
10.I j 0.4 g/IO'g HjO *
11.6 * 0.2 g/IO'g HjO *
09 g/10'g HtO *
09 g/ia'g NjO '
7i.l , 0.6 g/IO*g HtO *
01 g/IO*g H,0 *
6.0 * 0.
00.9 j 1.7 g/IO*g H,0 '
109.1 « 1.01 g/IO'o H,0 '
6.0 t 0.9 g/IO'g HjO *
II.0 t O.I g/10'0 HtO *
6.0 t 0.9 g/IO'g H,0 *
6.0 , 0.9 g/IO*g H,0 *
6.0 i 0.9 g/IO'g H,0 *
6.0 t 0.9 g/IO'g tt,0 '
Not round
Not round
Not round
Not round
•olltng Order i* » imtng of hydrocarbons In 1 he order of boiling point.
Solubilities arc »l I4*C unless olhcrHtse staled. Also for dilute solutions g/IO*g mg/1.
Price f!97b(.
Sol I on md Calder
-------
SOlUBIllllfS Or CASOIINI COHTOHINIS
00
Component fora»lt
SublULul£4_ArfiUllca (Continued)
tHMthyl-)-t-bulylb«ni«nt CiiHu
l.l-dlMthyl-t-elhylb«ni«n« C|«HM
l.l-dlMlhyl-1-tlhylbtni«nt C|«M|«
1,1-tHMethyl-I-olhylbenMM C|»HM
l-«*lnyl-4-l-butjlbtnient CiiHu
I.I-dlMlhyl-l-elhylbenitno C)aH|«
I.I.4.l-l«lrMMlhylbtnicnt Cioth*
I.I.I.S-ttjtrftMthylbgniene CieH|<
Isoptnlylbeitiene C||N|(
•-•clhyllndin CI«NU
4-4Mlhyllndin Ci.H,,
n-Hntylb«LkH k««.4 C|B|» flallft
Balling Point
Boiling Order1 °f
IS!
»l
lt«
Itl
Itl
Itl
its
Itt
itr
It!
It«
119
III
III
III
IIS
of boiling point.
for dilute solutions g/lo'g «g/l.
Solubility1
Nol found
Not found
Not found
Nol found
Nol found
Nol found
1.49 t 0.19 g/IO*(
Nol found
Nol found
Nol found
Nol found
Nol found
Nol found
Nol found
It .1 | 0.< g/io't
11.69 , 0.11 g/la'
11. I , O.It g/IO(|
14. 4 t g/IO*g AtQ
11.1 , g/IO*g H|0
Nol found
1 H,0 •
M,0 «
10 M|0 •
ltM,0 •
•
-------
SOLUBILITIES of GASOLINE COHMMIHTS
Component
Substituted Cvclnpentanet
1,1.1-trtMthylcyclepinltna
l-cli-l-d»«elhylcydopentana
l-lrans-2-dlMthylcyclopantana
l-trani-1-dlMthylcydopcntana
l-cts-1-dlMthylcyclopantana
1 . 1 -dtMthylcyclopcntanc
l-«ethy1-cl3-l-clhrlcyclopenttne
l-cls-2-cls-l i. iMthytcydopcntan*
l-Mlhyl-l-«thylcrcUpcntana
I^Mthyl-cll-l-«lhy)cyClopenline
l-Mthyl-tranf-2-tlhylcycloptntane
)-«Mthyl-trant-l-ethytCpclop«ntaM
l.1-cls-2-lrans-4-tetraMlhylcyclopentan*
l-elS-l-CtS-4-lrlBKlhyleytlopentana
l-cls-2-trans-l-tt t*«thylcyclopcntaM
l-cls-l-trans-4 trUctnylCyclopcntan*
1.1.2-trlMthylcyclopentanc
l-trans-2-cls-l-lrlMthyUyclopcntant •
l-trans-2-cls-4-trlMlhylcyclopcntan«
Fowl a
CiHM
CiNi«
CiHi«
C|H|4
C |H|4
c.Ml(
C,HU
C.HI,
C.MK
C,H,(
C.MI,
Ct«U
C.M,.
C.H..
C.H,«
C.H..
C.HI(
C.«u
Boiling Order1
101
99
00
74
71
1*9
MI
146
Ml
MO
119
0olltn0 faint
°f
210
211
197
I9S
190
2S1
ISO
2SO
2SO
2SO
.01
.16
.10
.19
.12
.4
.74
.00
.2
.0
Solubility*
1.7J
7.07
7.07
7.07
7.07
7.07
1.
1.
1.
1.
1,
1.
71
71
71
71
71
71
111 Mot
129
110
111
117
111
109
241
241
141
210
22«
.0
.s
.12
.4
.72
1.
1.
1.
1.
1.
1.
71
71
71
71
71
,0.17 «/IO*a HiO *
t 1.0 0/10*0 HjO 4
, 1.0 0/10*0 H,0 4
t 1.0 0/10*0 MjO 4
, 1.0 0/10*0 H,D 4
t 1.0 0/10*0 H«0 *
t O.S2 0/10*0
* 0.
J 0.
* o.
» o.
t 0.
found
r 0.
t 0.
t 0.
t 0.
1 0.
71 , 0.
S2
12
>2
S2
12
S2
SI
SI
S2
SI
SI
0/10*0
0/10*0
0/10*0
0/10*0
0/10*0
0/10*0
0/10*9
9/10*9
9/10*0
0/10*0
0/10*0
H,0
H,0
HjO
HjO
HjO
HjO
HjO
HjO
H,0
H,0
HjO
HjO
*
4
•
4
4
4
4
4
4
4
4
4
Order Is • listing of hydrocarbons In the order of boiling point.
J Salubtlttlcs «rc «t IS'C unless olherwlsc stated. Also for dilute solutions 9/10*9
1 Price
-------
SOiuaiiiiics or GASOUNC COMPONENTS
Component
SuB2llUilc4_£yclahcjuncs
l-tr.ns-4-dlaclhylcyclohextne
l.l-dle«thylcyc)oh«Mnc
1-ClS-l-dlMlhylcycloheMne
l-trint-2-dlMthylcycloheMne
l-CtS-4-o'lMthylcycloriONBno
l-trens-1-dlMthylcycloheMnt
1.1-dlMlhylcyclalMjMino
l-Ctl-2-dUcthylcyclohOMno
> 1.1.1-trleMthylcyclohOMnt
N)
0 1.1.4-trlMthylcyclolMMno
l-cll-l-ctS-J-lrlewlhylcyclohe«ene
Mrans-l-trini-4-lrtMlhylcycloliMMM
1-cll-l-trini-S-trleMthylcyclohoMno
l-trans-2-cls-4-lrlMthy1cyc1ohgHing
l-trans-2-cts-l-trlMthy1cyclohe«BM
l.l.2-trlMthy1cyc1ohe«en«
l-cls-2-lrins-l-trlnethylcycloheMng
l-«ethyl-trans-4-tsopropylcyclohe«int
l-Mthyl-clS-4-tSopropylcycloh.H.ne
foretile
C.H,.
C.Hu
C.H,.
C.Mu
C.NU
C.N,,
C.NI.
C.N,«
C.N..
C.H,.
C.H,.
C.H,.
C.H,.
C.H,.
C.N,.
C.M,.
C.H,.
C|0H,0
C|«H1V
Boiling folnt
Botllnf Order1 °f
114
IIS
III
l«7
III
ISI
IS7
I6«
171 II7.9I
IM
lift
111 21*. 20
III
1*1
196
191 191.40
20S
229 119.296
21ft 119.29*
Solubility*
l.M , O.tl i/lo'g HtO '
l.ll . O.S2 i/loS N«0 *
1.71 • 0.12 •/!«•• H,0 «
1.71 . O.S2 •/!••• H,0 *
1.71 t «.»2 •/!•*• HiO *
1.71 * I. (2 o/ll'i NiO *
1.71 * O.S2 t/ll'g HjO *
t.O t O.I 8/IO*g N|0 *
1.77 . 0.01 s/IO** HjO *
l.7» t 0.2S I/IOS H,0 *
l.7» t 0.21 |/IO% H|0 *
I.7S , 0.2S f/IO1! NtO 4
1.7ft t 0.2» l/IO'f NiO 4
1.7ft i 0.2t (/lO'c HjO *
I.7S , 0.2S t/10'i H,0 *
Nol found
l.7i t 0.2ft e/IO'g MjO *
0.72 t 0.10 •/lo'e H,0 *
0.72 , o.io •/IO'B M,O *
Boiling Order t« . list Ing of hrrdocarbont In the order of balling point.
Solublllltct ire »l 2ft*C unless alhiiMtse Steteil. Also for dilute solutions g/io'g wg/l.
Price M17fc).
IstlMled solubility using no«ogreph: Kebedl and O.nncr (19791.
-------
APPENDIX B
REVIEW OF THE LITERATURE ON GASOLINE VAPOR
MEASUREMENTS AT AND NEAR SERVICE STATIONS
B-l
-------
B-2
-------
APPENDIX B
REVIEW OF THE LITERATURE ON GASOLINE VAPOR
MEASUREMENTS AT AND NEAR SERVICE STATIONS
The literature on gasoline vapor measurements and emission rates can be classified
into the following groups: self-service refueling exposure, occupational exposure, nearby
neighborhood exposure, and other exposure parameters. Each of the following studies is
labeled as being in one of those groups. Where data from a study are considered in
forming scenarios 1,2, and 3, the mean and upper limit values are given in Table 5-1 of
Chapter 5.
Philadelphia Air Management Services Study (Self-Service Refueling)
The Philadelphia Department of Health, Air Management Services (AMS)
conducted a short-term survey of benzene concentrations at roadway intersections and
service stations (Ellis and Obendorfer, 1984; U.S. EPA, 1986b). The study was done in
two parts. Fixed sampling locations were used near six roadway intersections, one of
which had gasoline stations on three of its corners. A second phase of work involved
samples drawn at various distances from a refueling nozzle at a city gas station removed
from roadway traffic. Quality assurance procedures seem thorough. The use of glass
containers for the benzene samples, however, caused some sample loss. Glass has a
strong affinity for benzene and side wall losses can be up to 50 percent (Intersociety
Committee, 1989) for low concentration samples. The AMS seemed aware of this
problem, but made no effort to calculate a correction factor. The study period of February
to April presented a wide range of temperatures (31 to 69°) and wind speeds. Due to the
container loss problem, the resulting measurements are probably underestimated.
Unfortunately, because of the very limited number of samples, AMS was unable to
separate out the traffic versus gasoline marketing influences on benzene concentrations.
Thus, the data from the fixed locations are not applicable to this study. The measurements
made near a refueling nozzle were mostly taken at distances which are too close to represent
inhalation exposure during refueling. The four samples taken at more representative
distances of 2 to 2.5 feet away ranged from 6 to 612 ppb benzene, averaged 164 ppb
benzene, and were collected over periods of 1 to 2 minutes. No meteorological data were
reported for the days when these samples were collected. These measurements, though
few in number and possibly underestimated, will be considered in scenarios 1 and 2.
B-3
-------
GM Refueling Vapor Study (Self-Service Refueling)
General Motors has reported (Tironi et al., 1986; Williams, 1987) on a program
they conducted to measure vapor exposure of people during gasoline refueling. A total of
17 measurements were made during the summer (August and September) and 15
measurements during the winter (February) at a refueling location in southern Michigan.
Personal samplers and Tedlar bags were used to collect 0.5 to 2 minute integrated samples
from the breathing zone during a refueling event that ranged from 4 to 16 gallons. All
measurements were made at a reasonable distance of 2 to 4 feet from the vehicle filler pipe.
A commercial grade of unleaded gasoline was used in the study. Ambient temperatures and
wind speeds were typical for the Northeast states.
The program appears to have been well planned with adequate quality assurance
procedures. Two different positions were assumed by the person doing the refueling. In
most of the tests, the person kept his hand on the pump handle depressing the trigger
mechanism. In other cases, a trigger-latch was set and the person apparently did not lean
over to reach the pump handle. The measurements made in the latter case do not represent
exposures at self-service stations where, for safety reasons, the trigger-latch is removed.
Excluding these latch data, a total of 19 samples applicable to scenario 1 were examined.
The authors report that the THC distribution of all measurements is highly skewed
and that a 90 percentile statistic is a better measure of likely exposure than either the median
or mean. The skew is due mostly to the mixing of data taken in the two different refueling
positions. Eliminating the much lower measurements when the person is not holding the
pump handle produces a more symmetric distribution. Thus, for this study, we believe the
mean concentration of the 19 samples is a reasonable measure of typical exposure.
The authors report vapor concentrations in ppm C (ppm Carbon) and these were
convened to ppm v (ppm volume) by dividing through by the number of carbon atoms per
molecule: benzene (6), toluene (7), xylene (8), and THC (4.7 average). The convened
concentrations, mean and maximum, values are shown in Table B-l.
U.S. EPA Refueling Vapor Study (Self-Service Refueling)
A gasoline vapor exposure study was carried out by the U.S. EPA Environmental
Monitoring Systems Laboratory in Raleigh, North Carolina (Bond et al., 1986).
Measurements were taken under meteorological conditions which are typical for the
Northeast states: temperatures ranged from 4° to 16° C and wind speeds were moderate. A
number of measurements were taken at five fixed distances from the vehicle filler pipe over
a 4-day period. Five of these samples were taken in the Breathing Zone (BZ), a point
B-4
-------
TABLE B-1
GASOLINE VAPOR CONCENTRATIONS MEASURED DURING
REFUELING WITH A SELF-SERVE TYPE PUMP NOZZLE
(ppmv)
Sample No.
Summer
14
15
16
17
Winter
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Mean
Maximum
Benzene
1.5
2.1
0.9
2.3
2.9
2.2
0.2
0.1
0.5
0.04
1.3
0.1
1.7
1.3
1.4
0.01
0.1
0.01
2.9
1.1
2.9
Toluene
1.2
2.0
0.8
1.7
1.5
1.4
0.1
0.1
0.4
0.05
0.7
0.1
0.8
0.6
0.7
0.02
0.1
0.02
1.4
0.7
2.0
Xylene
0.1
0.2
0.1
0.1
0.1
0.1
0.03
0.03
0.1
0.02
0.04
0.04
0.04
0.04
0.05
N.D.
N.D.
N.D.
0.1
0.1
0.2
Total HC
173.4
213.9
111.6
259.5
682.8
466.2
23.8
23.1
103.5
3.0
244.3
17.6
400.6
316.0
306.8
0.7
32.3
1.5
567.4
207.8
682.8
SOURCE: Tironi et al., 1986.
B-5
-------
which was 3 feet from the filler pipe and at breathing height. A commercial grade unleaded
gasoline was used in the study and approximately 10 gallons were pumped over a 90
second period.
Stainless steel canisters were used to collect samples and adequate quality assurance
procedures were followed. Four of the five samples at the BZ location were for a wind
direction blowing directly from the vehicle to the sampler intake (labeled parallel winds).
One measurement was for perpendicular winds. The resulting hydrocarbon concentrations
were not significantly different for the two wind conditions at the BZ location, indicating
that the turbulent wake around the vehicle and refueling island will obscure differences in
concentration due to wind direction at very close distances.
The authors report their vapor concentrations in ppm C and these were convened to
ppm v using sample average carbon numbers reported by the authors. The convened
concentrations, mean and maximum values, are shown in Table B-2.
Clayton Gasoline Vapor Study (Self-Service Refueling)
A gasoline exposure study was conducted for the American Petroleum Institute by
Clayton Environmental Consultants, Inc. (Clayton, 1983) at 13 service stations across the
U.S. Two of these stations, in Philadelphia, are close to the Nonheast states.
Measurements at these stations occurred under summer conditions: temperatures of 27 to
28°C and moderate winds. Samples were collected in charcoal tubes attached to air pumps
worn by the person pumping gas. Although no information is provided on the distance of
the pump intake from the filler pipe, it is assumed that these data represent breathing zone
exposures. Data are reported separately for leaded, unleaded, and premium gasoline. Only
samples from the latter two categories are considered here.
The accuracy of the Clayton data is difficult to assess because the text describing the
sampling and analysis procedures is extremely shon and provides no information on
several key items. The authors explain that "breakthrough" was tested for only in a portion
of the sample tubes with "higher concentration". Breakthrough is generally defined as the
back section of the charcoal tube containing 10 percent or more of the hydrocarbon mass
found in the front section of the tube. When sufficient mass appears in the back section
there is a real possibility of sample loss (i.e., hydrocarbons have blown through both
sections and out of the tube). Breakthrough does not uniquely occur when the front section
of a sample tube has "high" concentrations. It can also occur at lower levels when
relatively high air flows are pulled through a sample tube, as was the case here. The fact
that the back sections of some of the tubes were not even analyzed suggests that the
B-6
-------
TABLE B-2
GASOLINE VAPOR CONCENTRATIONS IN THE BREATHING ZONE
MEASURED DURING REFUELING
(ppmv)
Sample No.
Parallel
1
2
3
4
Perpendicular
2
Mean
Maximum
Benzene
0.33
0.16
0.20
0.23
0.24
0.23
0.33
Toluene
0.26
0.16
0.18
0.19
0.18
0.19
0.26
Xylene
0.12
0.19
0.07
0.14
0.11
0.13
0.19
Total HC
32.1
18.3
18.8
23.4
24.3
23.4
32.1
SOURCE: Bond et al., 1986.
B-7
-------
hydrocarbon concentrations are biased low. An independent analysis by Shiller (1987) has
also concluded that hydrocarbon concentrations were underestimated in the Clayton study.
There are other problems with the sampling and analysis procedure. Complete
desorption of benzene from a charcoal tube is difficult and the solvent used, carbon
disulfide, can be contaminated with benzene. Thus it is very important that both blank
charcoal tubes and ones spiked with a known standard of benzene be analyzed to determine
the extent of any such problems. The authors did not report that any of this quality
assurance work was done. In summary, we believe the Clayton data underestimates the
actual exposures which occurred, but have considered them in Scenario 1. Relevant data
are listed in Table B-3. There are 35 samples, each of which represents an integrated
measurement over four t o eight refueling events.
GARB Benzene Control Study (Neighborhood)
As technical support for a proposed benzene control measure, the California Air
Resources Board estimated benzene exposures from service station activities (Ames et al.,
1987). Exposure factors for vehicle refueling used data from four papers in the literature
(Tironi et al., 1986; McDermott and Vos, 1979; Harde, 1980; CEC, 1983). Based on
these papers, CARD selected a number for the short-term benzene exposure to self-service
customers during refueling as 1,500 ppb for 2 minutes per week. This compares favorably
with the benzene concentration range of 0.9 to 4.2 ppm developed in this study (see Table
5-2). CARS assumes that the benzene content of gasoline will increase in the next decade
and as a result, they assume an increase from 0.6 percent to 0.8 percent in the weight
percentage of benzene in THC vapors. By comparison, the vapor concentration data in this
study (see Table 5-2) represent a range of 0.6 to 1.0 percent benzene, by weight. Using
the higher weight fraction of 0.8 percent in the year 2000, CARB estimates a service station
with Stage I controls will emit 0.044 g/gallon of benzene. That number is consistent with
the range of emission rates, 0.040 to 0.047 g/gallon, assumed in this study (see Table C-l)
for emission that are transported into nearby neighborhoods.
CARB estimates, from dispersion modeling, that people living in neighborhoods 20
to 200 meters from a service station with Stage I controls that pumps 1 million gallons per
year are exposed to an incremental annual benzene level of 0.04 ppb. In this study, the
ensemble average annual concentration for receptors from 30 to 200 meters from a
comparably sized service station yields the very same benzene level-0.04 ppb-confirming
the similarity in methods between this study and the CARB report In addition, this study
calculates an upper limit benzene concentration for a person living 30 meters from a station
as 0.16 ppb. Since both this study and the CARB study are based on similar methods and
B-8
-------
TABLE B-3
GASOLINE VAPOR CONCENTRATIONS MEASURED DURING
REFUELING AT A SERVICE STATION IN PHILADELPHIA
(ppmv)
Sample No.
01-01-U
01-03-U
01-04-U
01-05-U
01-07-U
01-08-U
01-10-U
01-12-U
01-13-U
01-14-U
01-15-U
01-16-U
01-17-U
01-19-U
01-20-U
01-21-U
02-0 1-U
02-03-U
02-05-U
02-07-U
02-08-U
02-10-U
02-1 1-U
02-12-U
02-14-U
02-18-U
02-20-U
02-2 1-U
02-22-U
02-23-U
02-25-U
02-26-U
02-28-U
02-16-P
02-17-P
Mean
Maximum
Benzene
0.70
0.89
0.20
0.30
0.10
0.20
0.84
4.2
1.7
2.8
1.2
0.72
3.5
0.73
1.5
0.54
0.59
2.6
2.4
0.68
0.30
0.20
0.75
0.45
0.50
0.56
0.20
0.59
0.41
1.5
0.10
0.20
0.10
<0.07
<0.08
0.92
4.2
Toluene
1.3
0.81
0.51
0.42
0.45
0.27
1.2
1.3
0.47
0.83
0.36
0.29
1.3
0.24
0.43
0.20
0.29
0.76
1.1
0.29
0.20
0.08
0.20
0.20
0.20
1.1
0.43
0.84
0.20
0.65
0.44
0.30
0.31
2.4
<0.50
0.59
2.4
Xylene
0.45
0.50
0.30
0.35
0.26
0.26
0.67
1.1
0.36
0.60
0.29
0.10
0.98
0.20
0.42
0.10
0.10
0.58
0.43
0.10
<0.10
<0.09
0.10
0.08
<0.10
0.64
0.33
0.48
<0.10
0.20
0.33
0.10
0.27
<0.40
<0.40
0.32
1.1
Total HC
120
110
81
79
35
57
150
260
100
190
74
47
210
43
110
32
31
160
96
34
18
11
42
25
30
200
97
190
27
48
88
27
70
95
21
86
260
SOURCE: CEC, 1983.
B-9
-------
some of the same literature sources, it is not surprising that the calculated exposure levels
are very similar. Therefore, the GARB study presents no new or independent empirical
data, it could not be used in developing scenarios 1-3.
API Occupational Exposure Study (Occupational)
The American Petroleum Institute measured occupational exposure to gasoline
vapors, including exposure of service station attendants at unnamed locations during one
unnamed season of the year. An analysis of the data is reported in Rappaport et al. (1987).
A total of 49 samples related to a service station attendant being exposed over an 8-hour
workday were collected at four unnamed locations. The measurements are valuable in that
an integrated 8-hour sample collected on the person was taken. But, the representativeness
of the samples cannot be judged since no information is provided on geographical
locations, seasons, meteorological conditions, or other site factors that affect hydrocarbon
concentrations.
The sampling and analysis method used was charcoal tubes with carbon disulfide
desorption. This method can underestimate benzene concentrations because of the
difficulty in desorbing benzene from charcoal and the possibility that the CS2 solvent has
benzene contamination. Because no information on quality assurance procedures was
provided, these problems become more likely.
The source in this case does not provide the actual measurements, but only mean
values and statistical measures. Thus no information on maximum exposure can be taken
from the API database. The situation is complicated by the fact that data from service
stations with and without Stage n vapor recovery are mixed together, except for total
hydrocarbons. The 33 samples without Stage II controls give a mean value of 94.5 mg/m3
THC exposure over 8 hours (33 ppm THC). For individual components of the THC
vapor, mean values are only given for mixed data. Scaling these by the THC ratio of non-
Stage II data to all data, rough estimates of non-Stage n mean values are as follows:
benzene 0.872 mg/m3 (273 ppb), toluene 1.013 mg/m3 (269 ppb), and xylene 0.571
mg/m3 (132 ppb). Although they lack documentation of representativeness and are likely
underestimated, these mean values will be considered in scenario 2.
The study also goes on to determine whether average vapor component to THC
ratios can be developed that accurately predict the component concentrations if only THC is
measured. The authors (Rappaport et al., 1987) conclude the "approach is viable" and
"relatively good agreement between predicted and observed values" is obtained. We
disagree. The actual results of a test of this method, applied to the GM database (Tironi et
al., 1986), show that individual components are predicted to be from 1/10 to 11 times the
B-10
-------
actual measured values, a spread of 2 orders of magnitude. The aromatic components are
overpredicted by factors of 2 to 11 in this one test and no estimates of the stability of the
factors in different test environments is given.
Hartle Benzene Exposure Study (Occupational')
A study of the occupational exposure to benzene for service station attendants was
performed at a variety of service stations across the U.S. The results, reported by Hartle
(1980), give data on both 8-hour time-weighted average (TWA) exposures collected with
personal samplers and area-average concentrations measured at fixed sampling points on
top of the refueling island. A total of 64 TWA samples were collected, but only 21 are
judged to be usable for this study. Samples were excluded where climatic conditions were
outside the normal range for the Northeast states, where Stage n controls were being used,
and where suspect laboratory analysis had been performed. Charcoal tubes were used for
sample collection. This method can underestimate benzene concentrations. Since no
information on quality assurance procedure was provided, problems with underestimation
must be considered likely.
The source in this case does not provide the actual measurements but only mean
values and other statistical measures. The mean value for the 21 TWA samples is 122 ppb
and the maximum 8-hour value is 237 ppb. The pump area concentrations averaged 38 ppb
with a maximum value of 55 ppb. These numbers, although possibly underestimated, will
be considered in scenario 2.
Mobil Oil Vapor Exposure Study (Occupational)
A high-volume service station in Pennsylvania was selected for an exposure study
during 1 week in the spring of 1983 (Kearney and Dunham, 1986). Eight-hour TWA
samples were taken on workers at the station using personal samplers. Pump area samples
were also collected over an 8-hour period. The air samples were collected on charcoal
tubes and the authors performed an adequate quality assurance program to correct for
sample loss and interferences. Although air temperatures were given, wind speeds are not
reported. This is of concern since the authors hint that winds may have been higher than
normal during the week of sampling. Such conditions could result in underestimated
exposures.
The actual measurements are not shown. Means and ranges are given for total
hydrocarbons. Vapor components are reliably reported only for concentrations
B-ll
-------
substantially above the method's detection limit of 100 ppb. Given the high detection limit,
none of the specific results for aromatic hydrocarbons are considered accurate. Worker
exposures to THC averaged 1,500 ppb with a maximum 8-hour value of 14,300 ppb for
one attendant. Pump area THC concentrations averaged 1,600 ppb over eight-hours.
These numbers, although possibly underestimated, will be considered in Scenario 2.
Shell Oil Vapor Exposure Study (Occupational)
Seven high-volume service stations were selected for a study of worker attendant
exposure to benzene and THC in the spring and summer of 1977 (McDermott and Vos,
1979). A mixture of leaded and unleaded gasolines were in use and no separate
information for unleaded gasoline is given. Since leaded gas generally has a lower benzene
content, these results may underestimate exposure from a station that pumps primarily or
only unleaded gasoline.
Two of the stations, in Connecticut and Dlinois, have weather conditions
considered typical of the Northeast stations. Thus, only data from those two stations are
considered here. No information on measured wind and temperature conditions were
reported by the authors, adding uncertainty to how representative the data are. Personal
samplers were worn for an 8-hour work shift and carbon was used to collect the
hydrocarbon vapors. An incomplete description of analytical techniques and quality
assurances procedures makes it difficult to assess the accuracy of the results. The actual
measurements are not reported, only summary statistics.
At the two stations, 29 TWA samples were taken. One sample has been excluded
here that had an extremely high benzene level (2,080 ppb) and for which there is evidence
to suggest tampering. The remaining 28 benzene samples had a mean value of 128 ppb and
a maximum value of 840 ppb. For THC, the mean value was 13,500 ppb. A maximum
value, after excluding suspect sampling, was not obtainable from the source document
These numbers are possibly underestimations of future exposures due to the selection of
collection medium and the use of leaded gasoline. The 28 samples are considered,
however, in forming Scenario 2.
Amoco Vapor Exposure Study (Occupational)
Amoco Corp. conducted a study of worker exposure to gasoline vapors at a variety
of locations including one service station plaza (Haider et al., 1986). The
representativeness of the 21 TWA samples is difficult to assess since no information is
B-12
-------
given on the geographic location, volume of gasoline sales, or meteorological conditions.
An incomplete description of quality assurance procedures also makes it difficult to assess
accuracy. No actual measurements are shown. The authors state that measurements were
adjusted to an 8-hour TWA by assuming zero exposure during nonmonitored periods.
Since most samples were collected over less than 8 hours, the TWA values are definitely
underestimated for a full workday exposure. The mean benzene exposure of 300 ppb and a
maximum value of 1,300 ppb are considered in scenario 2.
Other Exposure Parameters for Scenarios 1 and 2
GM Refueling Habits Study
Two surveys of the refueling habits of service station customers are reported by
General Motors (Lombardo and Behrens, 1987). The first survey, which questioned
1,182 customers in the Detroit area, found a mean fill volume of 9.8 gallons and a 98
percentile volume of 20 gallons. Lombardo and Behrens also reported data from an oil-
industry survey which questioned several thousand customers. These data for the
Northeast region of the U.S. show a mean fill volume of 9.9 gallons and a 97 percentile
volume of 20 gallons. Based on this information, scenarios 1 and 2 will assume a mean fill
volume of 10 gallons and an upper limit fill volume of 20 gallons.
Ford Motor Company Paper (Vehicle Gasoline Consumption)
Shiller (1987) reports that the average motor vehicle gasoline consumption is 700
gallons per year. This paper reviews the Clayton, U.S. EPA, and GM data on vapor
exposure during vehicle refueling. The validity of the Clayton data was questioned on the
basis of unusually low THC/benzene ratios. The author also reports on a conversation
with Clayton Environmental Consultants which confirms that CEC had problems with
sample loss from their charcoal tubes during that study and that THC concentrations are
underestimated.
B-13
-------
API Temperature Survey (Gasoline Temperature Measurement)
The American Petroleum Institute conducted a large number of gasoline temperature
measurements across the U.S. in 1975 (McAnally and Dickerman, 1976). The database
for the Northeast states shows that dispensed gasoline varied in temperature from 6°C to
26°C with a mean value of 17°C. At the same locations, the annual average underground
tank temperature was 14°C, the average ambient air temperature was 12°C, and the average
retail refueling volume was 10 gallons.
U.S. EPA Refueling Study (Air concentration other than during refueling)
Another interesting aspect of the Bond et al. study is that measurements were taken
at a point 9 feet away from the filler pipe, perpendicular to the vehicle, under three different
wind conditions: parallel, reverse parallel, and perpendicular. Parallel winds are those that
blow directly from the filler pipe to the sampler. Reverse parallel winds blow from the
sampler toward the filler pipe, and perpendicular winds move perpendicular to the filler
pipe-sampler axis. At the distance of 9 feet, total hydrocarbon levels were essentially equal
to measured background levels for perpendicular and and reverse parallel winds, and
benzene concentrations were below detectable limits. For parallel winds, average benzene
and THC exposures at 9 feet were about 25 percent of breathing zone levels and were
decreasing with distance at a rate of 1-2 percent per additional foot At 20 feet, the level
would be roughly 10 percent of the breathing zone value. During the time a customer is in
the station, but not pumping their own gas, they are exposed to vapors from other, on-
going refueling operations at varying distances; a reasonable average distance of 20 feet is
assumed. Also assuming an equal distribution of winds relative to the filler pipe-exposed
person axis, the parallel case will occur only 25 percent of the time on average. Thus,
concentrations for non-pumping time spent at a service station could be approximated as 2
percent, on average, of breathing zone levels and 10 percent, maximum, of such levels.
GM Tank Vapor Composition Study (Neighborhood)
Engineers from General Motors (Furey and Nagel, 1986) have reported on a test
program which measured gasoline vapors in vehicle tanks before refueling and expelled
from tanks during refueling. Three gasolines were used. Only one of these, blended by
Chevron Oil Company, closely represents the composition of commercial unleaded gas.
Unfortunately, only a few of the tests were carried out using the Chevron gasoline.
The sampling and analysis procedure is well documented and included quality
assurance checks. The Chevron gasoline was heated to temperatures of 11°C and 23°C in
the vehicle tanks. Equilibrium vapors in the tanks under these conditions are insensitive to
B-14
-------
temperature and consisted of 0.77 percent of benzene by mass, 1.0 to 1.1 percent of
toluene, and 0.3 percent xylene, all mean values.
Shell Oil Tank Vapor Composition Study (Neighborhood)
A set of 95 gasoline vapor samples drawn from the air displaced during tank truck
loading operations was analyzed by McDermott and Killiany (1978). The study examined
an unspecified mix of both leaded and unleaded gasolines. No information is given
regarding sampling and analysis methods, quality assurance procedures, geographical
locations, or temperature of the gasoline. Vapors consisted of 0.7percent benzene by
volume, 1.8 percent of toluene, and 0.5 percent of xylene, all mean values.
EPA Refueling Emissions Study (Neighborhood Exposure for Scenario 3)
Two important studies by the U.S. EPA Mobile Source Emissions Research
Branch (Braddock et al., 1986; Braddock, 1988) measured THC and component emissions
from unleaded refueling operations under controlled conditions. The refueling experiments
were conducted using a sealed housing for evaporative determination (SHED) technique,
and investigated the hydrocarbon emission rate as a function of many variables.
Conclusions from these studies include:
0 The total mass of hydrocarbon vapors emitted during vehicle refueling is directly
proportional to the amount of fuel delivered to the tank.
0 Emissions are strongly dependant on the temperature of both the delivered fuel
and the pre-existing fuel in the vehicle tank.
0 Relative composition of the vapors emitted are independent of temperature.
0 Emissions are relatively insensitive to Reid Vapor Pressure. Minimum THC
emissions increased with RVP but average THC emissions actually decreased
with increasing RVP.
These results demonstrate that refueling emission rates can be reasonably predicted
on the basis of gallons of fuel pumped and also that emissions of THC and components
such as benzene can be reasonably predicted as a function of temperature. The results for
RVP are different than the 11 to 12 percent change in emissions per 1 psi that other
investigators have found (Hochhauser and Campion, 1976; Smith, 1972). The fact that
RVP has an intercorrelation with other descriptive parameters may explain these
differences.
U.S. EPA data for the Nonheast states (Rothman and Johnson, 1985) show that
the typical temperature of dispensed fuel varies from 52°F (winter) to 78°F (summer) with
an average of 62°F and that RVP varies from 11 to 15 psi over the year. Within fuel and
B-15
-------
tank temperature ranges characteristic of the Northeast, the U.S. EPA study data in Table
A-4 show values for emissions from refueling with an unleaded gasoline RVP of 11.4 psi.
The gasoline vapor components as a percentage of the mass of THC reported by
U.S. EPA (Table A-4) are consistent with similar data published by other researchers
(Furey and Nagel, 1986; McDermott and Killiany, 1978) for benzene and toluene. The
xylene fraction of 0.9 percent is about twice the level reported elsewhere and hence the
xylene emission rate may be a slight overestimate for a wide range of gasolines and
conditions. These data will be used in scenario 3.
U.S. EPA Emission Factors (Neighborhood)
The published U.S. EPA AP-42 emission factors for gasoline service station
operations (U.S. EPA, 1986a) give estimated rates for underground storage tanks with
Stage I controls of 1.3 lb/1,000 gallons throughput This includes both tanker truck filling
(0.3 Ib) and tank breathing and working losses (1.0). Vehicle refueling operations without
Stage II controls are estimated to be 11.7 lb/1,000 gallons throughput. This number
includes displacement losses (11.0 Ib) and spillage (0.7 Ib) from nozzle drip, spit-back,
and overflow from the vehicle's fuel tank filler pipe. These factors assume an RVP of 10
psi and a fuel temperature of 60°F. The total refueling loss factor of 11.7 lb/1,000 gallons
equals 5.3 g/gal, which is close to the average rate of 5.4 g/gal measured by U.S. EPA in
more recent studies (Braddock et al., 1986; Braddock, 1988). The AP-42 estimates are
used in conjunction with those from the more recent U.S. EPA studies to estimate mass
emission rates for scenario 3.
B-16
-------
APPENDIX C
ISCLT DISPERSION MODEL OUTPUT
C-l
-------
C-2
-------
APPENDIX C
ISCLT DISPERSION MODELING OUTPUT
A resident living near a service station is exposed to vapors whenever the wind
direction is such that they are downwind of the station. The U.S. EPA Industrial Source
Complex Long Term (ISCLT) model was used to calculate ambient air concentrations in
residential neighborhoods near a station. The upper limit exposure calculation uses the
maximum modeled annual average concentration, corresponding to a distance of 30 meters,
the closest distance a house could reasonably be to a service station. The mean exposure
calculation uses the average modeled annual concentration in an area defined by rings of 30
meters and 200 meters downwind.
ISCLT modeling is based on an emission rate of 1 g/s. Using the mass emission
rates in Table C-l and an annual gasoline volume of 1,000,000 gallons, annual average
emissions in g/s are as follows:
Mean Upper Limit
THC 1.8x10-1 2.2x10-1
Benzene 1.3 x 10-3 1.5 x 10-3
Toluene 3.5x10-3 4.1x10-3
Xylene 1.6x10-3 1.9x10-3
For a 1.0 g/s emission rate, the ISCLT output in Appendix C shows a maximum annual
concentration of 346.6 ug/m^ and an ensemble average for all 144 receptors of 82.2
ug/m3. Average concentrations at the four distances over all 36 directions are: 220.4
ug/m3 (30 m); 79.1 ug/m3 (50 m); 23.4 ug/m3 (100 m); 6.1 ug/m^ (200 m).
C-3
-------
TABLE C-l
GASOLINE VAPOR MASS EMISSION RATES
FOR THE NORTHEAST STATES
(g'gal)
Range Average % by Mass
THC
Benzene
Toluene*
Xylene*
3.5 - 6.3
0.023 - 0.043
0.063 - 0.12
0.030 - 0.056
5.4
0.036
0.10
0.047
100.
0.7
1.9
0.9
* Estimated from benzene emissions using detailed vapor composition data for a fuel
temperature of 59°F and a tank temperature of 72°F.
C-4
-------
ISCLT (DATED 88167)
AH AIR QUALITY DISPERSION H3DEL. IN
SECTION 1. GUIDELINE MODELS
IN UNAHAP (VERSION 61 JUNE 8B.
SOURCE: FILE 7 ON UNANAP MAGNETIC TAPE FPOH HIIS.
9
LA
-------
ISCLT •••••• NESCAUM Benzen* Risk Assessment •»•••••• pif-g | •*•»
- ISCLT INPUT DATA -
NUMBER OF SOURCES = 1
NUMBER OF X AXIS GRID SYSTEM POINTS - 4
NUMBER OF V AXIS GRID SYSTEM POINTS - 36
NUMBER Or SPECIAL POINTS « 0
NUMBER OF SEASONS = 1
NUMBER or WIND SPEED CLASSES « 6
NUMBER or STABILITY CLASSES » 6
NUMBER or WIND DIRECTION CLASSES = 16
FILE NUMBER OF DATA FILE USED FOR REPORTS - 1
THE PROGRAM IS RUN IN URBAN MODE 3
CONCENTRATION (DEPOSITION) UNITS CONVERSION FACTOR -0.10000000E+07
ACCELERATION OF GRAVITY (METERS/SEC**2) = 9.BOO
HEIGHT or MEASUREMENT OF WIND SPEED (METERS) - 10.000
CORRECTION ANGLE FOR GRID SYSTO1 VERSUS DIRECTION DATA NORTH (DEGREES) « 0.000
DECAY COEFFICIENT =O.OOOOOOOOE-MJO
ALL SOURCES ARE USED TO FORM SOURCE COMBINATION 1
RANGE X AXIS GRID SYSTEM POINTS (METERS )»
AZIMUTH
BEARING V AXIS GRID SYSTEM POINTS (DEGREES)-
60.00. 70.00, BO. 00,
160.00, 170.00, 180.00,
260.00, 270.00, 280.00,
90.00,
190.00,
290.00,
30.00.
0.00.
100.00,
200.00,
300.00,
50.00.
10.00,
110
210
310
.00,
.00,
.00,
100.00,
20.00,
120
220
320
.00.
.00.
.00,
200.00.
30.00,
130
230
330
.00,
.00,
.00,
40.00,
140.00,
240.00,
340.00,
50.00,
150.00,
250.00,
350.00,
O\ - AMBIENT AIR TEMPERATURE (DEGREES KELVIN) -
STABILITY STABILITY STABILITY STABILITY STABILITY STABILITY
CATEGORY I CATEGORY 2 CATEGORY 1 CATEGORY 4 CATEGORY 5 CATEGORY 6
SEASON 1 286.3000 286.3000 288.3000 284.0000 279.6000 279.6000
- MIXING LAYER HEIGHT (METERS) -
SEASON 1
WIND SPEED WHO SPEED HIND SPEED HIND SPEED WIND SPEED HIND SPEED
CATEGORY 1 CATEGORY 3 CATEGORY J CATEGORY 4 CATEGORY 5 CATEGORY 6
STABILITY CATEGORY 10.189000E+040.169000E+040.189000E+040.189000E4040.1B9000E+040.189000E+04
STABILITY CATEGORY 20.126000E+040.126000E+040.126000E4040.126000E4040.126000E+040.126000E+04
STABILITY CATEGORY 30.126000E*040.126000E+040.126000E+040.126000E+040.126000E+040.126000E+04
STABILITY CATEGORY 40.126000E+040.126000E+040.12COOOE4040.126000E4040.126000E4040.126000E+04
STABILITY CATEGORY 50.100000E+050.IOOOOOC+050.100COOC+050.100000E+050.100000E+050.100000E+05
STABILITY CATEGORY 60. IOOOOOC+050.100000E+OSO. 100000E+OSO. lOOOOOE-f050.100000E+050.100000E+05
-------
ISCLT
NESCAUM Benzene Risk Assessment
- ISCLT INPUT DATA (COOT.) -
- FREQUENCY OF OCCURRENCE OF WIND SPEED. DIRECTION AND STABILITY -
SEASON 1
STABILITY CATEGORY 1
WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPCED WIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
DIRECTION
(DEGREES)
0.000
22.500
45.000
67.500
90.000
112.500
135.000
157.500
180.000
202.500
225.000
247.500
270.000
292.500
315.000
937.500
( l.JOOOMPSM 2.7000MPSM 4.4000MPSM 6.7000HPSK 9 . 5000HPS M 1 2 . 7000MPS )
0.00000000
0.00000000
0.00001000
0.00000000
0.00001000
0.00001000
0.00003000
0.00004000
0.00003000
0.00001000
0.00001000
0.00004000
0.00003000
0.00001000
0.00001000
0.00003000
0.00000000
0.00000000
0.00002000
0.00000000
0.00005000
0.00005000
0.00009000
0.00002000
0.00000000
0.00005000
0.00002000
0.00005000
0.00009000
0.00002000
0.00002000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
SEASON 1
STABILITY CATEGORY 2
WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
DIRECTION
(DEGREES)
0.000
22.500
45.000
67.500
90.000
112.500
135.000
157.500
160.000
202.500
225.000
247.500
270.000
292.500
315.000
337.500
( l.JOOOMPSM 2.7000MPSH 4.4000MPSH 6.7000HPSM 9.5000MPSM12.7000KPS)
0.00023000
0.00015000
0.00003000
0.00012000
0.00023000
0.00012000
0.00027000
0.00017000
0.00014000
0.00006000
0.00016000
0.00016000
0.00025000
0.00021000
0.00012000
0.00012000
0.00064000
0.00016000
0.00014000
0.00034000
0.00057000
0.00043000
0.00096000
0.00041000
0.00032000
0.00021000
0.00018000
0.00062000
0.00091000
0.00032000
0.00043000
0.00032000
0.00025000
0.00009000
0.00009000
0.00030000
0.00073000
0.00112000
0.00210000
0.00041000
0.00021000
0.00009000
0.00016000
0.00055000
0.00123000
0.00062000
0.00046000
0.00027000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0 . 00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
-------
•••• ISCLT ••••
NESCAUH Benzene Risk Assessment
- ISCLT INPUT DATA (COOT.) -
PAGE
J ••••
n
00
- FREQUENCY OP OCCURRENCE OF WIND SPEED, DIRECTION AND STABILITY -
SEASON 1
STABILITY CATEGORY 3
WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
DIRECTION
(DEGREES)
0.000
22.500
45.000
67.500
90.000
112.500
135.000
157.500
180.000
202.500
225.000
247.500
270.000
292.500
315.000
337.500
I 1.3000MFSH
0.00010000
0.00009000
0.00013000
0.00004000
0.00013000
0.00012000
0.00006000
0.00002000
0.00010000
0.00017000
0.00008000
0.00007000
0.00013000
0.00013000
0.00009000
0.00008000
2.7000HPSH 4.4000MPSM 6.7000HP3M
0. 00110600
0.00037000
0.00030000
0.00025000
0.00064000
0.00043000
0.00073000
.00288000 0.00018000
.00116000
.00100000
.00178000
.00559000
.00518000
.00571000
0.00039000 0.00116000
0.00057000 0.00237000
0.00050000 0.00091000
0.00064000 0.00285000
0.00096000 0.00400000
0.00171000 0.00749000
0.00130000 0.00502000
O.OOOS4000 0.00379000
.00007000
.00030000
.00041000
.00281000
.00253000
.00130000
.00021000
.00050000
.00057000
.00078000
.00153000
.00279000
.00139000
.00130000
0.00064000 0.00253000 0.00064000
9.SOOOMPS) (12.7000HPS
.00000000
.00000000
.00002000
.00000000
.00011000
0.00009000
0.00002000
0.00000000
0.00007000
0.00011000
0.00014000
0.00005000
0.00021000
0.00007000
0.00007000
0.00009000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00002000
0.00002000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
SEASON 1
STABILITY CATEGORY 4
WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
DIRECTION
(DEGREES)
0.000
22.500
45.000
67.500
90.000
112.500
135.000
157.500
180.000
202.500
225.000
247.500
270.000
292.500
315.000
337.500
i 1.3000MPSH 2.7000HPSM 4.4000HPSM 6.7000MPSH 9.5000MPS( (12.7000MPS
0.00092000
0.00057000
0.00081000
0.00062000
0.00097000
0.00060000
0.00052000
0.00040000
0.00093000
0.00024000
0.00024000
0.00013000
0.00037000
0.00016000
0.00032000
0.00035000
0.00708000
0.00420000
0.00477000
0.00413000
0.00550000
0.00379000
0.00420000
0.00311000
0.00575000
0.00256000
0.002S8000
0.00203000
0.00228000
0.00242000
0.00279000
0.00295000
0.02487001
0.00817000
0.00922000
0.01055000
0.01621000
0.01212000
0.00948000
0.00632000
0.01824000
0.01397000
0.01875000
0.01850000
0.01512000
0.01603000
0.01468000
0.01299000
0.02450000
0.00895000
0.01226000
0.01050000
0.01731000
0.01505000
0.00562000
0.00242000
0.01363000
0.02199000
0.03156001
0.03843001
0.04660001
0.04631001
0.03491001
0.02053001
0.00345000
0.00192000
0.00527000
0.00253000
0.00358000
0.00206000
0.00064000
0.00039000
0.00176000
0.00343000
0.00425000
0.00379000
0.00774000
0.00996000
0.00772000
0.00295000
0.00059000
0.00030000
0.00180000
0.00094000
0.00155000
0.00046000
0.00016000
0.00018000
0.00064000
0.00046000
0.00048000
0.00037000
0.00158000
0.00212000
0.00162000
0.00053000
-------
•••• ISCLT
NESCMJM Benzene Risk Assessment
- ISCLT INPUT DATA (COOT.) -
PACE
- FREQUENCY OF OCCURRENCE OF WIND SPEED. DIRECTION AND STABILITY -
SEASON 1
STABILITY CATEGORY 5
WIND SPEED HIND SPEED HIND SPEED HIND SPEED HIND SPEED HIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
DIRECTION
(DEGREES)
0.000
22.500
45.000
67.500
90.000
112.500
135.000
157.500
180.000
202.500
225.000
247.500
270.000
292.500
915.000
337.500
( 1.3000MPSM 2.7000MPSH 4.4000MPSH 6.7000MPSH 9.5000MPSM12.7000MPS!
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00285000
0.00169000
0.00121000
0.00110000
0.00162000
0.00169000
0.00137000
0.00178000
0.00292000
0.00219000
0.00201000
0.00158000
0.00180000
0.00167000
0.00123000
0.00116000
0.00806000
0.00064000
0.00064000
0.00059000
0.00169000
0.00174000
0.00206000
0.00128000
0.00395000
0.00550000
0.01105000
0.01160000
0.01123000
0.01818000
0.01318000
0.00847000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0 .00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
SEASON 1
STABILITY CATEGORY 6
HIND SPEED HIND SPEED HIND SFEED HIND SPEED HIND SPEED HIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
DIRECTION
(DEGREES)
0.000
22.500
45.000
67.500
90.000
112.500
13S.OOO
157.500
180.000
202.500
225.000
247.500
270.000
292.500
315.000
337.500
( 1.3000HPSH 2.7000KPSH 4.4000MPSH 6.7000MPSH 9 . SOOOMPS ) ( 1 2 . 7000MPS
0.00114000
0.00081000
0.00047000
0.00061000
0.00057000
0.00062000
0.00062000
0.00065000
0.00147000
0.00082000
0.00083000
0.00078000
0.00098000
0.00092000
0.00078000
0.00056000
0.00329000
0.00139000
0.00071000
0.00082000
0.00132000
0.00112000
0.00130000
0.00130000
0.00386000
0.00290000
0.00409000
0.00306000
0.00281000
0.00338000
0.00301000
0.00210000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0 . OOOPOOOO
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
0.00000000
-------
•••• ISCLT NESCAUM Benzmw Risk Assessment •••••... ffaB
- ISCLT INPUT DATA (COOT.) -
- VERTICAL POTENTIAL TEMPERATURE GRADIENT (DEGREES KELVIN/METER) -
WIND SPEED HIND SPEED WIND SPEED WIND SPEED WIND SPEED WIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
STABILITY CATEGORY 10.OOOOOOE+000.OOOOOOE+000. OOOOOOE+000.OOOOOOE+000. OOOOOOE-f000. OOOOOOE+00
STABILITY CATEGORY 20.OOOOOOE-f000.OOOOOOE+000.OOOOOOE+000.OOOOOOE-f000.OOOOOOE-f000.OOOOOOE-f00
STABILITY CATEGORY 30.OOOOOOE+000.OOOOOOE+000.OOOOOOE+000.OOOOOOE+000.OOOOOOE+000.OOOOOOE-f00
STABILITY CATEGORY 40.OOOOOOE+000.OOOOOOE+000.OOOOOOE+000.OOOOOOE+000.OOOOOOE+000.OOOOOOE+00
STABILITY CATEGORY S0.200000E-OI0.200000E-010.200000E-010.200000E-010.200000E-010.200000E-01
STABILITY CATEGORY 60.350000E-010.3SOOOOE-010.350000E-010.3SOOOOE-010.350000E-010.350000E-01
- HIND PROFILE POWER LMf EXPONENTS -
HIND SPEED HIND SPEED WIND SPEED HIND SPEED HIND SPEED HIND SPEED
CATEGORY 1 CATEGORY 2 CATEGORY 3 CATEGORY 4 CATEGORY 5 CATEGORY 6
STABIUTY CATEGORY 10.150000E+000.1SOOOOE+000.1SOOOOE+000.150000E+000.1SOOOOE+000.150000E+00
STABILITY CATEGORY 20.150000E+000.150000E+000.150000E+000.150000E+000.150000E+000.150000E+00
STABILITY CATEGORY 30.200000E+000.200000E+000.200000E+000.200000E+000.200000E+000.200000E+00
STABILITY CATEGORY 40.2SOOOOE+000.250000E+000.2SOOOOE+000.2SOOOOE+000.2SOOOOE+000.250000E+00
STABILITY CATEGORY SO. 300000E+000.300000E+000.300000E+000.300000E+000.300000E+000.300000E+00
STABILITY CATEGORY 60.300000E+000.300000E+000.300000E+000.300000E+000.300000E+000.300000E+00
-------
••" ISCLT ••••«» HcscMJN Benrene Risk Assessment ••• PAGE
- SOURCE INPUT DATA -
C T SOURCE SOURCE X Y EMISSION BASE /
A A NUMBER TYPE COORDINATE COORDINATE HEIGHT EICV- / - SOURCE DETAILS DEPENDINQ ON TYPE -
R P (M) (M) (M) ATION /
D E IN) /
X 1 VOLUME 0.00 0.00 0.00 0.00 STANDARD DEVIATION OT THE CROSSWIND SOURCE DISTRIBUTION 1M)= 0.15
STANDARD DEVIATION OF THE VERTICAL SOURCE DISTRIBUTION IH>" 0.10
- SOURCE STRENGTHS ( GRAMS PER SEC )
SEASON 1 SEASON 2 SEASON 3 SEASON 4
l.OOOOOE+00
O
-------
•••• ISCLT HESCAUH Benzene Risk Assessment
*»••!•*•
•• ANNUM. GROUND LEVEL CONCENTRATION < KICROGRAMS PER CUBIC METER
- GRID SlfSTEM HECEPTDRS -
- X AXIS (RANGE , METERS)
100.000 200.000
- CONCENTRATION -
ALL SOURCES COMBINED
10.000 SO.000
V AXIS (AZIMUTH BEARING. DEGREES )
9
350.000
340.000
3)0.000
320.000
310.000
300.000
290.000
280.000
270.000
260.000
250.000
240.000
230.000
220.000
210.000
200.000
190.000
180.000
170.000
160.000
150.000
140.000
130.000
120.000
110.000
100.000
90.000
80.000
70.000
60.000
50.000
40.000
30.000
20.000
10.000
0.000
178.241577
123.046059
114.000771
130.487625
148.311783
157.640228
171.341080
186.722946
207.201248
166. 7702 13
138.695053
128.662445
130.221832
130.12225)
127.835846
156.981873
227.523911
309.285004
262.037170
219.129684
232.119461
262.442719
303.680389
329.672882
346.572876
339.023956
345.898987
310.488556
295.415110
289.277100
285.230499
262.041351
233.695801
210.515045
226.778976
245.949966
70.061401
47.028492
44.400913
50.918404
57.415162
61.271187
66.157845
72.424004
79.225426
65.500221
53.229607
49.920199
50.517918
50.554039
49.797794
59.803471
89.112446
120.523180
102.622963
84.839005
90.625519
103.213727
118.557487
129.459152
135.991516
132.548279
131.322693
121.109013
115.194118
113.277550
111.764259
103.203903
91.386696
82.150238
88.975266
95.649719
19.013832
12.458734
11.944114
13.707568
15.346677
16.418854
17.629908
19.362690
20.916403
17.518650
14.109665
13.353589
13.519629
13.S40294
13.385107
15.779002
24.093996
32.43B164
27.741705
22.721668
24.446255
28.027493
31.985006
35.105518
36.858227
35.781082
34.526657
32.714825
31.010518
30.610434
30.254345
28.061289
24.694801
22.159197
24.108721
25.725828
.0(5299
.2(8217
.171151
.630633
.044761
.333225
.638253
.098501
.467931
.613135
.700694
.519601
.562915
.573672
.546695
.145143
.416141
.621946
.391904
.028219
.512109
.494414
.524451
.384795
.851379
.526522
.047768
.698159
.253644
.169125
.081341
.516406
.590228
.907900
.442310
.849172
-------
•••• ISCLT NESCAUH Benzene Risk Assessment
PAGE
•• ANNUAL GROUND LEVEL CONCENTRATION ( MICROGBAMS PER CUBIC METER ) FROM ALL SOURCES COMBINED (CONT.) •*
- PROGRAM DETERMINED MAXIMUM 10 VALUES -
X
COORDINATE
RANGE
(METERS)
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
30.00
V
COORDINATE
AZIMUTH
BEARIN3
(DEGREES)
110.00
90.00
100.00
120.00
80.00
180.00
130.00
70.00
60.00
50.00
CONCENTRATION
346.572876
345. 898987
339.023956
329.672882
310.488556
309.285004
301.680389
295.425110
289.277100
285.230499
n
!—•
LO
-------
END OP ISCLT PROGRAM. 1 SOURCES PROCESSED
O
-------
APPENDIX D
DERIVATION OF QUANTITATIVE RISK ASSESSMENT CRITERIA
D-l
-------
D-2
-------
APPENDIX D
DERIVATION OF QUANTITATIVE RISK
ASSESSMENT CRITERIA
IDENTIFICATION OF CRITICAL HEALTH EFFECTS STUDIES
In the following discussion, critical studies are identified for the health effects
which have been reviewed in the previous chapters. No observed adverse effect levels and
lowest observed adverse effect levels are determined They are presented in a standard
mg/kg/day, human equivalent dose form to allow for comparisons. Gasoline is evaluated
first, followed by benzene, toluene, and xylene. For each substance, critical health effects
studies for each significant health endpoint are identified for acute, subacute/subchronic,
and chronic exposure durations.
GASOLINE
Principal adverse health effects associated with gasoline exposure are: lethality,
kidney and liver toxicity, pulmonary toxicity, neurotoxicity, hematotoxicity, cancer, and
reproductive and developmental effects. Studies relevant to the derivation of quantitative
health criteria for these effects are presented below.
Lethality
Effects of Acute Exposure
One human case study determined a single lethal dose for the oral ingestion of
gasoline of 140 mg/kg (Machle, 1941). Two studies (Wang and Irons, 1961; Ainsworth,
1960) estimated lethal doses of gasoline vapors following exposures for approximately 5 to
10 minutes; an aircraft mechanic exposed to between 5,000 and 16,000 ppm of gasoline,
and a 3 year old boy exposed to at least 10,000 ppm of gasoline vapor. If it is assumed
that the mechanic was exposed to 5,000 ppm (approximately 15,000 mg/ m3), and that his
alveolar ventilation was two-thirds of his total ventilation (1.2 m^/hour), his inhaled dose is
determined as follows:
15,000 mg/m3 x 0.8 m3/hr x 0.08 hrs/ 70kg = 14 mg/kg/5 minutes
= 3 mg/kg/minute
D-3
-------
For the 3-year-old boy, assuming a weight of 15 kg, and an alveolar ventilation rate of
two-thirds of his total ventilation (4 m-Vday; 0.17 m^/hr) the lethal dose estimate is as
follows:
30,000 mg/m3 x 0.1 m3/hr x 0.08 hrs/ 15 kg = 16 mg/kg/5 minutes
= 3 mg/kg/minute
Thus, the lowest effect level for acute lethality in humans is 3 mg/kg/min. While
this level is based on limited information, it may be significant to note that two cases on
accidental inhalation exposures produced similar results. Also, the human case studies
support the animal studies which indicate that inhalation of gasoline vapors in more toxic
than oral ingestion (Berkow, 1982).
The acute oral lethality (LDso) of gasoline in Sprague-Dawley rats was estimated to
be 18.75 ml/kg (Elars Bioresearch Laboratories, 1980). Assuming a density for gasoline
of 1 gram/ml, the LDso (expressed as a weight to weight ratio) is 18.75 g/kg. Gerarde
(1963) noted that two male albino Wister rats receiving 0.2 ml of gasoline intratracheally
died almost immediately. Assuming a weight for these rats of 250 grams, this exposure
corresponds to a dose of 200 mg/0.25 kg = 800 mg/kg.
These animal studies provide limited information concerning the acute lethal effects
of exposure to gasoline vapors. For both exposure routes, the human studies indicate
lower lethal doses than do the animal studies. Thus, lethality criteria for both oral and
inhalation exposures should be based on the human data: the lowest observed acute lethal
level is 140 mg/kg for oral ingestion, and about 15 mg/kg (or 3 mg/kg/minute for 5
minutes) for inhalation exposure.
Pulmonary Toxicitv
Effects of Acute Exposure
There is very limited information regarding the acute effects of gasoline exposure
on pulmonary toxicity in human beings. The Ainsworth (1960) case study noted above
observed edema alveolar and hemorrahage, as well as hyperemia of the trachea and bronchi
of a 3 year old boy who died from gasoline vapor poisoning. This exposure was at least
21 mg/kg. Sprague-Dawley rats exposed to a single oral dose 10 ml/kg (approximately 10
grams/kg) of gasoline developed mild irritation and congestion of the lung (Elars
Bioresearch Laboratories, 1980). Higher doses (up to approximately 25 grams/kg)
produced more marked edema in the lower lung.
D-4
-------
CD-I mice were exposed to vaporized unleaded gasoline at a concentration of 104
mg/liter for 6 minutes (1 minute exposure, followed by a 10 minute recovery, followed by
a 5 minute exposure); (Phillips, 1984). Significant decreases in respiratory rate (indicative
of a significant sensory irritant response) were observed in most of the mice. Assuming a
weight of 25 grams and an initial respiratory rate of 0.02 liters/min (based on Guyton's
equation (Guyton, 1947), this exposure corresponds to a dose of 500 mg/kg/6 minute
exposure, or about 80 mg/kg/minute for 6 minutes.
LOAEL (Sensory Irritation - Unleaded Gasoline - Mice)
104 mg/liter x 0.02 liters/min x 6 rnin/0.025 kg = 500 mg/kg/6 minutes
= 80 mg/kg/minute
Because the gasoline caused up to a 50 percent reduction in the respiratory rate, the actual
dose may have been lower by as much as 50 percent. Yet, even assuming a lower dose
(e.g., 250 mg/kg), this study indicates that rodents are less sensitive than the human beings
to the acute pulmonary toxicity of gasoline.
Effects of Subacute and Subchronic Exposures
No subacute or subchronic exposure data were identified which are relevant to the
assessment of pulmonary toxicity. Pulmonary function changes (increased minute
volumes) were observed in male monkeys exposed to 374 ppm of leaded gasoline or 1,552
ppm of unleaded gasoline for 6 hrs/day, 5 days/week, for 13 weeks (Kuna and Ulrich,
1984). No pulmonary changes were observed in these monkeys following exposures to
either 103 ppm of leaded gasoline or 384 ppm of unleaded gasoline. Female monkeys also
showed pulmonary changes only at the highest exposure groups. Leaded fuel exposure at
374 ppm produced a decreased tidal volume while unleaded fuel exposure at 1,552 ppm
produced a decreased respiratory rate. Assuming a weight of 5 kilograms for the monkeys,
and that 1 ppm of gasoline vapors equals 3 mg/m^, a LOAEL and a NOAEL for these
pulmonary function changes can be calculated as follows:
LOAEL (Ventilation changes - Leaded Gasoline - Monkeys)
1122 mg/m3 x 0.001 mVmin x 60 min/hr x 6 hr/day x 5 days/7 days/5 kg
= 58 mg/kg/dav
D-5
-------
NOAEL (Ventilation changes - Leaded Gasoline - Monkeys)
309 mg/ 3 x 0.001 m3/min x 60 min/hr x 6 hr/day x 5 day/7days/5 kg
= 16 mg/kg/day
LOAEL (Ventilation changes - Unleaded Gasoline - Monkeys)
4656 mg/m3 x 0.001 m3/min x min/hr x 6 hr/day x 5 days/7days
= 239 mg/kg/dav
NOAEL (Ventilation changes - Unleaded Gasoline - Monkeys)
1152 mg/m3 x 0.001 m3/min x 60 x 6hr/day x 5 days/7 days/5 kg
= 59 mg/kg/dav
Le Mesurier et al. (1980) and Lykke et al. (1979) observed reduced pulmonary
surfactant levels followed by irreversible fibrotic and schlerotic changes in the alveoli of
rats as a result of inhalation exposures to leaded gasoline. Exposures took place for 5
days/week for up to 12 weeks. Decreased surfactant levels were observed after only 5
days of exposure. Lung structure changes were observed after 6 weeks of exposure. The
dose corresponding to this effect is 53 mg/kg/day.
LOAEL (Lung Pathology - Rats)
300 mg/m3 x 0.13 x 10'3 m3/min x 60 min/hr x 8 hr/day x 5 days/7 days/0.25 kg
= 53 mg/kg/day
While this LOAEL is similar to the one found by Kuna and Ulrich (1984) for
pulmonary function changes, its interpretation is limited by the lack of a dose-response
curve and by the lack of experimental results for unleaded gasoline. Also, the lung
structure changes observed by Lykke et al. (1979) are more significant than pulmonary
function changes, in that they indicate the potential for irreversible pulmonary damage to
occur as a result of subchronic exposure to gasoline vapors. Thus, despite the limitations
of this study, it warrants consideration because it investigates a sensitive health endpoint.
Effects of Chronic Exposure
No studies were identified in the literature with adequate quantitative information to
assess the effects of chronic gasoline exposures on the respiratory system.
D-6
-------
Kidnev Toxicitv
Effects of Acute Exposure
No human or animal data were identified which are relevant to the assessment of
acute exposure to gasoline on the kidney.
Effects of Subacute and Subchronic Exposures
A LOAEL of 29 ppm for unleaded gasoline vapors, a LOAEL of 34 ppm for full
range alkylate naphtha (FRA), and a NOAEL of 3 ppm for full range alkylate naphtha were
determined for Fischer 344 rats exposed to these mixtures via inhalation for 21 days
(Haider et al., 1984). No significant kidney toxicity was observed in female rats. The
exposure regimen included 10 or 20 animals per group, and exposures lasted for 6 hours
per day, 5 days per week. The conversion of these exposures to a mg/kg/day dose is
derived as follows (assuming that 1 ppm of the gasoline fractions equals 3 mg/m3 and a
weight for the rats of 135 grams):
LOAEL (Kidney Toxicity - Unleaded Gasoline - Male Rats)
87 mg/m3 x 0.1 liters/min x 0.001 m^/l x 60 min/hr x 6 hrs x 5 days/7days/
0.135 kg = 17 mg/kg/dav
LOAEL (Kidney Toxicity - Full range alkylate naphtha - Male Rats)
102 mg/m3 x 0.1 liters/min x 0.001 m3/l x 60 min/hr x 6 hrs x 5 days/
7 days/0.135kg = 19mg/kg/dav
NOAEL (Kidney Toxicity - Full range alkylate naphtha - Male Rats)
9 mg/m3 x 0.1 liters/min x 0.001 m^/l x 60 min/hr x 6 hrs x 5 days/7 days/
0.135kg=1.7mg/kg/dav
Because these effects are systemic, a scaling factor is needed in order to estimate the
human equivalent dose. This factor, as previously presented, is proportional to perfusion
if flow limited clearance is assumed. The use of liver perfusion as an estimator of
metabolic and clearance differences between species results in a scaling factor based on
body weight raised to the 0.74 power (Fiserova-Bergerova and Hughes, 1983). The
scaling factor relevant to these subchronic kidney toxicity studies can be derived by using
the following equations.
D-7
-------
DA x BWA/BWA0-74 = DH x BWH/BWR0.74
DA( 135/1 350-74) = DH x (70,000/70,0000-74)
Where:
DA = Dose to the animal
DH = Dose to the human
= animal body weight
= human body weight
Thus LOAELs and NOAELs determined in the Haider et al. (1984) should be scaled by a
factor of 0.2 to reflect the human equivalent dose.
LOAEL (Kidney Toxicity - Unleaded Gasoline - Male Rats; Human Equivalent Dose)
17 mg/kg/day x 0.2 = 2.6 mg/kg/dav
LOAEL (Kidney Toxicity - Full Range Alkylate Naphtha - Male Rats;
Human Equivalent Dose)
19 mg/kg/day x 0.2 = 4 mg/kg/day
NOAEL (Kidney Toxicity - Full Range Alkylate Naphtha - Male Rats;
Human Equivalent Dose)
1.7 mg/kg/day x 0.2 = 0.3 mg/kg/day
The kidney toxicity of more volatile gasoline fractions (e.g., those most likely to
represent the evaporative emissions from gasoline stations) has also been evaluated.
Haider et al. (1986b) examined the nephrotoxic effects of a 25 percent blend of n-butane,
n-pentane, isobutane, and isopentane in Sprague-Dawley rats. Rats were exposed to up to
4,437 ppm of the mixture for 6 hours/day, 5 days/week, for 3 weeks, with no effect. The
NOAEL for this experiment is calculated as follows:
D-8
-------
NOAEL (Kidney Toxicity - 25% mixture - Rats)
13,311 mg/m3 x 0.1 liters/min x 0.001 m3/! x 60 min/hr x 6 hrs x 5 days/7 days/0.25 kg
= 1370 mg/kg/day
The human equivalent dose is derived by multiplying this dose level by 0.2.
NOAEL (Kidney Toxicity - 25% mixture - Rats; Human Equivalent Dose)
1370 mg/kg/day x 0.2 = 274 mg/kg/day
Subchronic exposure effects of the 0 to 145°F distillation fraction of unleaded
gasoline were studied using Fischer 344 rats (Aranyi et al., 1986). Exposures were to
either 1,209 or 5,229 ppm of this fraction, 6 hours/day, 5 days/week, for up to 91 days. A
slight, but significant increase in the relative kidney weights of the female rats was
observed following the high exposure regimen; no other statistically significant differences
in kidney or liver weights were observed for any other treatment group. Thus, the 5,229
ppm represents a LOAEL for the female rats and a NOAEL for the male rats, while the
1,209 ppm exposure represents a NOAEL for the female rat
LOAEL (Kidney Toxicity - 0 - 145°F Distillation Fraction - Female Rats)
NOAEL (Kidney Toxicity - 0 - 145°F Distillation Fraction - Male Rats)
15,687 mg/m3 x 0.1 liters/min x 0.001 m3/! x 60 min/hr x 6 hrs x 5 days/7 days/0.25 kg
= 1610 mg/kg/day
NOAEL (Kidney Toxicity - 0 - 145°F Distillation Fraction - Female Rats)
3,627 mg/m3 x 0.1 liters/min x 0.001 m3/! x 60 min/hr x 6 hrs x 5 days/7 days/0.25 kg
= 370 mg/kg/day
The human equivalent doses are derived by multiplying these levels by 0.2
D-9
-------
LOAEL (Kidney Toxicity - 0 - 145°F Distillation Fraction - Female Rats;
Human Equivalent Dose)
NOAEL (Kidney Toxicity - 0 - 145°F Distillation Fraction - Male Rats;
Human Equivalent Dose)
1610 mg/kg/day x 0.2 = 320 mg/kg/day
NOAEL (Kidney Toxicity - 0 - 145°F Distillation Fraction - Female Rats;
Human Equivalent Dose)
370 mg/kg/day x 0.2 = 74 mg/kg/day
Effects of Chronic Exposure
The chronic exposure human data are inadequate to use for quantitative dose
response assessment A LOAEL of 67 ppm for kidney toxicity was identified in a 2-year
study on male Fischer 344 rats (MacFarland et al., 1984). Two female rats exposed to
unleaded gasoline vapors exhibited medullary mineralization following a chronic exposure
to 2,056 ppm, but not following exposure to either 292 or 67 ppm. Exposures were for 6
hours per day, 5 days per week. The chronic LOAEL and NOAEL for the male and female
rats are calculated as follows, assuming a rat body wight of 250 grams.
LOAEL (Kidney Toxicity - Unleaded Gasoline - Male Rats)
200 mg/m3 x 0.1 liters/ min. x 0.001 m3/liter x 60 min/hr x 6 hours/day x
5 days/7days/0.25kg = 21 mg/kg/day
NOAEL (Kidney Toxicity - Unleaded Gasoline - Female Rats)
876 mg/m3 x 0.1 liters/min. x 0.001 m3/liters x 60 min/hr x 6 hrs/day x
5 days/7 days/0.25 kg = 88 mg/kg/dav
For these experiments, the scaling factor required to produce a human equivalent
dose is also about 0.2 (i.e., there is no appreciable difference in the scaling factors based
on 135 gram rats versus 250 gram rats). Thus, human equivalent doses for these effect
levels are derived as follows.
D-10
-------
LOAEL (Kidney Toxicity - Unleaded Gasoline - Male Rats;
Human Equivalent Dose)
21 mg/kg/day x 0.2 = 4 mg/kg/day
LOAEL (Kidney Toxicity - Unleaded Gasoline - Female Rats;
Human Equivalent Dose)
634 mg/kg/day x 0.2 = 127 mg/kg/day
NOAEL (Kidney Toxicity - Unleaded Gasoline - Female Rats;
Human Equivalent Dose)
88 mg/kg/day x 0.2 = 18 mg/kg/day
It is important to note that the female rats are far less sensitive to the effects of
unleaded gasoline vapors than are the male rats. It is currently unknown the degree to
which this disparity reflects interspecies or interindividual differences in nephrotoxic
responses to unleaded gasoline exposure. It is also important to note that the chronic
LOAEL for the male rats is higher than the subchronic LOAEL (see above). The LOAEL
exposure concentration in this chronic study (67 ppm), however, was the lowest level
tested. Thus, it is possible that testing at lower concentrations under this exposure regimen
would have also produced kidney toxicity.
Neurotoxicitv
Effects of Acute Exposure
Limited data on humans exist concerning the effects of acute exposure to gasoline
vapors on sensory irritation and central nervous system depression. The effective doses
vary with exposure concentration and exposure duration. Dizziness and deep anesthesia
were produced in adults following gasoline vapor exposure of 10,000 ppm for 4 to 10
minutes (e.g., a potentially lethal exposure) (Poklis and Burkett, 1977). Nose and throat
irritation were observed after 2 minutes. Exposure to 500 to 1,000 ppm of gasoline vapors
for 0.5 to 1 hour produced varying degrees or irritation and slight dizziness (Drinker et al.,
1943; Machle, 1941; Poklis and Burkett, 1977). Subjects exposed to 200 ppm of gasoline
D-ll
-------
vapors for 30 minutes experienced eye irritation. These exposures correspond to the
following doses.
LOAEL (Dizziness, Deep Anesthesia - Humans)
10,000 ppm for 4 minutes
30,000 mg/m3 x 0.013 m3/min x 4 minutes/70 kg = 22 me/kg
= 6 mg/kg/min
LOAEL (Nose and Throat Irritation - Humans)
10,000 ppm for 2 minutes
30,000 mg/m3 x 0.015 mg3/min x 2 minutes/70 kg = 13 me/kg
= 6 mg/kg/min
LOAEL (Irritation, Slight Dizziness- Humans)
500 ppm exposure for 0.5 hour
1500 mg/m3 x 0.015 m3/min x 30 min/70 kg = 10 mg/kg
= 0.3 mg/kg/min
Subjects exposed to 140 to 270 ppm of gasoline vapors for 8 hours experienced eye
irritation. The lowest dose level eliciting this response for this exposure duration is
described below.
LOAEL (Eye Irritation - Humans)
200 ppm for 0.5 hours (eye irritation)
600 mg/m3 x 0.015 m3/min x 30 minutes/70 kg = 4 me/kg
= 0.1 mg/kg/min
LOAEL (Eye Irritation, Some Gastrointestinal Discomfort - Humans)
140 ppm for 8 hours
420 mg/m3 x 0.015 m3/min x 480 minutes/70 kg = 43mg/kg
= 0.1 mg/kg/min
D-12
-------
Thus, it appears on the basis of this information, that concentration, not exposure
duration, is critical when assessing the acute neurotoxic effects of gasoline vapor exposure
on humans.
Exposure studies in animals provide no additional insight regarding the acute
neurotoxic effects of gasoline. A single oral dose of 15 to 25 grams (approximately 5
grams/kg) produced narcosis in rabbits (Lewis, 1888, as cited in Browning, 1953). A
dose of 0.1 milliliters of unleaded gasoline applied to the right eye of the rabbit produced
no eye irritation (Elars Bioresearch Laboratories, 1982).
Effects of Subacute and Subchronic Exposures
Only animal data are available for assessing the subacute and subchronic exposure
neurotoxic effects of gasoline. A dose of 1 ml of unleaded gasoline per 100 grams body
weight in rats (approximately 10 grams/kg) produced no brain wave
(electroencephalogram) changes after ten days of exposure (Saito, 1973). Significant
changes in electroencephalogram readings were observed in rats, however, after a similar
exposure to leaded gasoline. The authors attributed the effects of this exposure to the
tetraethyl lead present in the leaded gasoline. Thus, 10 grams/kg represents a NOAEL for
unleaded gasoline and a LOAEL for leaded gasoline.
Exposure of squirrel monkeys to either 1552 ppm of unleaded gasoline or 374 ppm
of leaded gasoline produced no changes in the visual evoked response after subchronic
exposure (Kuna and Ulrich, 1984). The exposure regimen was for 6 hrs/day, 5
days/week, for 90 days. The inhaled doses for these NOAELs are derived as follows.
NOAEL (Visual Evoked Response - Unleaded Gasoline - Squirrel Monkey)
4656 mg/m3 x 1.2 liters/min x 0.001 m3/liter x 60 min/hr x 6 hrs/day x 5 days/7 days/5 kg
= 287 mg/kg/day
NOAEL (Visual Evoked Response - Leaded Gasoline - Squirrel Monkey)
1122 mg/m3 x 1.2 liters/min x 0.001 m3/liter x 60 min/hr x 6 hrs/day x 5 days/7 days/5 kg
= 69 me/kg/dav
D-13
-------
Because these are systemic effects, scaling is required to estimate a human equivalent dose
from the monkey studies. This scaling factor is derived in the same manner as it was for
the rat.
DA x BWA0.74 = DH x BWR0-74
DA x (5,000/5,000°-74) = DH x 70,000/70,0000-
DA x 9.2 = DHx 18.1
/ , ).74
= DHxl8.1
DH = 0.5
Applying this scaling factor results in the following human equivalent doses.
NOAEL (Visual Evoked Response - Unleaded Gasoline; Human Equivalent Dose)
257 mg/kg/day x 0.5 = 144 mg/kg/day
NOAEL (Visual Evoked Response - Leaded Gasoline; Human Equivalent Dose)
69 mg/kg/day x 0.5 = 35 mg/kg/day
Effects of Chronic Exposures
No adequate chronic exposure data were available concerning the neurotoxic effects
of gasoline on humans. Studies on human exposure to solvent mixtures indicate an
enhanced effect of the mixture on sensory irritation and central nervous system depression
relative to what would be predicted on the basis of the individual chemicals. Enhanced
central nervous system toxicity was observed when aromatic solvents (e.g., benzene,
toluene, and xylene) were present in a mixture. For the purposes of this assessment,
however, it is assumed that the effects of chemicals in a mixture is additive, especially
when considering low level, ambient air exposures. Thus, public health protection is
assumed to be provided if ambient air exposure criteria are derived for the individual
aromatic compounds. While the lack of understanding regarding possible chemical
interactions at low exposure levels does present a data gap, adequate information does not
currently exist to translate this concern into quantitative criteria.
Fischer 344 rats were exposed to 2056 ppm of unleaded gasoline, 6 hrs/day, 5
days/week, for 8 to 18 months (Spencer et al., 1982). No functional changes were
observed in the animals, nor was there any histopathological evidence of distal neuropathy.
This exposure regimen, therefore, represents a no observed adverse effect level for these
D-14
-------
parameters in these animals under the condition of the test The dose is calculated as
follows.
NOAEL (Peripheral neuropathy, Behavior changes - Unleaded Gasoline - Rats)
6168 mg/m3 x 0.1 liters/min x 0.001 m^/liter x 60 min/hr x 6 hrs/day x 5 days/7 days
70.25 kg = 635 mg/kg/dav
The human equivalent dose for this endpoint is obtained by scaling this dose by a factor of
0.2.
NOAEL (Peripheral neuropathy, Behavior changes - Unleaded Gasoline - Rats;
Human Equivalent Dose)
635 mg/kg/day x 0.2 = 127 mg/kg/day
Hematotoxicitv
Effects of Acute Exposure
No human or animal data were found on the hematotoxic effects of acute gasoline
exposures.
Effects of Subacute and Subchronic Exposures
No human data were found concerning the hematotoxic effects of subacute or
subchronic exposure to gasoline. Kuna and Ulrich (1984) observed a statistically
significant hematological changes in rats subchronically exposed to leaded gasoline vapors
at a concentration of 374 ppm, but not at 103 ppm. Elevated thrombocyte counts in male
rats and elevated reticulocyte counts in female rats were observed following subchronic
exposure to unleaded gasoline vapors at 1552 ppm, but not at 384 ppm. Exposures were 6
hrs/day, 5 days/week for 13 weeks. From these data, LOAELs and NOAELs for leaded
and unleaded gasoline can be estimated.
LOAEL (Changes in red blood cell, Decreased WBC -Leaded Gasoline- Rats)
1122 mg/m3 x 0.1 L/min x 0.001 m$/L x 60 min/hr x 6 hr/day x 5 days/7days/ 0.25 kg
= 116 mg/kg/dav
D-15
-------
NOAEL (Changes in red blood cell, Decreased WBC -Leaded Gasoline- Rats)
309 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hr/day x 5 days/7days/ 0.25 kg
= 32 mg/kg/day
LOAEL (Elevated thrombocyte (M) and redculocyte counts (F) -Unleaded Gasoline - Rats)
4656 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7 days/0.25 kg
= 480 mg/kg/dav
NOAEL (Elevated thrombocyte (M) and reticulocyte counts (F) -Unleaded Gasoline - Rats)
1152 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7 days/0.25 kg
= 117 mg/kg/dav
Scaling adjustments to these doses generate the following human equivalent doses.
LOAEL (Changes in red blood cell, decreased WBC -Leaded Gasoline- Rats;
Human Equivalent Dose)
116 mg/kg/day x 0.2 = 23 mg/kg/day
NOAEL (Changes in red blood cell, decreased WBC -Leaded Gasoline- Rats;
Human Equivalent Dose)
32 mg/kg/day x 0.2 = 6.4 mg/kg/day
LOAEL (Elevated thrombocyte (M) and reticulocyte counts (F) -Unleaded Gasoline - Rats;
Human Equivalent Dose)
480 mg/kg/day x 0.2 = 96 mg/kg/day
D-16
-------
NOAEL (Elevated thrombocyte (M) and rcticulocyte counts (F) -Unleaded Gasoline - Rats;
Human Equivalent Dose)
117 mg/kg/day x 0.2 = 23 mg/kg/day
Effects of Chronic Exposure
As presented in the gasoline toxicity section, several reports link human exposure to
hydrocarbon solvents with various adverse hematological changes (e.g., leucopenia,
decreased antibody tilers, red blood cell abnormalities). There are also some indications
that exposure to certain of these solvents may lead to chemical hypersensitivity, particularly
autoimmune disorders. There are, however, no adequate human or animal data on which
to base quantitative dose-response estimates for such effects as they relate to gasoline
exposure. At the present time, assessment of hematoxicity is restricted to the evaluation of
individual compounds.
Genetic Toxicitv
No adequate quantitative data were found on which to base quantitative dose-
response estimates for gasoline-induced genetic toxicity.
Reproductive and Developmental Toxicities
The human data are inadequate to assess the quantitative relationship between
gasoline exposure and either reproductive of developmental toxicity.
Abstracts of two Soviet animal studies indicate that subchronic exposure to 300
mg/m3 of gasoline may cause decreased spermatogenesis in male rats, and influence the
rate of uterine myometrical contractions in female rats. There is inadequate exposure
information from which to estimate the dose given to the male rats. The female rats were
exposed for 4 hours/day for 30 to 34 days. This corresponds to the following dose.
LOAEL (Reproductive effects - Gasoline - Female Rats)
300 mg/m3 x 0.13 L/min x 0.001 m3/liter x 60 min/hr x 4 hrs/day/0.25 kg
= 28 mg/kg/dav
Scaling by a factor of 0.2 produces a human equivalent dose of 6 mg/kg/day.
D-17
-------
LOAEL (Reproductive Effects- Gasoline - Female Rats; Human Equivalent Dose)
28 mg/kg/day x 0.2 = 6 mg/kg/day
A developmental effects study on gasoline vapors (Litton Bionetics, 1978b)
observed fetotoxic effects (skeletal abnormalities, low total weight) in rats. Pregnant rats
were exposed to unleaded gasoline vapors at 0,400, or 1600 ppm, for 6 hrs/day, on days
6 through 15 of gestation. The effects were observed at 1600 ppm, but not at 400 ppm.
The LOAEL from this study can be derived as follows.
LOAEL (Fetotoxic Effects - Gasoline vapors - Female Rats)
1200 mg/m3 x 0.13 L/min x 0.001 m3/Liter x 60 min/hr x 6 hrs/day/0.25 kg
= 225 mg/kg/day
Scaling by a factor of 0.2 produces a human equivalent dose of 45 mg/kg/day.
LOAEL (Fetotoxic Effects - Gasoline vapors - Female Rats; Human Equivalent Dose)
225 mg/kg/day x 0.2 = 45 mg/kg/day
BENZENE
Lethality
A human fatality was recorded following a 5 to 10 minute benzene exposure of
66,000 mg/m3 (IARC, 1982). An estimate of this lethal dose is presented below.
LOAEL (Lethality - Benzene -Human)
66,000 mg/m3 x 0.8 m3/hr x 0.08 hr/70 kg = 60 mg/kg/5 min
= 12 mg/kg/min
An acute inhalation LC$Q in rats was determined from 13,700 ppm following a 4
hour exposure. The lethal doses can be estimated as described below (assuming 1 ppm
equals 3 mg/m3).
D-18
-------
LC50 (Lethality. Benzene - Rats)
28,500 mg/m3 x 0.1 L/min x 0.001 m3/Liter x 240 min/0.25 kg = 2700 mg/kg
= 11 mg/kg/min
Applying a scaling factor to this dose (0.2) results in human equivalent doses of
approximately 540 mg/kg, or 2 mg/kg/min
LC50 (Lethality - Benzene - Rats; Human Equivalent Dose)
11 mg/kg/min x 0.2 = 2 mg/kg/min
Pulmonary Toxicitv
Acute pulmonary toxicity (inflammation of the respiratory tract, lung hemorrhage)
has been observed in human autopsies following benzene poisoning. No other information
concerning the pulmonary effects of acute, subacute, subchronic, or chronic benzene
exposure was found in the literature.
Kidnev Toxicitv
Acute kidney toxicity (kidney congestion) has been observed in human autopsies
following benzene poisoning. No other information concerning the nephrotoxic effects of
acute, subacute, subchronic, or chronic benzene exposure was found in the literature.
Neurotoxicitv
Effects of Acute Exposure
No adequate information exists to assess the quantitative relationship of acute
benzene exposure and central nervous system toxicity at non-lethal concentrations. Limited
information exists regarding the irritant properties of benzene. A dose of 2 milligrams was
found to be severely irritating to the eyes of rabbits. Nielson and Alaric (1982) found that a
benzene exposure of 8,500 ppm for 30 minutes produced no significant upper or lower
respiratory tract irritation. This corresponds to a NOAEL of 612 mg/kg.
NOAEL (Respiratory Tract Irritation - Benzene - Mice)
22,500 mg/m3 x 0.02 L/min x 0.001 m3/L x 30 min/0.025 kg = 612 mg/kg/30min
= 20 mg/kg/min
D-19
-------
Effects of Subacute and Subchronic Exposures
No human data were found that are sufficient to describe the quantitative
relationship between subacute or subchronic benzene exposure and neurotoxicity. Animal
data describe relationships between benzene exposure and either neurobehavioral indices or
changes in brain catecholamine levels.
Conditional reflex activity was reduced in rats exposed to 20 ppm, but not 4 ppm,
of benzene, 6 hrs/day, 5 days/week, for 5.5 months. The LOAEL and NOAEL for this
effect are calculated as follows.
LOAEL (Reflex Activity - Benzene - Rats)
60 mg/m3 x 0.1 L/min x 0.002 m^/L x 60 min/hr x 6 hrs/day x 5 days/7 days/0.25 kg
= 6 mg/kg/day
NOAEL (Reflex Activity - Benzene - Rats)
12 mg/m3 x 0.1 L/min x 0.002 rr?IL x 60 min/hr x 6 hrs/day x 5 days/7 days/0.25 kg
= 1.2 mg/kg/day
Because this effect is systemic, a scaling factor of 0.2 is needed to estimate a human
equivalent dose.
LOAEL (Reflex Activity - Benzene - Rats; Human Equivalent Dose)
6 mg/kg/day x 0.2 =1.2 mg/kg/day
NOAEL (Reflex Activity - Benzene - Rats; Human Equivalent Dose)
1.2 mg/kg/day x 0.2 = 0.2 mg/kg/day
Cumulative wheel-turning behavior in mice was decreased following 10 and 100
ppm benzene exposures for 8 hours/day, 5 days/week, for up to 20 days. The LOAEL for
this experiment is calculated as follows.
D-20
-------
LOAEL (Wheel Turning - Benzene - Mice)
30 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7 days/0.025 kg
= 6 mg/kg/day
As this effect is also systemic, a scaling factor of 0.1 is needed to estimate the human
equivalent dose.
LOAEL (Wheel-Turning - Benzene - Mice; Human Equivalent Dose)
6 mg/kg/day x 0.1 = 0.6 mg/kg/day
Changes in rat brain catecholamine levels were observed following a 6 hr/day, 3
day exposure to 1500 ppm of benzene. No lower concentrations were tested. This
exposure corresponds to the following LOAEL in rats.
LOAEL (Changes in catecholamine levels - Benzene - Rats)
4500 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hr/day/0.25 kg = 648 me/kg/dav
Applying a scaling factor of 0.2 results in a human equivalent dose of 130 mg/kg/day.
LOAEL (Changes in catecholamine levels - Benzene - Rats; Human Equivalent Dose)
648 mg/kg/day x 0.2 = 130 mg/kg/day
By contrast, subacute oral daily benzene doses of 8 mg/kg/day produced changes in
mice brain caiecholamine levels. This latter effect level is more consistent with the
observed behavioral changes, and indicates that brain catecholamine levels induced by
benzene inhalation could occur at much lower levels than 1500 ppm. The effect level is
described below.
LOAEL (Changes in catecholamine levels - Benzene - mice)
8 mg/kg/dav
D-21
-------
Applying a scaling factor of 0.1 to this level results in the following human equivalent
dose.
LOAEL (Changes in catecholamine levels - Benzene - Mice; Human Equivalent Dose)
8 mg/kg/day x 0.1 = 0.8 mg/kg/day
Effects of Chronic Exposure
No data were found that are adequate to evaluate the quantitative dose-response
relationship of benzene neurotoxicity.
Hemafotoxicitv
Effects of Acute Exposure
No data on humans exists which evaluates the quantitative relationship between
acute benzene exposure and hematotoxicity. The lowest acute effect level which produced
hematotoxicity was 95 ppm of benzene in mice for 8 hours (the lowest level tested) (Toft et
al., 1982). This exposure substantially reduced total cellularity and granulopoietic colony
forming units in mouse bone marrow cells. This exposure corresponds to the following
LOAEL in mice.
LOAEL (Hematotoxicity - Benzene - Mice)
285 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.025 mg
= 82 mg/kg/6 hours
= 14mg/kg/hr
To derive a human equivalent dose from this study, this animal dose level is scaled by a
factor of 0.1.
LOAEL (Hematotoxicity - Benzene - Mice; Human Equivalent Dose)
14 mg/kg/day x 0.2 = 3 mg/kg/hr
D-22
-------
Effects of Subacute and Subchronic Exposures
No quantitative data on humans exists from which quantitative relationships
between subacute benzene exposures and hematotoxicity can be estimated. Subchronic and
chronic benzene exposure effects on humans are discussed in the chronic exposure effects
section (see below). A considerable amount of subacute and subchronic exposure effects
data exist on the hematotoxicity of benzene on laboratory animals. Summaries of these
studies are presented below.
Several animal studies have been conducted to study benzene effects at
concentrations between 1 and 100 ppm. A summary of studies which investigated the
effects of benzene at concentrations at or below 50 ppm is presented in Table D-l. Acute
and subacute studies indicated that no decreases in leucocyte levels or colony forming units
occurred when benzene concentrations were below 10 ppm. Lowest observed effect levels
for these cell depressions were between 15 and 20 ppm (Cronkite et al., 1985; Rosen et al.,
1984; Toft et al., 1982). Rosen et al. (1984) found significant depressions in lymphocyte
levels in mice exposed to approximately 10 ppm of benzene.
Two other effects identified in the animal studies were without a no observed acute
effect level. Elevations in micronuclei were observed in mice exposed continuously to 14
ppm of benzene for 1 to 8 weeks (Toft et al., 1982). Also, spleen weight and splenic
granulocytes were depressed in mice after a five day exposure to 1 ppm of benzene (Green
et al., 1981). These reduction in splenic granulocytes (50 percent) and spleen weight (85
percent of controls) were statistically significant. This finding is particularly important
given the observations of granulocytopenia in benzene-exposed workers. At 10 ppm, the
reduction in spleen weight was not statistically significant. It was, however, consistent
with a decreasing trend in spleen weight in response to increasing benzene concentrations.
Subchronic effects studies have identified lowest observed effect levels for benzene
of 15 ppm (spleen effects, specifically increased levels of hemosiderin pigments)
(Deichmann et al., 1963) and 10 ppm (depressed erythroid and burst colony forming units
in the bone marrow) (Baarson et al., 1984). The first finding is consistent with
observations of red blood cell anomalies in occupational cohorts. The second finding is
important given the biological importance of stem cells, the findings of the acute and
subacute animal studies, and the degree of depression observed (95 percent depression in
erythroid colony-forming units, and 40 percent depression in the burst colony-forming
units throughout the first half of exposure). These findings are clinically important.
Depressions of of stem cell levels to about 10 percent of normal values result in the
cessation of stem cell differentiation, as cell division within this compartment only
produces more stem cells (Gill et al., 1980).
D-23
-------
TABLE D-l
SUMMARY OF STUDIES THAT EXAMINED THE EFFECTS OF
BENZENE AT CONCENTRATIONS UP TO 50 PPM
Study: Baarson et al., 1984
Species: Male C57B1 mice
Benzene Concentration: lOppm
Exposure: 6 hrs/ day, 5 days/week, 178 days
Effects:
* Depression of CFU-E in bone marrow at 32 days (66 percent of controls) and
178 days (5 percent of controls). Statistically significant at both times.
* Depression of BFU-E in bone marrow at 32 days (66 percent of controls) and 66
days (66 percent of controls). Statistically significant at 66 days. BFU-E returned
to control values at 178 days.
* Depression or peripheral RBC at 66 days (80 percent of controls) and 178 days
(80 percent of controls). Statistically significant at both times.
* Depression of peripheral lymphocytes at 32,66, and 178 days (all at 70 percent
of controls). Statistically significant at all times.
* Depression of CFU-E in spleen at 32 days (70 percent of controls) and 178 days
(10 percent of controls). Statistically significant at 178 days.
* Elevation of BFU-E in spleen at 32 days (400 percent of controls), 66 days (150
percent of controls), and 178 days (260 percent of controls). None were
statistically significant.
* Depression of cellularity in spleen at 178 days (85 percent of controls).
Statistically significant.
* Elevation of nucleated red cells in spleen at 32 days (225 percent of controls) and
66 days (220 percent of controls). Not statistically significant. Significant
depression at 178 days (15 percent of controls).
D-24
-------
TABLE D-l
(continued)
Study: Cronkite et al., 1985
Species: Male and female C57B1 mice
Benzene Concentrations: 10 and 25 ppm
Exposure: 6 hrs/day, 5 days/week, 2 weeks
Effects:
* Depression of peripheral lymphocytes (75 percent of controls) at 25 ppm. No
effect at 10 ppm.
*No effects on peripheral granulocytes, bone marrow cellularity, CPU content in
bone marrow, or the fraction of stem cells in DNA synthesis at either 10 or 25 ppm.
Study: Rosen et al., 1984
Species: Male C57B1 mice
Benzene Concentrations: 10 and 31 ppm
Exposure: 6 hrs/day; 6 days
Effects:
* Depression of peripheral lymphocyte levels at 10 and 31 ppm (both about 60
percent of controls). Statistically significant at both concentrations.
* More severe depression of the above-mentioned parameters at higher
concentrations (100 and 300 ppm).
* Slight depression of peripheral RBC at 10 ppm at 31 ppm (both at about 90
percent of controls). Neither was statistically significant. Significant depressions
at higher concentrations, however.
* Depression of mitogen-induced B-Iymphocyte in bone marrow CFC at 10 and 31
ppm (both at about 30 percent of controls). Statistically significant at both
concentrations. Similar depressions observed at higher exposure concentrations.
* Slight depressions in B-lymphocyte numbers in bone marrow at 10 ppm (80
percent of controls) and 31 ppm (60 percent of controls). Neither was statistically
significant. Significant depressions at higher concentrations, however.
D-25
-------
TABLE D-l
(continued)
Study: Aoyama, 1986
Species: Male BALB/c mice
Benzene Concentration: 50 ppm
Exposure: 6 hrs/day, 7 or 14 consecutive days
Effects:
* Depressions in spleen weight/body weight at 7 days (85 percent of controls) and
14 days (68 percent of controls). Statistically significant. Greater depression at
200 ppm.
* Depression in thymus weight/body weight at 14 days (50 percent of controls).
Statistically significant. No effect at 7 days. Greater depression at 200 ppm.
* Depression of white blood cell counts at 14 days (54 percent of controls).
Statistically significant. No effect at 7 days. Greater depression at 200 ppm.
* Depression of total peripheral lymphocytes at 14 days (50 percent of controls).
Statistically significant No effect at 7 days.
* Depression of peripheral blood and spleen B-lymphocytes (as measured by
percentage of surface immunoglobulin positive cells) at 7 days (80 percent of
controls) and 14 days (62 percent of controls). Statistically significant at both
times. Absolute numbers depressed at 7 days (85 percent of controls) and 14 days
(16 percent of controls). Statistically significant at 14 days.
* Depression of peripheral blood and spleen T-lymphocytes (as measured by
percentage of Thy 1.2 positive cells) at 7 days (64 percent of controls) only.
Statistically significant. Depressions in absolute number of T-lymphocytes at 7
days (66 percent of controls) and 14 days (60 percent of controls). Statistically
significant at both times.
D-26
-------
TABLE D-l
(continued)
* Depression of IgM production (13 percent of controls) at 14 days of exposure
(immunization taking place on day 7) of IgG production (16 percent and 30 percent
of controls) at 14 days of exposure (immunization taking place on day 7 and day
14, respectively).
* No effect on cell mediated immunity.
Study: Toft et ah, 1982
Species: Male NMRI mice
Benzene Concentration: 10,14,21, and SO ppm
Exposure: Continuous and 8 hrs/day, 5 days/week, both for up to two weeks
Effects:
* Depression in cells/tibia (about 15 percent of controls) and CFU-C/tibia (about 40
percent of controls) at 96 hours continuous exposure to 21 ppm. Statistically
significant. Tibia cellularity increased slightly (to about 50 percent of controls) by
end of exposure. CFU-C/tibia continued to decline throughout exposure. No
effects seen at 10 ppm.
* Increase in micronuclei/500 PCE (10-fold) in mice exposed to 21 ppm for 96
hours. Increase in micronuclei (4-fold) in mice exposed to 14 ppm for 96 hours.
Statistically significant at both times. Increased exposure duration did not increase
response.
* Depressions in CFU-C/tibia (30 percent of controls) in mice exposed
intermittently to 21 ppm. No effect at 14 ppm. Cellularity depressed at 50 ppm (33
percent of controls).
* Increases in micronuclei/500 PCE (5-fold) in mice exposed intermittently to 21
ppm. No effect at 14 ppm.
D-27
-------
TABLE D-l
(continued)
Study: Green etal., 1981
Species: Male CD-I mice
Benzene Concentrations: 1 and 10 ppm
Exposure: 6 hrs/day, 5 days/week, 1 and 10 weeks
Effects:
* Depression in spleen weight (85 percent of controls) and splenic granulocyte
counts (SO percent of controls) after 1 week exposure to 1 ppm. Statistically
significant. Depression, not significant at 10 ppm, but exposure to higher doses
resulted in significant depressions that were dose-related. No decrease in femur
cellularity, femoral granulocytes, femoral lymphocytes, femoral nucleated red blood
cells, peripheral white or red bloods or lymphocytes.
* Increase in spleen weight (125 percent of controls), spleen cellularity (125 percent
of controls), and nucleated red blood cells in spleen (180 percent of controls) at 50
day exposure to 10 ppm. Statistically significant. Elevations, not significant, in
splenic lymphocytes (120 percent of controls) and splenic granulocytes (200
percent of controls), and in peripheral and bone marrow cell counts. Stem cells
were not elevated.
Study: Ward et ah, 1985
Species: Sprague
Benzene Concentrations: 1, 10 and 30 ppm
Exposure: 6 hrs/day, 5 days/week, 13 weeks
Effects:
* No changes in peripheral cell counts.
Study: Johnston et al., 1979
Species: Duroc-Jersey Pigs
Benzene Concentrations: 20 ppm
Exposure: 6 hrs/day, 5 days/week, for 3 weeks
Effects:
* No effects peripheral or bone marrow cells observed 72 hours after last exposure.
D-28
-------
The available data show that changes in peripheral blood cell counts and spleen
weights are reversible, although the tendency appears to be greater among lymphocytes
than among red blood cells or platelets (Goldstein, 1977). Changes in stem cell levels,
however, may not be completely reversible (Goldstein, 1977; Cronkite et al., 1985).
Particularly, effects on defective stem cell levels may not be completely reversible
(Goldstein, 1977), a finding that could have specific importance with regard to leukemia
and other irreversible blood disorders.
The effects on the thymus, though apparently reversible, could be critical to the
developing infant Maturation of T-lymphocytes takes place in the thymus early in life.
Thus, benzene effects on infants may be substantially different from those predicted from
studies on adult humans and rodents.
The mechanisms by which benzene induces toxicity are not fully understood. The
effects are believed to be mediated through one or more metabolites of this chemical.
Attention has focused on the chemical's clastogenic effects as well as its ability to inhibit
microtubule formation. This latter effect would result in arrested mitoses (through
inhibition of the spindle formation) and possibly other effects on the cytoskeleton.
Benzene may also specifically interfere with heme biosynthesis during the early stages of
erythrocyte maturation (Cohen et al., 1978). Direct interactions between polyhydroxy
benzene metabolites and specific cell membrane components have been identified.
While there is some variability in the hematoxic responses among these studies,
most subacute and subchronic studies identified 10 ppm exposures as the lowest observed
adverse effect levels for most cytopenic effects and subtle changes in the mouse spleen
were observed after 5 days of exposure to 1 ppm of benzene. The lowest adverse effect
level doses for these effects are calculated as follows.
LOAEL (Cytopenia - Benzene - Mice)
30 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.025 kg = 8.6 me/kg/dav
LOAEL (Spleen changes - Benzene - Mice)
3 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.025 kg = 0.9 mg/kp/dav
Scaling these doses by a factor of 0.1 results in human equivalent doses of 0.8 mg/kg/day
for cytopenic effects and 0.09 mg/kg/day for spleen effects.
D-29
-------
LOAEL (Cytopenia - Benzene - Mice; Human Equivalent Dose)
8.6 mg/kg/day x 0.1 = 0.9 mg/kg/day
LOAEL (Spleen changes - Benzene - Mice; Human Equivalent Dose)
0.9 mg/kg/day x 0.1 = 0.09 mg/kg/day
Cytopenia was observed in mice following a subchronic exposure to 8 to 180
mg/kg/day of benzene via drinking water (Hseih et al., 1988). This exposure corresponds
to the following LOAEL.
LOAEL (Cytopenia - Benzene - Mice)
8 mg/kg/day
Scaling this value by a factor of 0.1 results in the following human equivalent dose.
LOAEL (Cytopenia - Benzene - Mice; Human Equivalent Dose)
8 mg/kg/day x 0.1 = 0.8 mg/kg/day
Effects of Chronic Exposure
Occupational studies and laboratory investigations have shown that exposure to
benzene causes changes in the number, structure, and function of blood cells and blood-
forming organs. Pancytopenia and leukopenia are common effects of benzene exposure.
In animals, a decrease in stem cells and lymphocyte levels are common indicators of
benzene toxicity. Decreases in red blood cell levels have also been reported, although to a
lesser degree. In humans, decreases in red blood cells, granulocytes, and platelets are
usually observed as a consequence of benzene exposure. Lymphopenia has also been
frequently reported, although in some cases, lymphocytosis has been observed. On the
other hand, lymphopenia has been cited as a possible early marker of benzene toxicity in
humans (Moszczynski and Lisiewicz, 1984). Changes in spleen weight are often reported
in animal studies. The effect of benzene on spleen in humans is variable.
Abnormal cellular changes have often been observed in benzene-exposed workers.
Red blood cells exhibited abnormal size and shape, chromosomal anomalies, and increased
D-30
-------
osmotic fragility. Anemias identified in exposed workers were often associated with
hyperplastic or hypoplastic bone marrows. Chromosomal anomalies and morphological
alternations in lymphocytes of benzene-exposed workers have been found (Goldstein,
1977). Abnormal platelets have also been observed in workers with thrombocytopenia.
Benzene exposure may also affect the immune competence of blood cells as well as
the number of cells. Phagocytosis was depressed in one study of benzene-exposed
workers (Doskin, 1971). Animal studies have shown that benzene may inhibit
blastogenesis and antibody formation in B-lymphocytes, and alter membrane receptors on
B- and T-lymphocytes (Aoyama, 1986; Rosen et al. 1984). These effects may occur in the
absence of effects on lymphocyte blood levels (Aoyama, 1986). Alternations in T-
lymphocyte may specifically involve damage to suppressor cells, which play a role in cell
mediated immunity and, possibly, in the development of aplastic anemia (Goldstein, 1977).
Exposure data regarding the epidemiological studies are very limited. Studies on
shoeworkers in Turkey found benzene levels ranging from 30 to 200 ppm, with some
excursions as high as 640 ppm. Data on exposure to other chemicals were often not
available. No conclusions regarding dose-response relationships between benzene
concentration or exposure durations can be made from these studies.
Assuming that 30 ppm represents the lowest observed chronic effect level for
benzene-induced hematotoxicity, the dose associated with this exposure is as follows.
LOAEL (Leucopenia - Benzene - Humans)
90 mg/m3 x 0.8 m3/hr x 0.67* x 8 hrs/day x 5 days/7 days/70 kg = 6 mg/kg/dav
Genetic Toxicity
Effects of Acute Exposure
There is considerable human and animal evidence which shows that benzene causes
cytogenetic damage to chromosomes (clastogenicity) and at the genomic level (inhibition of
mitosis). The lowest benzene concentrations associated with adverse responses were
found in a study by Erexson et al., (1986)-. Significant increases in micronuclei formation
were observed in the bone marrow polychromatic erythrocytes of rats who were exposed to
1 ppm of benzene for 6 hours (Erexson et al., 1986). This same exposure regimen
protocol produced borderline increases in sister chromatid exchange frequency in rat
peripheral blood leucocytes (Erexson et al., 1986). Both of these parameters are measures
D-31
-------
of chromosome damage. Significant decreases in mitotic activity and significant increases
in sister chromatid exchange frequency were observed in rat peripheral blood leucocytes
following a 6-hour exposure to 3 ppm (Erexson et ah, 1986). No significant cytogenetic
changes were observed in rats exposed to either 0.1 or 0.3 ppm of benzene for 6 hours
(Erexson et al., 1986). Mice exposed to 10 ppm of benzene for 6 hours had significantly
adverse changes in all three of these cytogenetic parameters (i.e., micronuclei formation,
sister chromatid exchange frequency, decreased mitotic activity) (Erexson et al., 1986); no
lower concentrations were tested in this species. Thus, the rat responses are the most
sensitive measures of benzene-induced genetic toxicity. The LOAELs and NOAELs for
these effects are calculated below.
LOAEL Onhibition of Mitosis - Benzene - Rats)
9 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.25 kg = 1.2 mg/kg/dav
LOAEL (Clastogenetic effects - Benzene - Rats)
NOAEL (Inhibition of Mitosis - Benzene - Rats)
3 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.25 kg = 0.4 me/ke/dav
NOAEL (Clastogenic effects - Benzene - Rats)
0.9 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.25 kg = 0.1 mg/kg/dav
The human equivalent doses for these effect levels are derived by multiplying these animal
doses by 0.2.
LOAEL (Inhibition of Mitosis - Benzene - Rats; Human Equivalent Dose)
1.2 mg/kg/day x 0.2 = 0.2 mg/kg/day
LOAEL (Clastogenetic effects - Benzene - Rats; Human Equivalent Dose)
NOAEL (Inhibition of Mitosis - Benzene - Rats; Human Equivalent Dose)
0.4 mg/kg/day x 0.2 = 0.08 mg/kg/day
D-32
-------
NOAEL (Clastogenic effects - Benzene - Rats; Human Equivalent Dose)
0.1 mg/kg/day x 0.2 = 0.02 mg/kg/day
Effects of Subacute and Subchronic Exposures
Toth et al. (1982) observed that mice exposed continuously to 14 ppm of benzene
for 1 to 8 weeks had a significantly elevated frequency of micronuclei in polychromatic
erythrocytes (PCEs). No significantly elevated frequency in micronuclei was observed
following a 14 ppm benzene exposure, 8 hours/day, 5 days/week, for 2 weeks. These
exposure corresponds to a LOAEL of 48 mg/kg/day and a NOAEL of 16 mg/kg/day.
LOAEL (Micronuclei - Benzene - Mice)
42 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 24 hrs/day/0.025 kg = 48 mg/kg/dav
NOAEL (Micronuclei - Benzene - Mice)
42 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 8 hrs/day/0.025 kg = 16 mg/kg/dav
To derive a human equivalent dose, these levels are multiplied by a factor of 0.1.
LOAEL (Micronuclei - Benzene - Mice; Human Equivalent Dose)
48 mg/kg/day x 0.1 = 4.2 mg/kg/day
NOAEL (Micronuclei - Benzene - Mice; Human Equivalent Dose)
16 mg/kg/day x 0.1 = 1.6 mg/kg/day
Reproductive and Developmental Toxieities
Reproductive Toxicity
One reproductive toxicity study on benzene was identified in the literature (Ward et
al., 1985). Male and female mice were exposed to 300 ppm of benzene, 6 hours/day, 5
days/week, for 13 weeks. The males exhibited testicular atrophy and degeneration,
decreased numbers of epididymal sperm, and abnormal sperm morphology. Females
D-33
-------
developed ovarian cysts as a result of the exposure. The LOAEL for these effects is
determined as follows.
LOAEL (Reproductive Toxicity - Benzene - Mice)
900 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7days/0.025 kg
= 185 mg/kg/day
As this is a systemic effect, the human equivalent dose is derived by multiplying
this level by 0.1.
LOAEL (Reproductive Toxicity - Benzene - Mice; Human Equivalent Dose)
185 mg/kg/day x 0.1 = 18.5 mg/kg/day
Developmental Toxicitv
Persistent changes in motor activity and catecholaminergic function were observed
in rats exposed neonatally to 550 mg/kg of benzene.
Various measures of developmental toxicity associated with benzene exposure have
been identified. Those effects with data adequate for quantitative dose estimation include
low fetal birth weight and specific persistent alterations include in the hematopoietic and
nervous systems. Low fetal weight is itself a reversible effect, but may be an indicator of
other, more subtle and persistent effects.
Kuna and Ulrich (1981) exposed pregnant Sprague-Dawley rats to 10,50, and 500
ppm of benzene, 7 hours/day, from days 6 through 15 of gestation. Significantly
decreased weight gain was observed in the offspring of the 50 and 500 ppm exposed
animals. A LOAEL and NOAEL for this effect is calculated as follows.
LOAEL (Low Fetal Weight - Benzene - Rats)
150 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 7 hrs/day /0.25 kg = 25 mg/kg/dav
NOAEL (Low Fetal Weight - Benzene - Rats)
30 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 7 hrs/day /0.25 kg = 5 mg/kg/dav
D-34
-------
The human equivalent dose is derived by multiplying these dose levels by 0.2.
LOAEL (Low Fetal Weight - Benzene - Rats; Human Equivalent Dose)
25 mg/kg/day x 0.2 = 5 mg/kg/day
NOAEL (Low Fetal Weight - Benzene - Rats; Human Equivalent Dose)
5 mg/kg/day x 0.2 = 1 mg/kg/day
Keller et al. (1986) exposed pregnant Swiss-Webster mice to 0, 5,10, or 20 ppm
of benzene, 6 hours/day on days 6 through 15 of gestation. They found that fetuses of
mice (particularly the male fetuses) exposed to 5 ppm of benzene exhibited significant
changes in their hematopoietic stem cell levels (BFU-E, CFU-E, and GM-CFU-E). These
changes did not persist past parturition. In the male progeny of the 10 ppm exposed mice,
however, persistent changes in hematopoietic stem cells were observed following in utero
exposures. Moreover, re-exposure of these mice to 10 ppm of benzene at 10 weeks of age
resulted in significantly depressed stem cell responses relative to the mice exposed in utero
to air. Thus, the 5 ppm exposure regimen represents a LOAEL for fetotoxicity, while the
10 ppm exposure represents a LOAEL, and 5 ppm represents the NOAEL, for persistent
developmental changes.
LOAEL (Fetotoxic Hematological Effects - Benzene - Mice)
NOAEL(Developmental Hematological Effects - Benzene - Mice)
15 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day /0.025 kg = 4 mg/kg/dav
LOAEL (Developmental Hematological Effects - Benzene - Mice)
30 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day /0.025 kg = 8 mg/kg/dav
To derive a human equivalent dose, a scaling factor of 0.1 is employed.
D-35
-------
LOAEL (Fetotoxic Hematological Effects - Benzene - Mice; Human Equivalent Dose)
NOAEL(Developmental Hematological Effects - Benzene - Mice; Human Equivalent Dose)
4 mg/kg/day x 0.1 = 0.4 mg/kg/day
LOAEL (Developmental Hematological Effects - Benzene - Mice; Human Equivalent Dose)
8 mg/kg/day x 0.1 = 0.8 mg/kg/day
TOLUENE
Lethality
It is difficult to estimate an acute lethal toluene dose to humans from the available
literature. Exposure to 10,000 to 30,000 ppm of toluene for a few minutes produced
narcosis in human beings. Because doses sufficient to produce narcosis are often close to
lethal doses, an approximate lethal dose can be estimated from this information. It is
assumed for the purposes of this estimation that an adult is exposed to 10,000 ppm of
toluene for 15 minutes, and that 1 ppm of toluene equals approximately 3.75 mg/m3.
LOAEL (Lethality - Toluene - Humans)
37,500 mg/m3 x 0.8 m3/hr x 0.25 hour/70 kg = 107 mg/kg/15 min
= 7 mg/kg/minute
Rodent studies have determined an oral LDso ranges between 6 and 7.5 grams/kg.
One inhalation study on rats noted mortality in 1 of 6 rats exposed to 4,000 ppm of
toluene for 4 hours. This exposure corresponds to a LOAEL of 360 mg/kg/hr
(6 mg/kg/min).
LOAEL (Lethality - Toluene -Rats)
15,000 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 4 hrs/0.25 kg = 1440mg/kg/4 hrs
= 360 mg/kg/hr
= 6 mg/kg/min
The human equivalent dose is derived by multiplying this dose by 0.2.
D-36
-------
LOAEL (Lethality - Toluene - Rats; Human Equivalent Dose)
1440 mg/kg/4 hrs x 0.2 = 290 mg/kg
360 mg/kg/hr x 0.2 = 72 mg/kg/hr
6 mg/kg/min x 0.2 =1 mg/kg/min
Pulmonary Toxicitv
Effects of Acute Exposure
Information is inadequate to assess the quantitative relationship between acute
toluene exposure and pulmonary toxicity.
Subacute and Subchronic Exposures
Only one animal study exists which is adequate to assess the quantitative
relationship between subacute or subchronic toluene exposure and pulmonary toxicity.
Very limited information exists regarding the quantitative relationship between toluene
exposure and pulmonary toxicity. Aranyi et al. (1985) reported that CD-I mice exhibited
an increased mortality from respiratory infections following a 3-hour exposure to 1 ppm of
toluene, with no effect at 1 ppm. The LOAEL and NOAEL for this effect is derived below.
LOAEL (Respiratory Infections - Toluene - Mice)
9.4 mg/m3 x 0.02 L/min x 0.001 rn^/L x 60 min/hr x 3 hrs/day /0.025 kg
= 1.4mg/kg/dav
NOAEL (Respiratory Infections - Toluene - Mice)
3.75 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 3 hrs/day /0.025 kg
= 0.5 mg/kg/day
Effects of Chronic Exposure
One animal study provides information from which to assess the quantitative
relationship between chronic toluene exposure and pulmonary toxicity. No treatment-
related effects were observed in Fischer 344 rats exposed to 300 ppm of toluene for 6
D-37
-------
hours/day, 5 days/week, for 2 years. This exposure corresponds to a NOAEL of 116
mg/kg/day.
NOAEL (Pulmonary Toxicity - Toluene - Rats)
1125 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5days/7days AJ.25 kg
= 116 mg/kg/day
Kidnev and Liver Toxicitv
Effects of Acute Exposure
Information is inadequate to assess the quantitative relationship between acute
toluene exposure and kidney or liver toxicity.
Effects of Subacute and Subchronic Exposures
Ungvary et al. (1982) observed increased liver weight to body weight ratios in rats
and mice exposed to 7,500 mg/m3 of toluene, 8 hours/day, for 1 to 3 weeks. The
LOAELs for these exposures are calculated as follows.
LOAEL (Hepatotoxicity - Toluene - Rats)
7500 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 8 hrs/day /0.25 kg
= 1440 mg/kg/day
LOAEL (Hepatotoxicity - Toluene - Mice)
7500 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 8 hrs/day /0.025 kg
= 2880 me/ke/dav
The human equivalent doses for these levels are derived by applying a scaling factor of 0.2
to the rat dose and 0.1 to the mouse dose.
LOAEL (Hepatotoxicity - Toluene - Rats; Human Equivalent Dose)
1440 mg/kg/day x 0.2 = 290 mg/kg/day
D-38
-------
LOAEL (Hepatotoxicity - Toluene • Mice; Human Equivalent Dose)
2880 mg/kg/day x 0.1 = 290 mg/kg/day
Effects of Chronic Exposure
One animal study provides information from which to assess the quantitative
relationship between chronic toluene exposure and kidney or liver toxicity. No treatment-
related effects were observed in Fischer 344 rats exposed to 300 ppm of toluene for 6
hours/day, 5 days/week, for 2 years. This exposure corresponds to NOAEL of 116
mg/kg/day.
NOAEL (Kidney/Liver Toxicity - Toluene - Rats)
1125 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5days/7days /0.25 kg
= 116mg/kg/dav
The human equivalent dose is derived by multiplying this dose level by 0.2.
NOAEL (Kidney/Liver Toxicity - Toluene - Rats; Human Equivalent Dose)
116 mg/kg/day x 0.2 = 23 mg/kg/day
Neurotoxicitv
Effects of Acute Exposure
Human subjects exposed to 40 ppm (Andersen et al., 1983) and 50 ppm (von
Oettingen et al., 1942) during single, short-term durations reported no complaints of
sensory irritation. Andersen et al. (1983) exposed adult human subjects to 40 ppm for 6
hours. The subjects showed indications of central nervous system toxicity following
exposure to 375 mg/m3 of toluene. These findings correspond to a lowest effect level of
39 mg/kg (6.5 mg/kg/hr) and a no observed effect level of 15 mg/kg (2.5 mg/kg/hr).
LOAEL (Sensory Irritation - Toluene -Humans)
375 mg/m3 x 1.2 L/min x 6 hr/day / 70 kg = 39 me/kg/dav
= 6.5 mg/kg/hour
D-39
-------
NOAEL (Sensory Irritation - Toluene -Humans)
150 mg/m3 x 1.2 L/min x 6 hr/day / 70 kg = 16mg/kg/dav
= 2.5 mg/kg/hour
Neurotoxic effects of acute toluene exposure in laboratory animals have not generally been
observed except at lethal doses.
Taylor and Evans (1985) observed toluene-induced changes in neurological
endpoints following 20 to 50 minute exposures to 100 to 4500 ppm of toluene. Central
nervous system decrements involving attention and visual motor abilities were observed
following 25 to 50 minute exposures to 2,000 ppm or greater. Increases in carbon dioxide
expiration (a potentially sensitive indicator of combined behavioral, sensory, respiratory,
and metabolic effects) were observed following a 20-minute exposure to 100 ppm of
toluene (the shortest duration and the lowest concentration for which measurements were
taken). These exposures correspond to the following LOAELs and NOAELs.
LOAEL (Attention, Visual-motor abilities - Toluene - monkeys)
7500 mg/m3 x 0.8 L/min x 0.001 m3/L x 25 min/5 kg = 30 me/kg
= 1.2 mg/kg/min
NOAEL (Attention, Visual-motor abilities - Toluene - monkeys)
3,750 x mg/m3 x 0.8 L/min x 0.001 m3/L x 25 min/5 kg = 15mg/kg
= 0.6 mg/kg/min
LOAEL (CO2 expiration - Toluene - monkeys)
375 mg/m3 x 0.8 L/min x 0.001 m3/L x 20 min/5 kg =1.2 me/kg
= 0.06 mg/kg/min
The human equivalent dose for the effect levels is determined by multiplying the
dose by 0.5.
D-40
-------
LOAEL (Attention, Visual-motor abilities - Toluene - monkeys; Human Equivalent Dose)
30 mg/kg x 0.5 = 15 mg/kg
1.2 mg/kg/min x 0.5 = 0.6 mg/kg/min
NOAEL (Attention, Visual-motor abilities - Toluene - monkeys; Human Equivalent Dose)
15 mg/kg x 0.5 = 7.5 mg/kg
0.6 mg/kg/min x 0.5 = 0.3 mg/kg/min
LOAEL (CO2 expiration - Toluene - monkeys; Human Equivalent Dose)
1.2 mg/kg x 0.5 = 0.6 mg/kg
0.06 mg/kg/min x 0.5 = 0.03 mg/kg/min
Mice exposed to toluene in utero through 30 to 45 days of age exhibited signs of
neurological impairment (Kostas and Hotchin, 1981). Exposure to 72 mg/kg/day (but not
14.4 mg/kg/day) in drinking water resulted in significantly reduced adaptation to open field
behavior, while exposure to levels as low as 3 mg/kg/day significantly reduced rotorod
performance. These effect levels are described below.
LOAEL (Open field behavior effects - Toluene - mice)
75 mg/kg/day
NOAEL (Open field behavior effects - Toluene - mice)
14.4 mg/kg/day
LOAEL (Rotorod performance - Toluene - mice)
3 mg/kg/day
Multiplying these values by a scaling factor of 0.1 results in the following human
equivalent doses.
D-41
-------
LOAEL (Open field behavior effects - Toluene -mice; Human Equivalent Dose)
72 mg/kg/day x 0.1 = 7.2 mg/kg/day
NOAEL (Open field behavior effects - Toluene -mice; Human Equivalent Dose)
14.4 mg/kg/day x 0.1 = 1.4 mg/kg/day
LOAEL (Rotorod performance - Toluene - mice; Human Equivalent Dose)
3 mg/kg/day x 0.1 = 0.3 mg/kg/day
Effects of Subacute and Subchronic Exposures
The available data indicates that the neurotoxic responses of human beings to
subacute and subchronic toluene exposures are similar to the acute exposure responses.
Animal studies have assessed the effects of toluene exposure on behavioral changes and
brain neurotransmitter and receptor binding.
Adult mice exposed to 1,10,100, or 1,000 ppm of toluene, 6 hours/day, for up to
20 days, showed significantly reduced cumulative wheel turning rounds relative to air
exposed controls. The LOAEL of 1 ppm corresponds to the following dose.
LOAEL (Reduced Wheel Turning - Toluene - Mice)
3.75 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day /0.02 kg
= 1 mg/kg/day
The human equivalent dose is derived by multiplying this level by 0.1.
LOAEL (Reduced Wheel Turning - Toluene - Mice; Human Equivalent Dose)
1 mg/kg/day x 0.1 = 0.1 mg/kg/day
Adult rats (600 g) were exposed to 80 ppm of toluene 6 hours/day, 5 days/week,
for 3 months. Exposure-induced changes in receptor binding, protein phosphorylation,
norepinephrine levels in the hypothalamus, and serum prolactin levels were observed. The
LOAEL for these effects are derived below.
D-42
-------
LOAEL (Brain Neurotransmitter, Protein Phosphorylation, Receptor Binding Changes -
Toluene - Rats)
300 mg/m3 x 0.25 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7 days/0.6 kg
= 32 mg/kg/day
The human equivalent dose is derived by multiplying this value by 0.3.
LOAEL (Brain Neurotransmitter, Protein Phosphorylation, Receptor Binding Changes -
Toluene - Rats; Human Equivalent Dose)
32 mg/kg/day x 0.3 = 10 mg/kg/day
Effects of Chronic Exposure
No adequate lexicological study exists form which to evaluate the quantitative
relationship between chronic toluene exposure and neurotoxicity. Human epidemiological
studies are confounded by the fact that workers are often exposed to other neurotoxic
chemicals in the workplace besides toluene. Iregren (1982) found that workers exposed to
toluene at 300 mg/m3 exhibited significant decreases in simple reaction time. Assuming an
8 hour/day, 5 day/week exposure, the following LOAEL is derived.
LOAEL (Reduced Reaction Time - Toluene - Humans)
300 mg/m3 x 1.2 m3/L x 60 min/hr x 8 hrs/day x 5 days/7 days/70 kg
= 29 mg/kg/dav
Hematoxicitv
Effects of Acute Exposure
Information is inadequate to assess the quantitative relationship between acute
toluene exposure and hematotoxicity.
Effects of Subacute and Subchronic Exposures
Only one animal study is available to assess the quantitative relationship between
subacute or subchronic toluene exposure and hematotoxicity. Horiguchi and Inoue (1977)
D-43
-------
exposed male NA2 mice (20 g) to 1,10,100, and 1,000 ppm of toluene for 6 hours/day,
for 20 days. Red blood cell counts and certain bone marrow elements were depressed in
the 10,100, and 1,000 ppm exposure groups. White blood cell counts were transiently
elevated in all exposure groups midway through exposure. They approached control
values by the end of exposure except in the 1,000 ppm exposure group, which dropped to
about 50 percent of control values. The corresponding effect levels for these exposures are
described below.
LOAEL (Red Blood Cell Decreases, Bone Marrow Effects - Toluene - Mice)
375 mg/m3 x 0.02 L/min x 0.001 m*/L x 60 min/hr x 6 hrs/day /0.02 kg
= 110 mg/kg/day
LOAEL (Thrombocyte Decreases - Toluene - Mice)
NOAEL (Red Blood Cell Decreases; Bone Marrow Effects - Toluene - Mice)
37.5 mg/m3 x 0.02 L/min x 0.001 m*/L x 60 min/hr x 6 hrs/day /0.02 kg
= 11 mg/kg/dav
NOAEL (Thrombocyte Decreases - Toluene - Mice)
3.75 mg/m3 x 0.02 L/min x 0.001 rn^/L x 60 min/hr x 6 hrs/day /0.02 kg
= 1 mg/kg/day
The human equivalent doses for these effect levels are derived by multiplying these levels
by 0.1.
LOAEL (Red Blood Cell Decreases, Bone Marrow Effects - Toluene - Mice;
Human Equivalent Dose)
110 mg/kg/day x 0.1 = 11 mg/kg/day
D-44
-------
LOAEL (Thrombocyte Decreases - Toluene - Mice; Human Equivalent Dose)
NOAEL (Red Blood Cell Decreases; Bone Marrow Effects - Toluene - Mice;
Human Equivalent Dose)
11 mg/kg/day x 0.1 = 1.1 mg/kg/day
NOAEL (Thrombocyte Decreases - Toluene - Mice; Human Equivalent Dose)
1 mg/kg/day x 0.1 = 0.1 mg/kg/day
Effects of Chronic Exposure
Two human and one animal study provide information from which to assess the
quantitative relationship between chronic toluene exposure and hematotoxicity. Cohr and
Stockholm (1979) observed changes in leucocyte enzyme activities and lymphocyte
morphology among workers chronically exposed to 160 to 3000 mg/m^ of toluene.
Matsushita et al. (1975) observed a significantly increased number of Mommsen's toxic
granules in the neutrophils of workers chronically exposed to 375 mg/m3 of toluene.
Thus, 160 mg/m3 represents a lowest effect exposure for toluene-induced hematotoxicity in
chronically exposed workers. Assuming the workers were exposed for 8 hours/day, 5
days/week, the following LOAEL can be derived.
LOAEL (Hematoxicity -Toluene - Humans)
160 mg/m3 x 1.2 L/min x 8 hr/day x 5 days/7days/70 kg
= 16 mg/kg/day
No treatment-related effects were observed in male Fischer 344 rats exposed to 300
ppm of toluene for 6 hours/day, 5 days/week, for 2 years. Female rats showed
significantly increased mean corpuscular hemoglobin concentrations at this exposure;
female rats chronically exposed to 100 and 300 ppm of toluene had significantly reduced
hematocrit levels. These exposures correspond to the following dose levels.
D-45
-------
LOAEL (Mean Corpuscular Hemoglobin Increases - Toluene - Female Rats)
NOAEL (Hematotoxicity - Toluene - Male Rats)
1125 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7days/0.25 kg
= 116 mg/kg/dav
LOAEL (Hematocrit changes - Toluene - Female Rats)
NOAEL (Mean Corpuscular Hemoglobin Increases - Toluene - Female Rats)
375 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day x 5 days/7days/0.25 kg
= 39 mg/kg/day
The human equivalent doses for these effects are derived by multiplying the levels by 0.2.
LOAEL (Mean Corpuscular Hemoglobin Increases - Toluene - Female Rats;
Human Equivalent Dose)
NOAEL (Hematotoxicity - Toluene - Male Rats; Human Equivalent Dose)
116 mg/kg/day x 0.2 = 23 mg/kg/day
LOAEL (Hematocrit changes - Toluene - Female Rats; Human Equivalent Dose)
NOAEL (Mean Corpuscular Hemoglobin Increases - Toluene - Female Rats;
Human Equivalent Dose)
39 mg/kg/day x 0.2 = 8 mg/kg/day
Genetic Toxicitv
There is currently inadequate information to assess the genetic toxicity that may
result from toluene exposure.
Carcinopenicifv
There is currently inadequate information to assess the carcinogenicity that may
result from toluene exposure.
D-46
-------
Reproductive and Developmental Toxicities
Reproductive Toxicity
There is currently inadequate information to assess the potential effects of toluene
on the reproductive system.
Developmental Toxicity
The data on human beings are inadequate for the purpose of assessing the
quantitative relationship between toluene exposure and developmental effects. Fetotoxicity
and persistent neurotransmitter changes have been identified as developmental effects.
Nawrot and Staples (1979) reported that a maternal dose of 440 mg/kg/day (but not
260 mg/kg/day) on days 6 through 15 of gestation caused significant reductions in fetal
body weight in mice. Hudak and Ungvary (1980) reported that female mice exposed to
500 mg/m3 of toluene continuously on days 6 through 13 of gestation caused a
significantly reduced fetal weights in their offspring. These exposures correspond to the
following LOAELs.
LOAEL (Low Fetal Weight - Toluene - Oral Exposure - Mice)
440 mg/kg/day
NOAEL (Low Fetal Weight - Oral Exposure - Mice)
260 me/kg/dav
LOAEL (Low Fetal Weight - Inhalation Exposure - Mice)
500 mg/m3 x 0.02 L/min x 0.001 m^/L x 60 min/hr x 24 hrs/day/0.025 kg
= 580 mg/kg/day
The human equivalent dose for the inhalation exposure effect level is derived by
multiplying that level by 0.1.
D-47
-------
LOAEL (Low Fetal Weight - Inhalation Exposure - Mice;
Human Equivalent Dose)
580 mg/kg/day x 0.1 = 58 mg/kg/day
Persistent changes in neurotransmitter levels and turnover were observed following
neonatal exposure to toluene. Rat neonates (weights not given) were exposed to 80 ppm of
toluene, 6 hours/day, on days 1 through 7 of gestation. This exposure produced changes
in the levels and utilization of monoamine neurotransmitters (dopamine and norepinephrine)
in the brain, and was also found to alter responses of the catecholamine neurons to
subacute toluene exposure in adulthood. If it is assumed that the neonates weighed 25
grams (the average weight for a rat neonate), an approximate LOAEL can be calculated as
follows.
LOAEL (Persistent Neurological Effects - Toluene - Rats)
300 mg/m3 x 0.02 L/min x 0.001 m3/L x 60 min/hr x 6 hrs/day/0.025 kg
= 86 mg/kg/dav
The human equivalent dose is derived by scaling the rat neonate body weight relative to a 5
kilogram infant, as previously described.
DA(25/25°-74) = DH x (5,000/5,000°-74)
2.3DA = 9.1DH
= 0.25DA
LOAEL (Persistent Neurological Effects - Toluene - Rats; Human Equivalent Dose)
86 mg/kg/day x 0.25 = 22 mg/kg/day
XYLENE
Lethality
The available human information in acute lethal exposures to xylene indicate that
"prolonged" exposures to 10,000 ppm of xylene can be fatal. This exposure is similar to
D-48
-------
the lethal exposure for toluene. Lethality in laboratory animals was observed following
oral xylene doses of 2 to 6 grams/kg.
An animal study identified an LCsQ of 6350 ppm (27,305 mg/m^) in rats exposed
for 4 hours to xylene. This corresponds to the following dose.
LOAEL (Lethality - Xylene - Rats)
27,305 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 4 hrs/day/0.25 kg = 2620 mg/kg
= 11 mg/kg/min
The human equivalent dose is derived by multiplying this dose level by 0.2.
LOAEL (Lethality - Xylene - Rats; Human Equivalent Dose)
2620 mg/kg x 0.2 = 525 mg/kg
11 mg/kg/min x 0.2 = 2.2 mg/kg/min
Pulmonary Toxicitv
No pulmonary effects at non-lethal xylene dose levels were identified in the
literature.
Kidnev and Liver Toxicitv
Effects of Acute Exposure
Kidney and liver damage have been observed in humans exposed to near lethal
concentrations of xylene. Data on lower level exposures were not found.
A single interperitoneal ingestion of 1,000 mg/kg of xylene increased serum
omithine carbamyl transferase (OCT) activity to 18.4, compared to 2.0 IU in controls. No
hepatocellular necrosis was observed.
Effects of Subacute and Subchronic Exposures
No data in human beings were found on the subacute or subchronic effects of
xylene exposure. Mortality occur in the absence of liver toxicity when rats, guinea pigs,
dogs, and monkeys were exposed to 7,790 ppm of xylene 8 hours/day, 5 days/week, for
30 exposures, or to 78 ppm of xylene continuously for 90 days. No treatment-related
D-49
-------
gross or microscopic pathologic lesions were observed in rats or mice exposed to as much
as 1,000 mg/kg/day of xylene for 13 weeks. Mean body weights were lower (8 to 15
percent) than controls in the 1,000 mg/kg/day dosed rats.
Effects of Chronic Exposure
No data on humans pertaining to the chronic exposure effects of xylene were
identified. Fischer 344 rats exposed to 500 mg/kg/day for 2 years of mixed xylenes
exhibited small decreases in body weight (5 to 8 percent compared to controls). There
were no significant increases in the incidence of microscopic lesions in any site among rats
exposed to 500 mg/kg/day of xylene or among mice exposed to 1,000 mg/kg/day of
xylene.
Male CFY rats exposed to 1,090 ppm of o-xylene for 8 hours/day, 7 days/week,
for 1 year exhibited hepatomegaly and increased activity in several hepatic enzymes. This
corresponds to a LOAEL of 912 mg/kg/day.
LOAEL (Hepatotoxicity - Xylene - Rats)
4750 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 8 hrs/day/0.25 kg
= 912 mg/kg/day
A human equivalent dose is derived by multiplying this level by 0.2
LOAEL (Hepatotoxicity - Xylene - Rats; Human Equivalent Dose)
912 mg/kg/day x 0.2 = 180 mg/kg/day
Neurotoxicitv
Effects of Acute Exposure
Men exposed to near lethal concentrations of xylene reported mental confusion and
amnesia for events that occurred 24 hours before the exposure. Other, less severe
symptoms included headache, nausea, vomiting, vertigo, dizziness, and sensory irritation.
In animal studies, lack of coordination, prostration, and hunched posture was
observed in rats receiving 4,000 mg/kg of xylene. Tremors and/or slow breathing were
observed in mice receiving a 4,000 mg/kg xylene dose.
D-50
-------
Sensory irritation was observed in human beings exposed to 1 10 to 460 ppm (472
to!980 mg/m3) of xylene for 3 to 5 minutes (Sandmeyer, 1983). This corresponds to a
LOAELofO.l mg/kg/min.
LOAEL (Sensory irritation - Xylene - Humans)
472 mg/m3 x 0.02 m^/min x 3 min/70 kg = 0.4
= 0.1 mg/kg/min
Effects of Subacute and Subchronic Exposures
Weakness, lethargy, short and shallow breathing, tremors, and paresis were
observed in mice dosed with 2,000 mg/kg/day of xylene, 5 days/week, for 13 weeks.
These signs persisted for IS to 60 minutes after dosing.
Effects of Chronic Exposure
Hyperactivity was observed in mice dosed with 1,000 mg/kg.day of xylene, 5
days/week for 2 years.
Hematotoxicitv
No hematotoxic effects have been identified in associated with xylene exposure.
Genetic Toxicity
The information is inadequate to evaluate the potential effects of xylene exposure on
genes or chromosomes.
Carcinogcnicitv
The information is currently inadequate to evaluate the potential carcinogenicity of
xylene.
Reproductive and Developmental Toxicities
Reproductive Toxicitv
Pregnant female rats were exposed to 150 mg/m^ of xylene for 24 hours/day during
days 7 through 14 of gestation (Ungvary et al., 1980). Placenta! weights were decreased
in this exposure group relative to the controls. The effect level for this exposure is
calculated as follows.
D-51
-------
LOAEL (Decreased Placental Weight - Xylene - Rats)
150 mg/m3 x 0.1 L/min x 0.001 m3/L x 60 min/hr x 24 hrs/day/0.25 kg
= 86 mg/kg/day
The human equivalent dose is derived by multiplying this level by 0.2.
LOAEL (Decreased Placental Weight - Xylene - Rats; Human Equivalent Dose)
86 mg/kg/day x 0.2 = 17 mg/kg/day
Developmental Toxicity
Several measures of embryotoxicity and fetotoxicity (e.g, increased rcsorptions,
reduced fetal body weights, delayed ossification, or retarded skeletal development) have
been significantly increased in association with xylene exposure. It is difficult, however,
to identify these as primary effects or as effects secondary to maternal toxicity. The lowest
effect level associated with fetotoxicity in rats, for example, was 24-hour exposure to 150
mg/nv- This exposure was also toxic to the mothers, as described above in the
Reproductive Toxicity section. Thus, the lowest effect level for developmental toxicity is
the same as the lowest effect level for reproductive toxicity.
SUMMARY
A summary of the animal doses for the animal LOAELs and NOAELs derived in
this section is presented in Tables D-2a-d A summary of the human LOAELs and
NOAELs for gasoline, benzene, toluene, and xylene is presented in Tables D-3. When
systemic effects are evaluated, a scaling factor is applied to this dose to obtain a human
equivalent dose. Such scaling adjusts the animal dose by about a factor of 0.1 to 0.3.
Tables D-4a-d presents a summary of both the human and animal doses corresponding to
the various LOAELs and NOAELs. The equivalent concentrations are derived by
multiplying these dose levels by 70 kilograms, and by dividing either by 14.4 m3/day (for
systemic effects, corresponding to an alveolar ventilation rate of 10 L/minute) or by 21.6
m^/day (for respiratory effects, corresponding to a total ventilation rate of 15 L/minute).
The exceptions to this procedure is with the neonatal toluene effects, for which a reference
population is the 10 kilogram infant (alveolar ventilation rate of 1.7 m^/hour), and when
exposure durations less than 24 hours are being assessed. In this latter case, the ventilation
D-52
-------
rate is based on the exposure duration of the critical study. These are presented in Tables
D-5a-d.
D-53
-------
TABLE D-2a
SUMMARY OF ANIMAL LOAELS AND NOAELS DERIVED FROM GASOLINE HEALTH EFFECTS STUDIES
Endpoint
Type
Animal Dose Human Equivalent Exposure Reference
Dose Interval
Respiratory
Irritation
(Unleaded Gasoline)
Ventilation Changes
(Leaded Gasoline)
Ventilation Changes
(Leaded Gasoline)
Ventilation Changes
(Unleaded Gasoline)
Ventilation Changes
(Unleaded Gasoline)
Lung Pathology
(Leaded Gasoline)
Kidney Toxicity
(Unleaded Gasoline)
Kidney Toxicity
(Full Range Alkylate)
Kidney Toxicity
(Full Range Alkylate)
LOAEL 80 mg/kg/min 80mg/kg/min minutes Phillips, 1984
LOAEL 58 mg/kg/day 58mg/kg/day 24 hour Kuna & Ulrich, 1984
NOAEL 16mg/kg/day 16mg/kg/day
LOAEL
24 hour Kuna & Ulrich, 1984
LOAEL 239 mg/kg/day 239 mg/kg/day 24 hour Kuna & Ulrich, 1984
NOAEL 59 mg/kg/day 59 mg/kg/day
LOAEL 53 mg/kg/day 53 mg/kg/day
24 hour Kuna & Ulrich, 1984
24 hour LeMesurier et al., 1980
17 mg/kg/day 2.6 mg/kg/day 24 hour Haider etal., 1984
LOAEL 19 mg/kg/day 4 mg/kg/day
24 hour Haider et al., 1984
NOAEL 1.7 mg/kg/day 0.3 mg/kg/day 24 hour Haider etal., 1984
Kidney Toxicity
(25% Blend)
NOAEL 1370 mg/kg/day 274 mg/kg/day 24 hour Haider etal., 1984
-------
TABLE D-2a
(continued)
Endpoint
Kidney Toxicity
(0-145° Distillate)
Kidney Toxicity
(0-145° Distillate)
Kidney Toxicity
(Unleaded Gasoline)
Visual Evoked Response
(Unleaded Gasoline)
o
(I* Visual Evoked Response
** (Leaded Gasoline)
Neuropathy, Neurobehavior
(Unleaded Gasoline)
Hematotoxicity
(Unleaded Gasoline)
Hematotoxicity
(Unleaded Gasoline)
Hematotoxicity
(Leaded Gasoline)
Hematotoxicity
Type
LOAEL
NOAEL
LOAEL
NOAEL
NOAEL
NOAEL
LOAEL
NOAEL
LOAEL
NOAEL
Animal Dose
1610 mg/kg/day
370 mg/kg/day
21 mg/kg/day
287 mg/kg/day
69 mg/kg/day
635 mg/kg/day
480 mg/kg/day
1 17 mg/kg/day
116 mg/kg/day
32 mg/kg/day
Human Equivalent
Dose
320 mg/kg/day
74 mg/kg/day
4 mg/kg/day
144 mg/kg/day
35 mg/kg/day
127 mg/kg/day
96 mg/kg/day
23 mg/kg/day
23 mg/kg/day
6.4 mg/kg/day
Exposure
Interval
24 hour
24 hour
chronic
24 hour
24 hour
24 hour
24 hour
24 hour
24 hour
24 hour
Reference
Aranyi et al., 1986
Aranyi et al., 1986
MacFarland et al., 1984
Kuna & Ulrich, 1984
Kuna & Ulrich, 1984
Spencer etal., 1982
Kuna and Ulrich, 1984
Kuna and Ulrich, 1984
Kuna and Ulrich, 1984
Kuna and Ulrich, 1984
(Unleaded Gasoline)
-------
TABLE D-2a
(continued)
Endpoint Type Animal Dose Human Equivalent Exposure Reference
Dose Interval
Reproductive Effects LOAEL 28mg/kg/day 6mg/kg/day 24 hour
(Gasoline)
Fetotoxicity LOAEL 225mg/kg/day 4Smg/kg/day 24 hour
(Unleaded Gasoline)
-------
TABLE D-2b
SUMMARY OF ANIMAL LOAELS AND NOAELS DERIVED FROM BENZENE HEALTH EFFECTS STUDIES
Endpoint
Lethality
Respiratory Tract
Irritation
Reflex Activity
Reflex Activity
Neurobehavioral
o
Oi Neurotransmitter Changes
Neurotransmitter Changes
Hematotoxicity
(Cytopenia)
Hematotoxicity
(Cytopenia)
Hematotoxicity
(Cytopenia)
Hematotoxicity
(Spleen Changes)
Clastogenicity
Clastogenicity
Type
LC50
NOAEL
LOAEL
NOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
NOAEL
Animal Dose
1 1 mg/kg/day
20 mg/kg/day
6 mg/kg/day
1.2 mg/kg/day
6 mg/kg/day
648 mg/kg/day
8 mg/kg/day
82 mg/kg/day
8.6 mg/kg/day
8 mg/kg/day
0.9 mg/kg/day
0.4 mg/kg/day
0.1 mg/kg/day
Human Equivalent
Dose
2 mg/kg/day
20 mg/kg/day
1.2 mg/kg/day
0.2 mg/kg/day
0.6 mg/kg/day
130 mg/kg/day
0.8 mg/kg/day
16 mg/kg/day
0.9 mg/kg/day
0.8 mg/kg/day
0.09 mg/kg/day
0.08 mg/kg/day
0.02 mg/kg/day
Exposure
Interval
4 hours
30min
24 hours
24 hour
24 hour
24 hour
24 hour
hours
24 hours
24 hours
24 hours
24 hours
24 hours
Reference
IARC, 1982
Nielson and Alarie, 1!
Novikov, 1956
Novikov, 1956
Horiuchi et al., 1967
Dempster et al., 1984
Hsieh et al.,
Toft etal., 1982
Baarson et al., 1984
Hsieh et al.,
Green etal., 1981
Erexson et al., 1986
Erexson et al., 1986
-------
TABLE D-2b
(continued)
Endpoint
[Inhibition of Mitosis
nhibition of Mitosis
Reproductive Toxicity
Fetotoxicity
Developmental Effects
Type
LOAEL
NOAEL
LOAEL
LOAEL
LOAEL
Animal Dose
1.2 mg/kg/day
0.4 mg/kg/day
185 mg/kg/day
4 mg/kg/day
4 mg/kg/day
Human Equivalent
Dose
0.2 mg/kg/day
0.08 mg/kg/day
18.5 mg/kg/day
0.4 mg/kg/day
0.4 mg/kg/day
Exposure
Interval
24 hours
24 hours
24 hour
24 hour
24 hour
Reference
Erexson et al., 1986
Erexson et al., 1986
Ward et al., 1985
Keller etal., 1986
Keller etal., 1986
(Hematotoxic)
in
oo
-------
TABLE D-2c
SUMMARY OF ANIMAL LOAELS AND NOAELS DERIVED FROM TOLUENE HEALTH EFFECTS STUDIES
Endpoint
Lethality
Neurotoxicity
Attenuation, visual -motor
Neurotoxicity
Attenuation, visual -motor
Neurotoxicity
CO2 expiration
Neurotoxicity
Open field behavior
Neurotoxicity
Open field behavior
Neurotoxicity
Rotorod performance
Hepatotoxicity
Kidney/Liver Toxicity
Respiratory Infections
Respiratory Infections
Pulmonary Toxicity
Hematotoxicity
Type
LOAEL
LOAEL
NOAEL
LOAEL
LOAEL
NOAEL
LOAEL
LOAEL
NOAEL
LOAEL
NOAEL
NOAEL
LOAEL
Animal Dose
6 mg/kg/day
1.2 mg/kg/day
0.6 mg/kg/day
0.06 mg/kg/day
72 mg/kg/day
14.4 mg/kg/day
3 mg/kg/day
1440 mg/kg/day
116 mg/kg/day
1.4 mg/kg/day
0.5 mg/kg/day
116 mg/kg/day
11 mg/kg/day
Human Equivalent
Dose
1 mg/kg/day
0.6 mg/kg/day
0.3 mg/kg/day
0.03 mg/kg/day
7.2 mg/kg/day
1.4 mg/kg/day
0.3 mg/kg/day
290 mg/kg/day
23 mg/kg/day
1.4 mg/kg/day
0.5 mg/kg/day
116 mg/kg/day
1.1 mg/kg/day
Exposure
Interval
24 hour
25min
20min
24 hours
24 hours
24 hours
24 hour
24 hour
24 hour
24 hour
chronic
24 hour
Reference
Smyth et al., 1969
Taylor and Evans, 1985
Taylor and Evans, 1985
Taylor and Evans, 1985
Kostas and Hotchins,
Kostas and Hotchins,
Kostas and Hotchins,
Ungvary et al., 1982
CUT, 1983
Aranyietal., 1985
Aranyi et al., 1985
CUT, 1983
Horiuchi, 1977
-------
TABLE D-2c
(continued)
9
8
Endpoint
Hematotoxicity
Hematotoxicity
Neurobehavioral Effects
Fetotoxicity
Developmental Effects
Neurotransmitter Changes
Type
NOAEL
LOAEL
LOAEL
LOAEL
LOAEL
Animal Dose
1 mg/kg/day
39 mg/kg/day
1 mg/kg/day
580 mg/kg/day
86 mg/kg/day
Human Equivalent
Dose
0.1 mg/kg/day
8 mg/kg/day
0.1 mg/kg/day
58 mg/kg/day
22 mg/kg/day
Exposure
Interval
24 hour
chronic
24 hour
24 hour
24 hour
Reference
Horiuchi, 1977
CUT, 1983
Horiguchi ei al., 1977
Hudak and Ungvary, 1978
von Euler et al., 1989
-------
TABLE D-2d
SUMMARY OF ANIMAL LOAELS AND NOAELS DERIVED FROM THE XYLENE
HEALTH EFFECTS STUDIES
Endpoint
Type
Animal Dose Human Equivalent Exposure Reference
Dose Interval
Lethality LC50
Hepatotoxicity LOAEL
Reproductive Toxicity LOAEL
1 1 mg/kg/min
912mg/kg/day
86 mg/kg/day
2.2 mg/kg/min
180 mg/kg/day
17 mg/kg/day
4 hours Sandmeyer, 1983
chronic Tatrai et al., 1980
24 hour Ungvary et al., 1980
-------
TABLE D-3
SUMMARY OF HUMAN LOAELS AND NOAELS DERIVED FROM STUDIES ON GASOLINE.
BENZENE, TOLUENE, AND XYLENE
Substance
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Benzene
Benzene
Toluene
Toluene
Toluene
Toluene
Toluene
Xylene
Xylene
Endooint
Lethality
Dizziness
Anesthesia
Nose and
Throat Irritation
Irritation
Slight Dizziness
Eye Irritation
Lethality
Leucopenia
Lethality
Sensory Irritation
Sensory Irritation
Neurotoxicity
Hematotoxicity
Lethality
Irritation
Tvoe
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
LOAEL
NOAEL
LOAEL
LOAEL
LOAEL
LOAEL
Human Dose
3 mg/kg/min
6 mg/kg/min
6 mg/kg/min
0.3 mg/kg/min
0.1 mg/kg/min
12 mg/kg/min
6 mg/kg/day
7 mg/kg/day
6.5 mg/kg/hr
2.5 mg/kg/yr
29 mg/kg/day
16 mg/kg/day
7 mg/kg/day
0.1 mg/kg min
Exposure Interval
5 minutes
4 minutes
2 minutes
30 minutes
minutes-hours
5 minutes
24 hours
15 minutes
6 hours
6 hours
chronic
chronic
15 minutes
3-5 minutes
Reference
Wang and Irons, 1961
Poklis and Burkett, 1977
Poklis and Burkett, 1977
Drinker et al., 1943
Drinker et al., 1943
IARC, 1982
Askoy et al., 1976
U.S. EPA, 1983b
Andersen et al., 1983
Andersen et al., 1983
Iregren, 1982
Stockholm, 1979
Sandmeyer, 1983
Sandmeyer, 1983
-------
TABLE D-4a
SUMMARY OF HUMAN EQUIVALENT DOSES AND EQUIVALENT AIR
CONCENTRATIONS FOR LOAELS AND NOAELS DERIVED FOR
GASOLINE
Substance
Gasoline • NS
Gasoline • NS
Gasoline - NS
Gasoline - NS
Gasoline - NS
Gasoline - U
Gasoline - L
Gasoline - L
Gasoline - U
Gasoline - U
Gasoline - L
Gasoline - U
Gasoline - U
Gasoline - FRA
Gasoline - FRA
Gasoline Blend
Gasoline -0-145°
Gasoline - 0-145°
Gasoline - U
Gasoline - L
Endpoint
Lethality
Anesthesia
Irritation
Irritation
Eye Irritation
Irritation
Ventilation
Ventilation
Ventilation
Ventilation
Lung Pathology
Kidney Effects
Kidney Effects
Kidney Effects
Kidney Effects
Kidney Effects
Kidney Effects
Kidney Effects
Visual Effects
Visual Effects
Type
LOAEL-H
LOAEL-H
LOAEL-H
LOAEL-H
LOAEL-H
LOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
NOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
NOAEL-A
NOAEL-A
Human
Equivalent Dose
3 mg/kg/min
6 mg/kg/min
6 mg/kg/min
0.3 mg/kg/min
0.1 mg/kg/day
80 mg/kg/min
58 mg/kg/day
16 mg/kg/day
239 mg/kg/day
59 mg/kg/day
53 mg/kg/day
2.6 mg/kg/day
4 mg/kg/day
4 mg/kg/day
0.3 mg/kg/day
274 mg/kg/day
320 mg/kg/day
74 mg/kg/day
144 mg/kg/day
35 mg/kg/day
Equivalent Air
Cone fmg/m^l
15,000
30,000
30,000
1,500
500
280,000
190
52
775
191
172
13
19
19
1.5
1332
1560
360
700
170
D-63
-------
TABLE D-4a
(continued)
Substance
Gasoline - U
Gasoline - U
Gasoline - U
Gasoline - L
Gasoline - L
Gasoline - NS
Gasoline - U
Endpoint
Neuropathy,
Behavioral
Hematotoxicity
Hematotoxicity
Hematotoxicity
Hematotoxicity
Reproductive
Effects
Fetotoxicity
Type
NOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
Human
Equivalent Dose
127 mg/kg/day
96 mg/kg/day
23 mg/kg/day
23 mg/kg/day
6.4 mg/kg/day
6 mg/kg/day
45 mg/kg/day
Equivalent Air
Cone (mg/m^
617
467
112
112
31
29
218
Gasoline - NS: Gasoline (not specified)
Gasoline - U: Unleaded gasoline
Gasoline -L: Leaded gasoline
Gasoline - FRA: Full range alkylate
Gasoline - Blend: 24% blend of butane, isobutane, pentane, isopentane
Gasoline -1-145°: 0-145° distillate fraction of gasoline
D-64
-------
TABLE D-4b
SUMMARY OF HUMAN EQUIVALENT DOSES
AND EQUIVALENT AIR CONCENTRATIONS FOR LOAELS AND
NOAELS DERIVED FOR BENZENE
Endpoint
Lethality
Respiratory Irritation
Reflex Activity
Reflex Activity
Neurobehavioral
Neurotransmittor
Changes
Cytopenia
Cytopenia
Cytopenia
Leucopenia
Spleen Changes
Clastogenicity
Clastogenicity
Mitotic Effects
Mitotic Effects
Reproductive Effects
Fetotoxicity
Type
LOAEL-H
NOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-H
LOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
Human Equivalent Air
Equivalent Dose Concentration (me/m$)
12 mg/kg/min
20 mg/kg/day
1.2 mg/kg/day
0.2 mg/kg/day
0.6 mg/kg/day
0.8 mg/kg/day
3 mg/kg/day
0.9 mg/kg/day
0.8 mg/kg/day
6 mg/kg/day
0.09 mg/kg/day
0.08 mg/kg/day
0.02 mg/kg/day
0.2 mg/kg/day
0.08 mg/kg/day
18.5 mg/kg/day
5 mg/kg/day
66,000
70,000
5.8
1.0
2.9
3.9
350
4.4
3.9
29
0.4
0.4
0.1
1.0
0.4
90
19
Low birth wt.
Fetotoxicity
Low birth wt.
NOAEL-A
1 mg/kg/day
D-65
-------
Endpoint
Type
TABLE D-4b
(continued)
Human
Eauivalent Dose
Equivalent Air
Concentration fme/m')
Fetotoxicity
Hematotoxidty
Developmental
Toxicity - Hemat.
Developmental
Toxicity - Hemat.
LOAEL-A
LOAEL-A
NOAEL-A
0.4 mg/kg/day
0.8 mg/kg/day
0.4 mg/kg/day
1.9
3.8
1.9
D-66
-------
TABLE D-4c
SUMMARY OF HUMAN EQUIVALENT DOSES
AND EQUIVALENT AIR CONCENTRATIONS FOR LOAELS AND
NOAELS DERIVED FOR TOLUENE
Endpoint
Lethality
Neurotoxicity
Attenuation,
visual-motor
Neurotoxicity
Attenuation,
visual-motor
Neurotoxicity
CO2 expiration
Neurotoxicity
Rotorod perf .
Neurotoxicity
Open field behav.
Neurotoxicity
Open field behav.
Lethality
Hepatotoxicity
Kidney/ Liver Tox.
Respiratory Inf.
Respiratory Inf.
Pulmonary Tox.
Sensory Irritation
Sensory Irritation
Neurotoxicity
Type
LOAEL-H
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
NOAEL-A
LOAEL-H
NOAEL-H
LOAEL-H
Human Equivalent Air
Equivalent Dose Concentration (me/m^)
7mg/kg/min
0.6 mg/kg/min
0.3 mg/kg/min
0.03 mg/kg/min
0.3 mg/kg/min
7.2 mg/kg/min
1.4 mg/kg/min
1 mg/kg/min
290 mg/kg/day
23 mg/kg/day
1.4 mg/kg/day
0.5 mg/kg/day
1 16 mg/kg/hr
6.5 mg/kg/hr
2.5 mg/kg/hr
29 mg/kg/day
37,000
4,200
2,100
210
1.5
35
6.8
5,000
1,400
112
4.5
1.6
564
375
150
140
D-67
-------
Endpoint
Type
TABLE D-4c
(continued)
Human
Eauivalent Dose
Equivalent Air
Concentration
Neurobehavioral
Hematotoxicity
Hematotoxicity
Hematotoxicity
Hematotoxicity
Fetotoxicity
Developmental Effects
* •
LOAEL-A
LOAEL-H
LOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
0.1 mg/kg/day
16 mg/kg/day
1.1 mg/kg/day
0.1 mg/kg/day
8 mg/kg/day
58 mg/kg/day
22 mg/kg/day
0.5
78
5.3
0.5
39
282
107
Neuro.
D-68
-------
TABLE D-4d
SUMMARY OF HUMAN EQUIVALENT DOSES
AND EQUIVALENT AIR CONCENTRATIONS FOR LOAELS AND
NOAELS DERIVED FOR XYLENE
Endpoint
Type
Human
Equivalent Dose
Equivalent Air
Concentration
Lethality
Lethality
Irritation
Hepatotoxicity
Reproductive
Toxicity
LOAEL-H
LC50-A
LOAEL-H
LOAEL-A
LOAEL-A
7mg/kg/min
2.2 mg/kg/min
0.1 mg/kg/min
180mg/kg/day
17 mg/kg/day
37,000
15,400
470
875
83
D-69
-------
Substance
SUMMARY
Endpoint
TABLE D-5a
OF REFERENCE AIR CRITERIA FOR GASOLINE
Type
Human
Equivalent Dose
Uncertainty Reference Air Criteria (ug/m^)
Factor Adults Infants
Gasoline - NS
Gasoline - NS
Gasoline - NS
Gasoline - NS
Gasoline - NS
Gasoline - U
Gasoline - L
Gasoline - U
Gasoline - L
Gasoline - U
Gasoline - U
Gasoline - FRA
Gasoline Blend
Gasoline - 0-145°
Gasoline - U
Lethality
Anesthesia
Irritation
Irritation
Eye Irritation
Irritation
Ventilation
Ventilation
Lung Pathology
Kidney Effects
Kidney Effects
Kidney Effects
Kidney Effects
Kidney Effects
Visual Effects
LOAEL-H
LOAEL-H
LOAEL-H
LOAEL-H
LOAEL-H
LOAEL-A
NOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
NOAEL-A
NOAEL-A
NOAEL-A
NOAEL-A
3 mg/kg/min
6 mg/kg/min
6 mg/kg/min
0.3 mg/kg/min
0.1 mg/kg/day
80 mg/kg/min
16 mg/kg/day
59 mg/kg/day
53 mg/kg/day
2.6 mg/kg/day
4 mg/kg/day
0.3 mg/kg/day
274 mg/kg/day
74 mg/kg/day
144 mg/kg/day
100
100
100
100
100
500
100
100
500
500
500
100
100
100
100
210,000
420,000
280,000
21,000
4,700
746,000
520
1,910
340
26
38
15
13,300
3,600
7,000
150,000
300,000
222,000
15,000
3,700
600,000
400
1,500
266
20
30
11
10,150
2,700
5,300
-------
TABLE D-5a
(continued)
2
Substance
Gasoline - L
Gasoline - U
Gasoline - U
Gasoline - L
Gasoline - NS
Gasoline - U
Endpoint
Visual Effects
Neuropathy,
Behavioral
Hematotoxicity
Hematotoxicity
Reproductive
Effects
Fetntnxicitv
Type
NOAEL-A
NOAEL-A
NOAEL-A
NOAEL-A
LOAEL-A
inART^A
Human
Equivalent Dose
35 mg/kg/day
127 mg/kg/day
23 mg/kg/day
6.4 mg/kg/day
6 mg/kg/day
4S maflcoMav
Uncertainty
Factor
100
100
100
100
500
snn
Reference Air Criteria (uj
Adults Infants
1,700
6,200
1,120
310
38
41A
1,300
4,700
850
240
44
-------
TABLE D-5b
SUMMARY OF REFERENCE AIR CRITERIA FOR BENZENE
Substance
Lethality
Respiratory Irritation
Reflex Activity
Neurobehavioral
Neurotransmittor
Changes
Cytopenia
Cytopenia
Cytopenia
Leucopenia
Clastogenicity
Mitotic Effects
Mitotic Effects
Reproductive Effects
Fetotoxicity
Low birth wt.
Fetotoxicity
Hematotoxicity
Developmental
Type
LOAEL-H
NOAEL-A
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
LOAEL-H
LOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
LOAEL-A
NOAEL-A
Human
Equivalent Dose
12 mg/kg/min
20 mg/kg/day
0.2 mg/kg/day
0.6 mg/kg/day
0.8 mg/kg/day
3 mg/kg/day
0.9 mg/kg/day
0.8 mg/kg/day
6 mg/kg/day
0.08 mg/kg/day
0.2 mg/kg/day
0.08 mg/kg/day
18.5 mg/kg/day
1 mg/kg/day
0.4 mg/kg/day
0.4 mg/kg/day
Uncertainty
Factor
100
100
100
500
500
500
500
500
100
500
100
100
500
100
500
100
Reference Air Criteria (ug/nv*)
Adults Infants
840,000
930,000
10
5.8
7.8
700
8.8
6.8
290
0.8
1.0
3.9
180
49
3.8
19
600,000
740,000
7.4
4.4
6.0
600
6.6
6.0
220
0.6
0.7
3.0
138
Toxicity - Hemat.
-------
TABLE D-5c
SUMMARY OF REFERENCE AIR CRITERIA FOR TOLUENE
Substance
Lethality
Lethality
Type
LOAEL-H
LOAEL-A
Neurotoxicity NOAEL-A
Attenuation.visual-motor
Neurotoxicity
CO2 expiration
LOAEL-A
9 Neurotoxicity LOAEL-A
£j Rotorod performance
Hepatotoxicity
Kidney/ Liver Tox.
Respiratory Inf.
Pulmonary Tox.
Sensory Irritation
Neurotoxicity
Neurobehavioral
Hematotoxicity
LOAEL-A
NOAEL-A
NOAEL-A
NOAEL-A
NOAEL-H
LOAEL-H
LOAEL-A
LOAEL-H
Human Uncertainty
Equivalent Dose Factor
7 mg/kg/min
1 mg/kg/min
0.3 mg/kg/min
0.03 mg/kg/min
0.3 mg/kg/min
290 mg/kg/day
23 mg/kg/day
0.5 mg/kg/day
116mg/kg/hr
2.5 mg/kg/hr
29 mg/kg/day
0.1 mg/kg/day
16 mg/kg/day
100
500
100
500
500
500
100
100
100
10
100
500
100
Reference Air Criteria (ug/m:
Adults Infants
490,000
14,000
21,000
420
3.0
2,800
1,120
16
5,640
19,000
1,400
1.0
780
350,000
10,000
15,000
300
2.2
2,200
850
12.5
4,300
15,000
1,100
0.8
590
-------
TABLE D-5c
(continued)
Substance Type Human Uncertainty Reference Air Criteria (ug/m^)
Equivalent Dose Factor Adults Infants
Hematotoxicity
Hematotoxicity
Fetotoxicity
Developmental
NOAEL-A
LOAEL-A
LOAEL-A
LOAEL-A
0.1 mg/kg/day
8 mg/kg/day
58 mg/kg/day
22 mg/kg/day
100
500
500
500
5.4
78
564
.
4.1
60
-
160
Effects - Neuro.
-------
9
in
TABLE D-5d
SUMMARY OF REFERENCE AIR CRITERIA FOR XYLENE
Endpoint
Lethality
Irritation
Hepatotoxicity
Reproductive
Toxicity
Type
LOAEL-H
LOAEL-H
LOAEL-A
LOAEL-A
Human
Equivalent Dose
7mg/kg/rnin
0.1 mg/kg/min
180mg/kg/day
17 mg/kg/day
Uncertainty
Factor
100
100
500
500
Reference Air Criteria (ug/m-*)
Adults Infants
490,000
4,700
1750
166
350,000
3,700
1340
126
-------
REFERENCES
Adamson, R.H., and Seiber, S.M. 1981. Chemically induced leukemia in humans. Env.
Health Persp. 39:93-103.
Ainsworth, R.W. 1960. Petrol-vapour poisoning. British Med. Journal 1:1547-1548.
\ksoy, M. 1980. Different types of malignancies due to occupational exposure to
benzene: a review of recent observations in Turkey. Environ. Res. 23:181-190.
Aksoy, M., and Erdem, S. 1978. Follow-up study on the mortality and the development
of leukemia in 44 panocytopenic patients with chronic exposure to benzene. Blood
52:285.
Aksoy, M., Dincol, K., Akgun, T., Erdmen, S., and Dinkol, G. 1971. Haematological
effects of chronic benzene poisoning in 217 workers. Br. J. Ind. Med. 28:296-302.
Aksoy, M., Dincol, K., Erdem, S., and Dincol, G. 1972a. Acute leukemia due to chronic
exposure to benzene. Amer. J. Med. 52:160-166.
Aksoy, M., Dincol, K., Erdem, S., and Dincol, G. 1972b. Details of blood changes in
32 patients with pancytopenia associated with long term exposure to benzene. Br. J.
Ind. Med. 29:56-64.
Aksoy, M., Erdem, S., and Dincol, G. 1975. Two rare complications of chronic benzene
poisoning: myeloid metaplasia and paroxsymal noctural hemaglobinurea. Blut. 30:255-
260.
Aksoy, M., Erdem, S., and Dincol, G. 1976. Types of leukemia in chronic benzene
poisoning. A study in 34 patients. Acta Haematol. 55:65-72.
Aksoy, M., Erdem, S., Dincol, K., Hepyuksel, T., and Dincol, G. 1974. Chronic
exposure to benzene as a possible contributar etiologic factor in Hodgkin's disease.
Blut. 38:293-298.
Amacher, D. E. and Zelljadt, J. 1983. The morphological transformation of Syrian
hamster embryo cells by chemical reportedly non-mutagenic to Salmonella
typhimurium. Carcinogenesis (London) 4:291-295.
American Petroleum Institute (API). 1976. Evaluation of the three month inhalation
toxicitv of two motor fuels. Submitted by Huntingdon Research Center, New York, to
The American Petroleum Institute April 14,1976. API Medical Research Publication
27-32610. Washington, DC: American Petroleum Institute.
_. 1984. Information on unleaded gasoline. Submitted to the American
Petroleum Institute by S.M. Swanson, December 19, 1984.
. 1985a. Comments on information review - unleaded gasoline (IR-469).
Submitted to the American Petroleum Institute by S.M. Swanson, April 22,1985.
. 1985b. Laboratory study on solubilities of petroleum hydrocarbons in ground
water. API Publication 4395. Washington, DC: American Petroleum Institute.
R-l
-------
1985c. Literature survey. Hydrocarbon solubilities and attenuation
mechanics. API Publication 4414. Washington, DC: American Petroleum Institute.
, 1985d. Two generation reproduction and fertility study on a petroleum-
derived hydrocarbon (i.e., toluene). Volume 1. Washington, DC: American Petroleum
Institute.
. 1986a. Benzene in Florida groundwater. An assessment of the significance to
human health. Washington, DC: Florida Petroleum Council, A Division of the
American Petroleum Institute.
. 1986b. Exposure data on C7 and C8 aromarics during handling and
production of motor gasolines. API Publication 4439. Washington, DC: American
Petroleum Institute.
. 1988. Comments of the American Petroleum Institute on the draft document
"Evaluation of the Health Effects from Exposure to Gasoline Vapors." March 2,1988.
American Society for Testing and Materials (ASTM). 1984. Standard specifications for
automotive gasoline. In 1984 Annual book of ASTM standards. Petroleum products.
lubricants, and fossil fuels, pp. 207-223.
Amorati, A., Cacciari, C., and Troisi, P.M. 1952. Research on chronic toxic effects from
long exposure to vapors of pure gasoline. Ind. Med. Surg. 21:466-468.
Amsel, L. P., and Levy, G. 1969. Drug biotransfprmation in man. II. A pharmacokinetic
study of the simultaneous conjugation of benzoic and salicylic acids with glycine. I
Pharmaceut. Science 58:321-6.
Andelman, J.B. 1985. Inhalation exposure in the home to volatile organic contaminants of
drinking water. The Science of the Total Environment 47:443-460.
Andersen, M.E. 1981. Saturable metabolism and its relationship to toxicity. In CRC
critical reviews of toxicology, pp. 105-50. Boca Raton, FL: Chemical Rubber
Company (CRC) Press.
Andersen, M.E. 1983. Flow-limited clearance. In Modeling of inhalation elimination.
Vol. 2. Edited by V. Fiserova-Bergerova. Boca Raton, FL: Chemical Rubber
Company (CRC) Press.
Anderson, D., and Richardson, C. 1981. Issues relevant to the assessment of chemically
induced chromosome damage in vivo and their relationship to chemical mutagenesis.
Mutat. Res. 90:261-272.
Anderson, E., Browne, N., Duletsky, S., et al. 1984. Development of statistical
distributions or ranges of standard factors used in exposure assessment. Revised draft
final report. Washington, DC: U.S. Environmental Protection Agency, Office of
Health and Environmental Assessment, Exposure Assessment Group.
Anderson, I., Lundqvist, G.R., Molhare, L., Pedersen, O.F., Proctor, D.F., Vaeth, M.,
and Wyon, D.P. 1983b. Human exposure to controlled levels of toluene in six-hour
exposures. Scand. J. Work Environ. Health 9(51:405-418.
R-2
-------
Anderson, K., Nilsen, O.G., Toftgard, R., Eneroth, P., Gustafsson, J., Battinstini, N.,
and Agnati, L.F. 1983. Increased amine turnover in several hypothalamic
noradrenaline nerve terminal systems and changes in prolactin secretion in the male rat
by exposure to various concentrations of toluene. Neurotoxicologv 4:43056.
Anderson, N. T. 1987. Risk assessment document for toluene. Final report. Augusta,
ME: State of Maine Bureau of Health.
Andrews, L.S., Lee, E.W., Witmer, C.M., Kocsis, J.J., and Snyder, R. 1977. Effects
of toluene on the metabolism, disposition and hemopoietic toxicity of [3H]-benzene.
Biochem. Pharmacol. 26:293-300.
Anonymous. 1967. Am. Med. Assoc. News 10:1.
Aoyama, K. 1986. Effects of benzene inhalation on lymphocyte subpopulations and
immune response in mice. Toxicol. Appl. Pharmacol. 85:92-101.
Aranyi, C. 1984. Thirteen-week inhalation toxicity study of C4/C5 hydrocarbon blends in
rats. Final Report. Submitted by the ITT Research Institute, October 1984.
Washington, DC: The American Petroleum Institute.
Aranyi, C. 1985. Effects of toluene on the metabolism, disposition, and hemopoietic
toxicity of irritated benzene. Biochem. Pharmacol. 26:283-300.
Aranyi, C., O'Shea, W.J., Haider, C.A., Holdsworth, C.E., and Cockrell, B.Y. 1986.
Absence of hydrocarbon-induced nephropathy in rats exposed subchronically to volatile
hydrocarbon mixtures pertinent to gasoline. Toxicology and Industrial Health 2:85-98.
Arp, E.W., et al. 1983. Lymphocytic leukemia and exposures to benzene and other
solvents in the rubber industry. J. Occup. Med. 25:598-602.
Arthur, L.J.H., and Curnock, D.A. 1982. Xylene-induced epilepsy following innocent
glue sniffing, (letter) Br. Med. J. 284:1787.
Askergren, A. 1981. Studies on kidney function in subjects exposed to organic solvents.
in. Excretion of cells in the urine. Acta Med. Scand. 210:103-106.
Askergren, A. 1982. Organic solvents and kidney function. Adv. Mod. Environ.
Ioxi£2l.2:157-172. (CA 98:184830)
Askergren, A., Allgen, L.G., Karlsson, C., Lundberg, I., and E. Nyberg. 1981. Kidney
function in subjects exposed to organic solvents: 1. Excretion of albumin and B-2-
microglobulin in the urine. Acta Med. Scand. 209:479-484.
Astrand, I. 1975. Uptake of sol vents in the blood and tissues of man: a review. Scand. J.
Work Environ. Health 1:199-218.
Astrand, I. 1983. Effect of physical exercise on uptake, distribution, and elimination of
vapors in man. In Modeling of inhalation exposure to vapors: uptake, distribution.
and elimination. Vol. 2, pp. 107-30. Edited by V. Fiserova-Bergerova. Boca Raton,
FL: Chemical Rubber Company Press.
R-3
-------
Astrand, I., Engstrom, J., and Ovrum, P. 1978. Exposure to xylene and ethylbenzene. I.
Uptake, distribution and elimination in man. Scand. J. Work Environ. Health 4:185-
94.
Austin, H., Cole, P. and McCraw, D.S. 1986. A case-control study of leukemia at a
refinery. J. Occup. Med. 28:fll):l 169-1173.
Avilova, G.G., and Ulilova, I.P. 1975. Comparative characteristics of the effect of
benzene on the reproductive function of adult and young animals. Gig. Tr. Prof.
Zabol. 2:55-57. (Russian, as cited in IARC, 1982.)
Baarson, K.A., Snyder, CA., and Albert, R.E. 1984. Repeated exposure of C57B1 mice
to inhaled benzene at 10 ppm markedly depressed erythropoietic colony formation.
Toxicol. Lett. 20:337-342.
Baker, J.M. 1976. Ecological changes in Milford Haven during its history as an oil port.
In Marine ecology and oil pollution, pp. 55-66. Edited by J.M. Baker. New York:
John Wiley & Sons.
Bakke, O., and Schelilne, R. 1970. Hydroxylation of aromatic hydrocarbons in the rat.
Toxicology and Applied Pharmacology 16:691-700.
Balogh, T., Tatrai, E., Barcza, G., Lorincz, M., and Ungvary, G. 1982. Emryo-toxic
effect of xylene. Egeszsegtudomany 26:42-48.
Barrett, J.C. and Lamb, P.W. 1985. Tests for the Syrian hamster embryo cell
transformation assay. Pros. Mutat. Res. 5:623-628. (As cited in Dean, 1985b.)
Barrett, R.H. 1985. Assays for unscheduled DNA synthesis in HeLa S3 cells. In
Evaluation of short-term tests for carcinogens: report of the international program on
chemical safety collaborative study on in vitro assays, pp. 347-352. Edited by J.
Ashby, F.J. deSerres, M. Draper, M. Ishidata, B.H. Margolin, A.B.E. Matter, and
M.D. Shelby. Amsterdam: Elsevier. (As cited in Dean, 1985b.)
Barron, V., and Divine, B. 1985. The Texaco mortality study: case-control analysis of
brain tumors in white male refinery, petrochemical and research workers. Am. J.
Epidemiol. 122:159-520.
Barry, B. W., et al. 1985. Vapour and liquid diffusion of model penetrants through
human skin. J. Pharm. Phanmacol. 37:226-35.
Banlett, S., and Tapia, F. 1966. Glue and gasoline "sniffing", the addiction of youth.
Case Report 63:270-272.
Bartsch, H., Malaveille, C, Camus, A., Martel-Planche, G., Brun, G., Nautefeuelle, A.,
Sabadie, N., Barbin, A., Kuroki, T., Drevon, C., Piccpli, C., and Montesano, R.
1980. Validation and comparative studies in 180 chemicals with S. typhimurium
strains and V79 Chinese hamster cells in the presence of various metabolizing systems.
Mutat. Res. 76:1-50. (As cited in NTP, 1986.)
Battelle. 1985. Literature evaluation of gasoline vapor toxicology. Prepared for Health
Effects Institute. Richland, MI: Battelle Pacific Northwest Laboratories.
R-4
-------
Bauchinager, M., Schmid, E., Dresp, J., Kolin-Gerresheim, J., Hauf, R., and Suhr, E.
1982. Chromosome changes in lymphocytes after occupational exposure to toluene.
Mutat. Res. 102(4):439-445.
Beime, G.J., and Brennen, J.T. 1972. Glomerulonephritis associated with hydrocarbon
solvents. Arch. Environ. Health 25:365-369.
Benville, P.E., Jr., and Kom. S. 1977. The acute toxicity of six monocyclic aromatic
crude oil components to striped bass (Morone saxatilis) and bay shrimp (Crago
franciscorum). Calif. Fish & Game 63(4):204-209.
Berenblum, I. 1941. Skin painting studies of xylene. Cancer Res. 1:44-48.
Berry, W.O., and Brammer, J.D. 1977. Toxicity of water-soluble gasoline fractions to
fourth instar larvae of the mosquito Aedes aegypti L. Environ Pollut. 13:229-234.
Berry, W.O., et ah, 1978. Uptake of water-soluble gasoline fractions and their effect on
oxygen consumption in aquatic stages of the mosquito Aedes aegypti L. Environ.
Pollut. 15:1-22.
Bio/dynamics Inc. 1980. A dominant-lethal inhalation study with benzene in rats. Report
No. 79-7342. East Millstone, NJ.: Bio/dynamics, Inc.
Bio/dynamics Inc. 1982. Parental and fetal reproduction inhalation toxicity study in rats
with mixed xylenes. FYI submission FYI-AX-0982-0209. Submitted by the
American Petroleum Institute to U.S. Environmental Protection Agency.
Black, P.D. 1967. Reports of cases. Mental illness due to the voluntary inhalation of
petrol vapor. The Medical Journal of Australia 2:70-71.
Blank, I., and MacAuIiffe, D. 1985. Penetration of benzene through human skin. L
Invest. Derm. 85:522-6.
Bond, A., Thompson, V., Ortman, G., Black, F., and Sigsby, J. 1986a. Self service
station vehicle refueling exposure study by U.S. EPA. In Proceedings of the 1986
EPA/APCA symposium on measurement of toxic air pollutants. n.p.
Bond, G.G., McLaren, E.A., Baldwin, C.L., and Cook, R.R. 1986b. An update of
mortality among chemical workers exposed to benzene. Brit. J. Ind. Med. 43:685-
691.
Bonnet, P., Morele, Y., Raoult, G., Zissu, D., and Gradiski, D. 1982. Determination of
the median lethal concentration of the main aromatic hydrocarbons in the rat Arch.
Mai. Prof. 43:261-265.
Borriston Laboratories, Inc. 1984. Four-week oral nephrotoxicitv screening study in male
F344 rats. Borriston Project No. 1706. Draft final report. Submitted to the American
Petroleum Institute, July 31,1984. Washington, DC: The American Petroleum
Institute.
1985. Serial sacrifice study with unleaded gasoline in rat. Final report.
Submitted to the American Petroleum Institute, February 1,1985.Washington, DC:
The American Petroleum Institute.
R-5
-------
Bos, R.P., Brouns, R.M.E., Van Docer, R., Theuevs, J.L.G., and Henderson, P.T.
1981. Non-mutagenicity of toluene, o-, m-, and p-xylene, o-methylbenzyalcohol and
o-methylbenzylsulfate in the Ames assay. Mutat Res. 88:273.
Bosron and Li. 1980. Alcohol dehydrogenase. In Enzymaric basis of detoxication.
Edited by W. Jacoby. New York: Academic Press.
Boyd, R., Griffiths, J., Kindt, V., Snyder, R., Caro, J., and Erslev, A. 1982. Relative
toxicity of five benzene metabolites on CFU-GM cultures. The Toxicologist 2(1): 121.
(Abstract No. 429) (As cited in NRC, 1986.)
Boye, H., Benkel, W., and Heiniger, HJ. 1970. Study of leukopenia in bone marrow of
rats after chronic benzene inhalation. (German) Blut. 21:250-257.
Braddock, J. 1988. Factors influencing the composition and quantity of passenger car
refueling emissions - Part II. SAE Paper 880712. Presented at Society of Automotive
Engineers, International Congress. March 1988, Detroit, MI.
Braddock, J., Gavele, P., and Lemmons, T. 1986. Factors influencing the composition
and quantity of passenger car refueling emissions - Part I. SAE Paper 861558.
Presented at Society of Automotive Engineers, International Fuels and Lubricants
Meeting. October 1986. Philadelphia.
Brandt, L., Nilsson, P.p., and Mitelman, F. 1978. Occupational exposure to petroleum
products in men with acute non-lymphocytic leukemia. Br. Med. J. 1:553.
Bray, H., Humphris, B., and Thorpe, W. 1949. Metabolism of derivatives of toluene and
o-, m-, and p-xylenes. Biochem. J. 45:241-244.
Brenniman, G., et al. 1976. A continuous flow bioassay method to evaluate the effects of
outboard motor exhausts and selected aromatic toxicants on fish. Water Res. 10:165-
169.
Bridie, A.L., et al. 1979. The acute toxicity of some petrochemicals to goldfish. Water
Res. 13:623-626.
Bringmann, G., and Kuhn, R. 1977. The effects of water pollutants on Daphnia magna.
Z.F. Wasser-und Abwasser Forschung 10(5): 161-166.
. 1978. Threshold values of substances harmful to water for blue algae
(microcytis aeruginosa) and green algae (Scenedesmus quadricauda) in tests measuring
the inhibition of cellular progation. Vom Wasser 50:45-60.
_. 1980. Comparison of the toxicity threshold values of water pollutants to
bacteria, algae, and protozoa in the cell multiplication inhibition test. Water Research
14:231-241.
Brown, H.S., Bishop, D.R., Rowan, C.A. 1984. The role of skin absorption as a route
of exposure for volatile organic compounds (VOCs) in drinking water. American
Journal Public Health 74:479-483.
Browning, E. 1965. Toxicitv and metabolism of industrial solvents. Isted., pp. 150-
155. Amsterdam: Elsevier Publishing Company.
R-6
-------
Brunekreef, B. 1983. Blood lead levels of Dutch city children and their relationship to
lead in the environment. Journal Air Pollut. Control Association 33:9.
Buccafusco, R.J. et al. 1981. Acute toxicity of priority pollutants to bluegill. Bull.
Environ. Contam. Toxicol. 26:446-452.
Burnett, C.J., and Kontogiannis, J.E. 1975. The effect of crude oil fractions on the
survival of a tidepool copepod, Tigriopus calif ornicus.
Busey, W.M., and Cockrcll, B.Y. 1984. Non-neoplastic exposure-related renal lesions in
rats following inhalation of unleaded gasoline vapors. In Advances in modem
environmental toxicology. Vol. VII. Renal effects of petroleum hydrocarbons, pp. 57-
64. Edited by M.A. Mehlmann, G.P. Hempstreet, J.J. Thorpe, and N.K. Weaver.
Princeton, New Jersey: Princeton Scientific Publishers, Inc.
Cairns, J., Jr. 1983. Are single species tests alone adequate for estimating hazard?
Hvdrobiologia 100:47-57.
1988. Putting the eco in ecotoxicology. Regulatory Toxicology and
Pharmacology 8:226-38.
Caldwell, R.S., Cadarone, E.M., and Mallon, M.H. 1977. Effects of a seawater-soluble
fraction of cook. In Crude oil and its maior aromatic components on larval stages of
the Dungeness crab. Cancer magister. Fate and effects of petroleum hydrocarbons in
marine ecosystems and organisms, p. 210-220. Edited by D.A. Wolfe. New York:
Pergamon Press.
California Air Resources Board (CARB). 1987. Technical support document to proposed
airborne toxic control measure for emissions of benzene from retail service stations.
Prepared by Toxic Pollutants Branch, Stationary Source Division. Sacramento, CA:
California Air Resources Board.
Carlsson, A. 1981. Distribution and elimination of ^^C"xylene in rat. Scandinavian
Journal of Work. Environment, and Health 7:51-5.
Carpenter, C.P., Kinkead, E.R., Geary, D.L., Jr., Sullivan, L.J., and King, J.M. 1975.
Hydrocarbon toxicity studies. V. Animal and human response to vapors of mixed
xylenes. Toxicol. Appl. Pharmacol. 33(3):543.
Cesaro, A. N. 1946. Is percutaneous absorption of benzene possible? Med. Lavozo
4:151-6.
Charbonneau, M., Lock, E.A., Strasser, J., Short, B.C., and Bus, J.S. 1987a.
Nephrotoxicity of 2,2,4-trimethylpentane (TMP) metabolites in male Fischer 344 rats.
The Toxicologist 7:89.
Charbonneau, M., Lock, E.A., Strasser, J., Turner, M.J., Cox, M.G., and Swenberg,
J.A.. 1987b. Identification of the 2,2,4-trimethylpentane (TMP) metabolite bound to
male rat ct-2-microglobulin.
Chekoway, H., Wilcosky, T., Wolf, P., and Tyroler, H.. 1984. An evaluation of the
associations of leukemia and rubber industry solvent exposures. Amer. J. Ind. Med.
5:329-349.
R-7
-------
Chemical Engineering. 1982. Is there a health risk at Exxon refineries? Oct. 4,1982. p.
53.
Chemical Industry Institute of Toxicology (CHT). 1980. A twenty-four month inhalation
toxicology study in Fischer-344 rats exposed to atmospheric toluene. Executive
Summary and Data Tables. Conducted by Industrial Bio-Test Laboratories, Inc.,
Decatur, EL, and Experimental Pathology Laboratories, Inc., Raleigh, NC, October 15,
1980. Research Triangle Park, NC: CIIT.
Chemical Rubber Company (CRC). 1967. Handbook of chemistry and phvsics. 48th ed.
Cleveland: The Chemical Rubber Company.
.. 1968. Handbook of chemistry and phvsics. 49th ed. Cleveland: The
Chemical Rubber Company.
_. 1987. Handbook of chemistry and phvsics. 67thed. Cleveland: The
Chemical Rubber Company.
Chiang, C.Y., Chai, E.Y., Salanitro, J.P., Colthart, and Klein, C.L. 1987. Effects of
dissolved oxygen on the biodegradation of BTX in a sandy aquifer. Houston, TX:
Shell Development Co.
Choy, W.N., MacGregor, J.T., Shelby, M.D., and Maronpot, R.R. 1985. Induction of
micronuclei by benzene in B6C3F1 mice: retrospective analysis of peripheral blood
smears from the NTP carcinogenesis bioassay. Mutat. Res. 143:55-59.
Christie, D., Robinson, K., Gordon, I., and Rockett, I. 1984. Health watch: the
Australian petroleum industry health study surveillance programme. Med. J. Aus.
141:331-334.
Clare, M.G., Yardley-Jones, A., McLean, A.C., and Dean, B.J. 1984. Chromosome
analysis from peripheral blood lymphocytes of workers after acute exposure to
benzene. Br. J. Ind. Med. 41:249-253. (As cited in Dean, 1985b.)
Clayson, D.B., et al. 1985. Toxicological risk assessment. Vol. 1. Boca Raton, FL:
Chemical Rubber Company Press.
Clayton Environmental Consultants (CEC). 1983. Gasoline exposure study for the
American Petroleum Institute, August 25,1983. Submitted by N.K. Weaver to The
American Petroleum Institute, September 19,1983. FYI-AX-0983-0148 Supp.
Sequence L. Southfield, MI: Clayton Environmental Consultants.
Coates, M., Cpnnel, D.W., and Brown, D.M. 1985. Aqueous solubility and octanol
water partition coefficients of aliphatic hydrocarbons. Environ. Sci. Technol.
19(7):628-632.
Cohr, K. H., and Stockholm, J. 1979. Toluene: a lexicological review. Scand. J. Work
Environ. Health 5:71-90.
Cole, P. 1985. University of Alabama review of "Lifetime risks of kidney cancer
estimated from animal and epidemiologic studies" by P.E. Enterline. Washington, DC:
American Petroleum Institute.
R-8
-------
Cole, P., Hoover, R., and Friedell, G.H. 1972. Occupation and cancer of the lower
urinary tract. Cancer 29:1250-1260.
Coleman, W.E., Munch, J.W., Streicher, R.P., Ringhand, H.P., and Kepfler, F.C.
1984. The identification and measurement of components in gasoline, kerosene, and
No. 2 fuel oil that partition into the aqueous phase after mixing. Arch. Environ.
Contam. Toxicol. 13:171-178.
Conca, G. L., and Maltagliati, A. 1955. Study of the percutaneous absorption of
benzene. Med. Lavoro 46:194-8.
Connecticut Department of Environmental Protection. 1989. Research on gasoline-
contaminated water supply wells. February 1989. Hartford: Connecticut Department
of Environmental Protection, Water Compliance Unit
Contreras, C.M., Gonzalez-Estrada, T., Zarabozo, D., and Femandez-Guardiola, A.
1979. Petit mal and grand mal seizures produced by toluene or benzene intoxication in
the cat. Electroencephalogr. Clin. Neurlphvsiol. 46:290-301.
Coombs, M.M., and Croft, C.J. 1966. Carcinogenic derivatives of
cyclopenta(a)phenanthrene. Nature 210:1281-1282.
Cooper etal. 1982. Environmental quality research - Fate of toxic iet fuel components in
aquatic systems. Air Force Aerospace Medical Research Laboratory Report No.
AFAMRL-TR-82-64. Irvine, CA: University of California at Irvine.
Cotruvo, J., Simmon, V., and Spanggord, R. 1977. Investigation of mutagenic effects of
products of ozonation reactions in water. Ann. NY Acad. Sci. 298:124-140. (As cited
in NTP, 1986.)
Courtney, K.D., Andrews, J.E., Springer, J., Menache, M., Williams, T., Dalley, L.,
and Graham, J.A. 1986. A perinatal study of toluene in CD-I mice. Fundamental and
Applied Toxicology 6:145-154.
Crapp, G.B. 1971. The ecological effects of stranded oil. In The ecological effects of oil
pollution in littoral communities, pp. 181-186. Edited by E.B. Cowell. London:
Institute of Petroleum.
Crebilli, R., Bellincampi, D., Conti, G., Conti, L., Murpurgo, G., and Carere, A. 1986.
A comparative study on selected chemical carcinogens for chromosome malsegregation,
mitotic crossing-over and forward mutation induction in Aspergillus midulans.
Mutation Res. 172:139-149.
Cronkite, E., Drew, R., Inoue, T., and Bullis, J. 1985. Benzene hematotoxicity and
leukemogenesis. Amer. J. Ind. Med. 7:447-456.
Cronkite, E.P. 1986. Benzene hematotoxicity and leukemogenesis. Blood Cells 12:129-
137.
Cronkite, E.P., Inoue, T., Carsten, A.L., Miller, M.E., Bullis, I.E., and Drew, R.T.
1982. Effects of benzene inhalation on marine pluripotent stem cells. J. Toxicol. Env.
Health 9:411-421.
R-9
-------
Cupitt, L.T. 1980. Fate of toxic and hazardous materials in the air environment Research
Triangle Park, NC: U.S. Environmental Protection Agency, Environmental Sciences
Research Laboratory.
Danford, N.D. 1985. Tests for chromosome aberrations and anueploidy in the Chinese
hamster fibrpblast cell line CH1-L. In Evaluation of short-term tests for carcinogens:
report of the international program on chemical safety collaborative study on in vitro
assays, pp. 397-411. Edited by J. Ashby, F.J. fie Series, M. Draper, M. Ishidate, Jr.,
B.H. Margolin, B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier Publishing
Company. (As cited in Dean, 1985b.)
Davis, A., Schafer, L.J., and Bell, Z.G. 1960. The effects on human volunteers of
exposure to air containing gasoline vapor. Arch. Environ. Health 1:548-554.
Dean, B. 1969. Chemical-induced chromosome damage. Lab Anim. 3:157-174.
Dean, B. 1985a. Summary report on the performance of cytogenetic assays in cultured
mammalian cells. In Evaluation of short-term tests for carcinogens: report of the
international program on chemical safety collaborative study on in vitro assays, pp. 69-
83. Edited by J. Ashby, F.J. de Serres, M. Draper, M. Ishidate, Jr., B.H. Margolin,
B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier. (As cited in Dean, 1985b.)
Dean, B. 1985b. Recent findings on the genetic toxicology of benzene, toluene, xylenes,
and phenols. Mutat. Res. 154:153-181.
DeCoufle, P., Blattner, W.A., and Blair, A. 1983. Mortality among chemical workers
exposed to benzene and other agents. Environ. Res. 30:16-25.
DeGowin, R.L. 1963. Benzene exposure and aplastic anemia followed by leukemia 15
years later. J. Amer. Med. Assoc. 185:748-751.
DeGraeve, G.M., et al. 1980. Effects of naphthalene and benzene on fathead minnows
and rainbow trout. Trans. Amer. Fish. Soc. (submitted).
Deichmann, W.B., MacDonald, W.E., and Bemal, E. 1963. The hemopoietic tissue
toxicity of benzene vapors. Toxicol. ADD!. Pharmacol. 5:201-224.
Dell-Acqua, R., Bush, B., and Egan, J. 1976. Identification of gasoline contamination of
groundwater by gas chromatography. J. Chromatographv 128:271-280.
Delore, P., and Borgomano, C. 1928. Acute leukemia following benzene poisoning on
the toxic origin of certain acute leukemias and their relation to serious anemias.
(French) J. Med. Lvon 9:227-233.
Dempster, A.M., Evans, H.L., and Snyder, C.A. 1984. The temporal relationship
between behavioral and hematological effects of inhaled benzene. Toxicology and
Applied Pharmacology 76:195-203.
DGMK. 1985. Exposure data on C7 and C8 aromatics during handling and production of
motor gasoline. Project 250-1. German Society for Petroleum Sciences and Coal
Chemistry. Washington, DC: American Petroleum Institute.
R-10
-------
Diakun, R. 1983. Characterization of workplace exposures. Documentation of
presentation to EPA by API, submitted by N.K. Weaver, American Petroleum
Institute, July 15,1983. FYI-AX-0783-0148. Supp. Sequence H.
Diaz, M., Reiser, A., Braier, L., and Diez, J. 1980. Studies on benzene mutagenesis. I.
The micronucleus test. Exoerienria 35:297-299. (As cited in NTP, 1986.)
Diem, K., and Lentner, C. 1973. Documenta Geigv. Scientific Tables, Ciba-Geigy Ltd.
DiGuglielmo, G., and lannacione, A. 1958. Inhibition of mitosis and regressive changes
of erythroblasts in acute erythropathy caused by occupational benzene poisoning. Acta
Haematol. 19:144-147.
DiVencenzo, G.D., and Krasavage, W.J. 1974. Serum ornithine carbamyl transferase as
a liver response test for exposure to organic solvents. Am. Ind. Hvp. Assoc. J. 35:21-
29.
Divine, B.J., Barren, V., and Kaplan, S.D. 1985. Texaco mortality study. I. Mortality
among refinery, petrochemical, and research workers. J. Occup. Med. 27:445-447.
Dobashi, Y. 1974. Effects of benzene and its metabolites on the mitosis of cultures of
human cells. Sangvo Igaku 16:453.
Dobrokhotov, V.B. 1972. The mutagenic influence of benzene and toluene under
experimental conditions. Gig. Sanit. 37:36-39. (In Russian; evaluation based on an
English translation provided by the U.S. EPA.)
Dobrokhotov, V.B., and Enikeev, M.I. 1975. Mutagenic effect of benzene, toluene, and
a mixture of these hydrocarbons in a chronic experiment Gig. Sanit. 1:32-34. (In
Russian with English summary; evaluation based on an English translation provided by
the U.S. EPA.)
Domask, W.G. 1984. Introduction to petroleum hydrocarbons. Chemistry and
composition in relation to petroleum-derived fuels and solvents. Chapter 1 in Advances
in modern environmental toxicology. Vol. VII. Renal effects of petroleum
hydrocarbons, pp. 1-23. Edited by M.A. Mehlman, G.P. Hemstreet, J.J. Thorpe, and
N.K. Weaver. Princeton, NJ: Princeton Scientific Publishers, Inc. (As cited in EPA,
1985.)
Domenico, P.A., and Palciauskas, V.V. 1982. Alternative boundaries in solid waste
management. Groundwater 20(3) :303-311.
Domiano, S.F., Vena, J.E., and Swanson, J.K. 1985. Gasoline exposure, smoking, and
kidney cancer. J. OCCUD. Med. 27:389-399.
Donner, M., Maki-Paakkanen, J., Norppa, H., Sorsa, M., and Vainio, H. 1980. Genetic
toxicology of xylenes. Mutat. Res. 74:171-172.
Dooley, J.F., Skinner, M.J., Roy, T.A., Blackbuar, G.R., Schreiner, C.A., and
Mackerer, C.R. 1987. Evaluation of the genotoxicity of API reference unleaded
gasoline. In Polvnuclear aromatic hydrocarbons. Proceedings of the tenth international
symposium, n.p.: Battelle Press, (in press)
R-ll
-------
Doskin, T.A. 1971. Effect of age on the reaction to a combination of hydrocarbons.
Hvg. Sanit. 30:379-384.
Dourson, M., and Stara, J. 1983. Regulatory history and experimental support of
uncertainty (safety) factors. Regulatory Toxicology and Pharmacology 3:224-238.
Dowd, R.M. 1984. Leaking underground storage tanks. Environ. Sci. Technol.
18.-309A.
Dowty, B.J., Laseter, J.L., and Storer, J. 1976. The transplacental migration and
accumulation in blood of volatile organic constituents. Pediatr. Res. 10:696-701.
Drinker, P., Yaglou, C.P., and Warren, M.F. 1943. The threshold toxicity of gasoline
vapor. Journal of Industrial Hygiene and Toxicology 25:225-232.
Drozd, J., and Bockowski, E.J. 1967. Acute benzene poisoning. J. OCCUD. Med. 9(1 ):9-
1 1 •
Dubois, J.P. 1961. Three fatal cases of benzene poisoning. Rev. Med. Suisse Romande
12:851-856.
Duggan, M., and Williams, S. 1977. Lead dust in city streets. In Science of the total
environment. Amsterdam: Elsevier Publishing.
Dunstan, W.M., et al. 1975. Stimulation and inhibition of phytoplankton growth by low
molecular with hydrocarbons. Marine Biology 31:305-310.
Dworzanski, J., and Debowski, M. 1981. Semiautomated preparation technique of urine
samples for gas chromatographic determination of toluene and xylene metabolites.
Chemia Anal. 26:319-25.
Dynamac Corporation. 1986. A review of the epidemiologic evidence for a causal
association between unleaded gasoline and cancer risk. Report submitted to the U.S.
Environmental Protection Agency, April 18,1986. Rockville, MD: Dynamac
Corporation.
Egilsson, V., Evans, I., and Wilkie, D. 1979. Toxic and mutagenic effects of carcinogens
on the mitochondria of Saccaromyces cerevisiae. Mol Gen"Genet 174:39-46. (As
cited in NTP, 1986.)
Elars Bioresearch Laboratories. 1982. Acute toxicity tests API #PS-6 unleaded motor
gasoline. Sample 81-13. Prepared for The American Petroleum Institute. Fort Collins,
CO: Elars Bioresearch Laboratories, Inc.
Ellis and Obendorfer. 1984. Survey of benzene concentrations in ambient air.
Philadelphia: Philadelphia Department of Health, Air Management Services.
Elofsson, S.-A., Gamberale, F., Hindmarsh, T., Iregren, A., Isaksson, A., Johnsson, I.,
Knave, B., Lydahl, E., Mindus, P., Persson, H.E., Philipson, BM Steby, M.,
Struwe, G., Soderman, E., Wennberg, A., and Widen, L. 1980. Exposure to organic
solvents: A cross-sectional epidemiologic investigation on occupationally exposed car
and industrial spray painters with special reference to the nervous system. Scand. J.
Work Environ. Health 6:239-273.
R-12
-------
Elovaara, £., Engstrom, K., and Vainio, H. 1984. Metabolism and disposition of
simultaneously inhaled m-xylene and ethylbenzene in the rat. Toxicology and Applied
Pharmacology 75:466-478.
Engstrom, J., and Bjurstrom, R. 1978. Exposure to xylene and ethylbenzene. II.
Concentration in subcutaneous adipose tissue. Scand. J. Work Environ. Health 4:195-
203.
Engstrom, J., Husman, J.L., and Storer, J. 1979. Evaluation of occupational exposure to
xylene by blood, exhaled air and urine analysis. Arh. Hie. Rada Toksikol. 30
(Suppl.):483-438.
Engstrom, J., Husman, K., and Riihimaki, V. 1977. Percutaneous absorption of m-
xyleneinman. Int. Arch. OCCUP. Environ. Health 39:181-189.
Engstrom, J., Riihimaki, V., and Laine, A. 1984. Urinary disposition of ethylbenzene
and m-xylene in man following separate and combined exposure. Int. Arch. Occup.
Environ. Health 54:355-363.
Enterline, P. 1978. Internal report: Mortality among workers from Shell Oil Company's
lubricants sewaging unit at Deer Park, Texas, n.p., n.p.
. 1984. Lifetime risks of kidnev cancer caused by gasoline estimated from
animal and human studies. Prepared for the Motor Vehicle Manufacturers' Association
of the United States.
Enterline, P.E., and Henderson, V. 1985 . A report on the mortality experience of
workers from a petroleum refinery at Torrance. California. Prepared for Mobil Oil
Corporation. Pittsburgh, PA: University of Pittsburgh. (Unpublished)
Enterline, P.E., and Viren, J. 1985. Epidemiologic evidence for an association between
gasoline line and kidney cancer. Environ. Health Persp. 62:302-312.
Eppley Institute for Research in Cancer. 1983. Carcinogenic potential of key petroleum
products. Final report. Submitted to the American Petroleum Institute, May 10,1983.
Erexson, G.L., Wilmer, J.L., and Kligerman, A.D. 1985. Sister chromatid exchange
induction in human lymphocytes exposed to benzene and its metabolites in vitro.
Cancer Res. 45:2471-2477.
Erexson, G.L., Wilmer, J.L., Auman, C.D., and Kligerman, A.D. 1984. Induction of
sister chromatic exchanges and micronuclei in male DBA/2 mice after inhalation of
benzene. Environ. Mutagen. 6:406.
Erexson, G.L., Wilmer, J.L., Steinhagen, W.H., and Kligerman, A.D. 1986. Induction
of cytogenic damage in rodents after short-term inhalation of benzene. Environ. Muta.
8:20-40.
Eschenroader, A., Jaeger, R., Ospital, J., and Doyle, C. 1986. Health risk analysis of
human exposure to soil amended with sewage sludge contaminated with
polychlorinated diberzodioxins and dibenzofurans. Vet. Hum. Toxicol. 28(5):435-
442.
R-13
-------
Evans, E.L., and Mitchell, A.D. 1980. An evaluation of the effect of toluene on sister
chromatid exchange frequencies in cultured Chinese hamster ovary cells. Prepared by
SRI International, Menlo Park, CA, for the U.S. Environmental Protection Agency.
Research Triangle Park, NC: U.S. Environmental Protection Agency.
Evans, H.L., Dempster, A.M., and Snyder, C.A. 1981. Behavioral changes in mice
following benzene exposure. Neurobehavioral Toxicol. Teratol. 3:481-485.
Experimental Pathology Laboratories, Inc. (EPL). 1983. Review of kidnev and livers
from female mice used in a two-year motor fuel chronic inhalation study. Pathology
report submitted to the American Petroleum Institute, April 19,1983.
Fabre, R., et al. 1955. Recherches toxicolagiques sur les solvents de remplacement due
benzended. Archives Maladies Professionalles de Medicine du Travail et du Securite
Sociale 16:197-215. (As cited in Bergman, 1979.)
Farino, W., Spawn, P., Jaskinski, M., and Murphy, B. 1983. Evaluation and selection
of models for estimating air emissions from hazardous waste treatment, storage and
disposal facilities. Revised draft final report. Prepared for U.S. Environmental
Protection Agency, Office of Solid Waste. Land Disposal Branch. Bedford, MA:
GCA Corporation, CCA/Technology Division.
Federal Register. 1987. Preamble to final OSHA rule on occupation exposure to benzene.
52 FR 34460, September 11,1987. Occupational Safety and Health Reporter 17:619-
736.
Feliciano, D.V. 1984. Leaking underground storage tanks. A potential environmental
problem. Congressional Research Service, Report No. 84-508ENR. Washington,
DC: Library of Congress.
Feller, I. 1972. Gonadotropic and mutagenic action of grade BR-1 (Galosha) gasoline.
Gig. Tr. Prof. Zabol. 16:25-28.
Fielder, A.C., Katz, S., and Kinney, S.P. 1921. Prevention of oxygen breathing
apparatus bv gases and vapors. Technical Paper 272. U.S. Department of Commerce,
Bureau of Mines. (As cited in Machle, 1941.)
Fielder, R.J. 1982. Toxicitv review 4. Benzene. London: Health and Safety Executive.
Fiserova-Bergerova, V., and Hughes, H.C. 1983. Species differences on bioavailability
of inhaled vapor and gases. In Modeling of inhalation exposure to vapors: uptake.
distribution, and elimination. Vol. 2, pp. 97-106. Edited by V. Fiserova-Bergterova.
Boca Raton, FL: CRC Press.
Fishbein, L. 1985. An overview of environmental and lexicological aspects of aromatic
hydrocarbons. II. Toluene. Science of the Total Environment 42:267-288.
Florida Petroleum Council/ American Petroleum Council (FPC/API). 1986. Benzene in
Florida groundwater. An assessment of the significance to human health, n.p.: Florida
Petroleum Council, A Division of the American Petroleum Institute.
Florin, I., Rutberg, K., Curvall, M., and Enzell, C. 1980. Screening of tobacco smoke
constituents for mutagenicity using the Ames test Toxicology 15:219-232.
R-14
-------
Fluck, E.R., et al. 1976. Evaluation of a DNA polymerase-deficient mutant of E. coli for
the rapid detection of carcinogens. Chem. Biol. Inter. 15:219.
Food and Agriculture Organization of the United Nations. (FAO). 1977. Food and
agriculture organization of the United Nations. Impact of oil on the marine
environment, p. 250. IMCO/FAO/UNESCO/WMO/WHO/IAEA/UN Joint Group of
Experts on the Scientific Aspects of Marine Pollution (GESAMP) Rep. Stud.
GESAMP (6).
Forni, A., and Moreo, L. 1969. Chromosome studies in a case of benzene induced
erythroleukemia. Eur. J. Cancer 5:459-463.
Fomi, A., Pacifico, E., and Limonta, A. 1971. Chromosome studies in workers exposed
to benzene or toluene or both. Arch. Environ. Health 22(3):373-378.
Franchini, L, Cavatorta, A., Falzoi, M., Lucertini, S., and Mutti, A. 1983. Early
indicators of renal damage in workers exposed to organic solvents. Inter. Arch.
OCCUP. Environ. Health 52:1 -9.
Fredga, K., Davring, L., Sunner, M., Bengtsson, B.O., Elinder, C.G., Sigtryggsson, P.,
and Berlin, M. 1982. Chromosome changes in workers (smokers and nonsmokers)
exposed to automobile fuels and exhaust gases. Scand. J. Environ. Health 8:209-221.
Fredga, K., Reitalu, J., and Berlin, M. 1979. Chromosome studies in workers exposed
to benzene. In Genetic damage in man caused by environmental agents. Vol. 1, pp.
187-203. Edited by E.K. Berg. New York: Academic Press. (As cited in Dean,
1985b.)
Frei, J., and Kingsley, W.F. 1968. Observations on chemically induced regressing
tumors of mouse epidermis. J. National Cancer Inst. 41:322-325.
Frei, J.V., and Stephens, P. 1968. The correlation of promotion of tumor growth and of
induction of hyperplasia in epidermal two-stage carcinogenesis. Br. J. Cancer 22:83-
92.
Fribroska, A. 1973. Some cytochemical findings in the peripheral white blood cells in
workers exposed to toluene. Folia Haematol. (Leipzig) 99:233-237.
Fujikawa, K., Ryo, H., and Kondo, S. 1985. The Drosophila gene mutation and small
deletion assay using the zeste-white somatic eye colour system. In Evaluation of short-
term tests for carcinogens: report of the international program on chemical safety
collaborative study on in vitro assays, pp. 319-324. Edited by J. Ashby, F.J. de
Serres, M. Draper, M. Ishidate, Jr., B.H. Margolin, B.E. Matter, and M.D. Shelby.
Amsterdam: Elsevier.
Funes-Craviota, F., et al. 1977. Chromosome aberrations and sister-chromatic exchange
in workers in chemical laboratories and a rotoprinting factory and in children of women
laboratory workers. Lancet 2:322.
Furey, R.L., and Nagel, B.E. 1986. Composition of vapor emitted from a vehicle
gasoline tank during refueling. SAE Paper 860086. Presented at the Society of
Automotive Engineers Inc., International Congress. February 1986. Detroit, MI.
R-15
-------
Gad-el Karim, M.M., Harper, B.L., and Legator, M.S. 1984. Modifications in the
myeloclastogenic effect of benzene in mice with toluene, phenobarbitol, 3-
methylcholanthrene, Aroclor 1254 and Skf-525A. Mutation Res. 135:225-243. (As
cited in Dean, 1985b.)
Gad-el Karim, M.M., Ramanujam, V.M., Ahmed, A.E., and Legator, M.S. 1985.
Benzene myeloclastogenicity: a function of its metabolism. Amer. J. Ind. Med. 7:475-
484.
Gad-el Karim, M.M., Ramanujam, V.M., and Legator, M.S. 1986. Correlation between
the induction of micronuclei in bone marrow by benzene exposure and the excretion of
metabolites in urine of CD-I mice. Toxicol. Apnl. Pharmacol. 85:464-477.
Galavotti, B., and Troisi, P.M. 1950. Erythro-leukaemia myelosis in benzene poisoning.
Br. J. Ind. Med. 7:79-81.
Gallahan, M.A., Johnson, R.H., McGinnity, J.L., et al. 1983. Handbook for performing
exposure assessment. Draft Washington, DC: U.S. Environmental Protection
Agency, Office of Health and Environmental Assessment
Gammage, R.B., White, D.A., and Gupta, K.C. 1984. Residential measurements of high
volatility organics and their sources. In Indoor air, proceedings of the international
conference on indoor air quality climate, pp. 157-62. 34d; Vol. 4, Iss. PB85-104214.
Gamer, R.C. 1985. Summary report on the performance of gene mutation assays in
mammalian cells in culture. In Evaluation of short-term tests for carcinogens: report of
the international proeram on chemical safety collaborative study on in vitro assays, pp.
85-94. Edited by J. Ashby, F.J. de Serres, M. Draper, M. Ishidate, Jr., B.H.
Margolin, B.E. Matter, and M.D. Shelby. Amsterdan: Elselvier. (As cited in Dean,
1985b.)
Gates, D.M. 1985. Energy and ecology. Sunderland, MA: Sinauer Associates, Inc.
Gerarde, H.W. 1960. Toxicology and biochemistry of aromatic hydrocarbons. New
York: Elsevier Publishing Company. (As cited in NRC, 1986.)
Gerarde, H.W. 1963. Toxicology studies on hydrocarbons. IX. The aspiration hazard
and toxicity of hydrocarbons and hydrocarbon mixtures. Arch. Environ. Health 6:329-
341.
Gerner-Smidt, P., and Friedrich, U. 1978. The mutagenic effect of benzene, toluene and
xylene studies by the SCE technique. Mutat. Res. S8f2-3^: 313-316. (As cited in
NTP, 1986.)
Gifford, F.A. 1961. Uses of routine meteorological observations for estimating
atmospheric dispersion. Nuclear Safety 2.4:47-51.
Gill, D.P. and Ahmed, A.E. 1981. Covalent binding of 14C-benzene to cellular
organelles and bone marrow nucleic acids. Biochem. Pharmacol. 30:1127-1132.
Gill, D.P., Jenkins, V.K., Kempen, R.R., and Ellis, S. 1980. The importance of
pluripotential stem cells in benzene toxicity. Toxicology 16:163-171.
R-16
-------
Glauert, H.P., Kennan, W.S., Sattler, G.S., and Pitot, H.C. 1985. Assays to measure
the induction of unscheduled DNA synthesis in cultured hepatocytes. In Evaluation of
short-term tests for carcinogens: report of the international program on chemical safety
collaborative study on in vitro assays, pp. 371-373. Edited by J. Ashby, F.J. de
Serres, M. Draper, M. Ishidate, Jr., B.H. Margolin, B. E. Matter, and M.D. Shelby.
Amsterdam: Elsevier. (As cited in Dean, 1985b.)
Gleason, M.N., Gosselin, R.E., Hodge, H.C., and Smith, R.P. 1969. Clinical
toxicology of commercial products. 3rd ed. Baltimore, MD: Williams and Wilkins
Company.
Gofmekler, V.A. 1968. Effect on embryonic development of benzene and formaldehyde
in inhalation experiments. Hvg. Sanit. 33:327-332. (As cited in Fielder, 1982.)
Goguel, A. Cavigneaux, A., and Bernard, J. 1967. Benzene leukemias. (French) Bull
Inst. Nat'l. Sante Rich. Med. 22:421-441.
Goldie, I. 1960. Can xylene (xylol) provoke convulsive seizures? Ind. Med. Sure. 29:33-
35.
Goldstein, B.D. 1977. Hemototoxicity in humans. J. Toxicol. Env. Health 69:105.
Supplement 2.
Goldstein, B.D. 1985. Risk assessment and risk management of benzene by the EPA. In
Risk quantification and regulatory policy. Edited by D.G. Hall, R.A. Merill, and F.P.
Perere. Banbery Report, n.p.: n.p.
Gosta, A., Lunz, C., and Rylander, R. 1984. Exposure to solvents and outcome of
pregnancy in university laboratory employees. Brit. J. Ind. Med. 41:305-312.
Gottlieb, M.S. 1980. Lung cancer and the petroleum industry in Louisiana. J. Occup.
Med. 22:384-388.
Gottlieb, M.S., and Carr, J.K. 1981. Mortality studies on lung, pancreas, esophageal,
and other cancers in Louisiana. In Quantification of occupational cancer, pp. 195-204.
Edited by R. Peto and M. Schneiderman. Branbury Report 9. Cold Spring Harbor,
NY: Cold Spring Harbor Laboratory.
Gottlieb, M.S., and Pickle, L.W. 1981. Bladder cancer mortality in Louisiana. J. La.
State Med. Soc. 133:6-9.
Gottlieb, M.S., Pickle, L.W., Blot, W.J., and Fraumeni, J.F., Jr. 1979. Lung cancer in
Louisiana: Death certificate analysis. J. Natl. Cancer Inst. 63:1131-1137.
Green, D. 1982. Gasoline use in the United States. In U.S. Department of Energy.
Petroleum Supply Monthly. May 1982. Washington, DC: Energy Information
Administration.
Green, J.D., Leong, B.K.J., and Laskin, S. 1978. Inhaled benzene fetotoxicity in rats.
Toxicol. Appl. Pharmacol. 46:9-18.
Green, J.D., Snyder, C.A., Lobue, J., Goldstein, B.D., and Albert, R.E. 1981. Acute
and chronic dose/response effect of benzene inhalation on the peripheral blood, bone
marrow, and spleen cells of CD-I male mice. Toxicol. Appl. Pharmacol. 59:204-214.
R-17
-------
Green, M.H., et al. 1979. Cancer mortality among printing plant workers. Envir. Res.
20:66-73.
Groundwater Technology Inc. 1988. Personal communication regarding air monitoring
data.
Guyton, A.C. 1947. Measurement of the respiratory volumes of laboratory animals.
Amer. J. Phvsiol. 150:70-77.
Haggard, H.W. 1921. The anesthetic and convulsant effects of gasoline vapor. L
Pharm. EXD. Ther. 16:401-404.
Haglund, U., Lundberg, I., and Zech, L. 1980. Chromosome aberrations and sister
chromarid exchanges to Swedish paint industry workers. Scand. J. Work Environ.
Heahh 6(4):291-298.
Haider, C., Gorp, G.N., Hatoum, N.S., and Wame, T. 1986a. Gasoline vapor
exposures. Pan I. Characterization of workplace exposures. Am. Ind. Hvg. Assoc.
L47: 164-172.
Haider, C.A., Holdsworth, B.Y. Cpckrell. and Piccirillo, V.J. 1985. Hydrocarbon
nephropathy in male rats: identification of the nephrotoxic components of unleaded
gasoline. Toxicology and Industrial Health 1:67-87.
Haider, C.A., Van Gorp, G.S., Hatoum, N.S., and Warne, T.M. 1986b. Gasoline vapor
exposures. Part n. Evaluation of the nephrotoxicity of the major C4/C5 hydrocarbon
components. Am. Ind. Hvg. Assoc. J. 47:173-175.
Haider, C.A., Warne, T.M., and Hatoum, N.S. 1984. Renal toxicity of gasoline and
related petroleum naphthas in male rats. In Advances in modem environmental
toxicology. Volume VII. Renal effects of petroleum hydrocarbons, p. 73-88. Edited
by M.A. Mehlman, C.P. Hemstreet, J.J. Thorpe, and N.K. Weaver. Princeton:
Princeton Scientific Publishers, Inc.
Hanis, N.M., Holmes, T.M, Shallenberger, L.G., and Jones, K.E. 1982. Epidemiologic
study of refinery and chemical plant workers. J. Occup. Med. 24:203-212.
Hanis, N.M., Shallenberger, L.G., Donaleski, D.L., and Sales, E.A. 1985a. A
retrospective mortality study of workers in three major U.S. refineries and chemical
plants. Parti. Comparisons with U.S. population. J. Occup. Med. 21:283-292.
Hanis, N.M., Shallenberger, L.G., Donaleski, D.L., and Sales, E.A. 1985b. A
retrospective mortality study of workers in three major U.S. refineries and chemical
plants. Part n. Internal comparison by geographic site, occupation and smoking
history. J. Occup. Med. 27:361-369.
Hanis, N.M., Stavraky, K.M., and Fowler, J.L. 1979. Cancer mortality in oil refinery
workers. J. Occup. Med. 21:167-174.
Hanninen, H., Eskelinen, L., Husman, K., and Nurminen, M. 1976. Behavioral effects
of long-term exposure to a mixture of organic solvents. Scand. J. Work Environ. &
Health 2:240-255.
R-18
-------
Hansen, K.S. and Sharp, F.R. 1978. Gasoline sniffing. Lead poisoning and myoclonus.
J. Am. Med. Assoc. 240:1375-1376.
Hartle, R. 1980. Occupational exposure to benzene at automotive service stations.
M.S.P.H. thesis, University of Washington, 1980.
Health Effects Institute (HEI). 1985. Gasoline vapor exposure and human cancer:
Evaluation of existing scientific information and recommendations for future research.
September 1985. Supplement, January 1988. Cambridge, MA: Health Effects
Institute, Health Review Committee.
Heinonen, T., et al. 1983. Effect of various solvents on the xenobiotic biotransformation
in the liver and the kidneys of the rat: a comparative study. In Extrahepatic drug
metabolism and chemical carcinogenesis. pp. 29-31. Edited by J. Rydstrom, J.
Montelius, and M. Bengsson. New York: Elsevier Science Publishers.
Heitmuller, P.T., et al. 1981. Acute toxicity of 54 industrial chemicals to sheepshead
minnows. Bull. Environ. Contam. Toxicol. 27(5):596-604.
Hendry, D.G. and Kenley, R.A. 1979. Atmospheric reaction products of organic
compounds, n.p.: U.S. Environmental Protection Agency.
Hermens, J., et al. 1984. Quantitative structure activity relationships and toxicity studies
of mixtures of chemicals with anaesthetic potency: acute lethal and sublethal toxicity to
Daphnia magna. Aquatic Toxicol. 5:143-154.
Higginson, J., Muir, C, and Buffein, P.A. 1984. The epidemiology of renal carcinoma
in humans with a note on the effect of exposure to gasoline. In Renal effects of
petroleum hydrocarbons. Advances in modem environmental toxicology. Edited by
M. Mehlman et al. Princeton, NJ: Princeton Scientific Publishers.
Hinchee, R.E., and Reisinger, J.J. 1987. A practical application of multiphase transport
theory to ground-water contamination problems. GWMR, pp. 84-92.
Hoag, G., Bruell, C.J., and Marley, M.C. 1985. Study of the mechanisms controlling
gasoline hydrocarbon partitioning and transport in groundwater systems. Govt.
Reports Announcements & Index (GRA&I), Issue 26,1985.
Hochhauser, A. and Campion, R. 1976. An experimental study of vehicle refueling
emissions. SAE Paper 760307. In Society of Automotive Engineers, Automotive
Engineering Congress. February 1976. Detroit, Michigan.
Holmberg, P.C. 1979. Central-nervous-system defects in children born to mothers
exposed to organic solvents during pregnancy. Lancet 2:177-179.
Horiguchi, K., Horiguchi, S., and Morioka, S. 1967. Maximum allowable concentration
of benzene in an animal experiment. Osaka Citv Med. J. 13:1-8.
Horiguchi, S., and Inoue, K. 1977. Effects of toluene on the wheel-turning activity and
peripheral blood findings in mice - an approach to the maximum allowable
concentration of toluene. J. Toxicol. Sci. 2(4):363-372.
R-19
-------
Horowitz, A., and Adas, R.M. 1977. Response of microorganisms to an accidental
gasoline spillage in an artic freshwater ecosystem. Applied and Environmental
Microbiology 33(6): 1252-1258.
Howe, G.R., Burch, J.D., Miller, A.B., Cook, G.M., Esteve, J., Morrison, B., Gordon,
P., Chambers, L.W., Fodor, G., and Winsor, G.M. 1980. Tobacco use, occupation,
coffee, various nutrients, and bladder cancer. J. Natl. Cancer Inst. 64:701-713.
Hsieh, G.C, Parker, R.D.R., and Sharma, R.P. 1988. Subclinical effects of
groundwater contaminants. II. Alternation of regional brain monoamine
neurotransmitters by benzene in CD-I mice. Arch. Environ. Contain. Toxicol. 17:799-
805.
Hudak, A., and Ungvary, G. 1978. Embryotoxic effects of benzene and its methyl
derivatives: toluene, xylene. Toxicology 11:55-63.
Hunter, A.G. Thompson, D., and Evans, J.A. 1979. Is there a fatal gasoline syndrome?
Teratology 20:75-80.
Hunter, C.G. 1968. Solvents with reference to studies on the pharmacodynamics of
benzene. Proc. R. Soc. Med. 61:913-915.
Hutchinson, T.C., Hellebust, J.A, Tarn, D., Mackay, D., Mascarenhas , R.A., and Shiu,
W. Y. 1980. The correlation of the toxicity to algae of hydrocarbons and halogenated
hydrocarbons with physical-chemical properties. Environ. Sci. Res. 16:577-586.
ICF, Inc. 1987. Petroleum underground storage tank risk assessment procedures manual.
Draft, February 1987. Washington, DC: ICF, Inc.
IEP, Inc. 1989. Personal communication regarding gasoline-contaminated groundwater
measurements to Peter Guldberg, Tech Environmental.
nT Research Institute. 1985. Thirteen-week inhalation toxicity study of a 0 to 145°F
gasoline distillate fraction in rats. Sponsored by the American Petroleum Institute.
Chicago, IL: ITT Research Institute.
Infante, P.F., Rinsky, R.A., Wagoner, J. K., and Young, R.J. 1977a. Leukaemia in
benzene workers. Lancet 2:76-78.
Infante, P.P., Rinsky, R.A., Wagoner, J. K., and Young, R.J. 1977b. Benzene and
leukaemia. Lancet 2:866-869.
International Agency for Research on Cancer (IARC). 1982. Benzene. In Some industrial
chemicals and dvestuffs. IARC monographs on the evaluation of carcinogenic risk of
chemicals to humans. Vol. 29, pp. 93-148. Lyon, France: International Agency for
Research on Cancer, World Health Organization.
International Research and Development Corporation (IRDC). 1983. Motor fuel chronic
inhalation study. 6 vols. Submitted to American Petroleum Institute on September 15,
1983.
_. 1984. Amendment to the final report on motor fuel chronic inhalation study,
September 21,1984. Submitted to American Petroleum Institute by D.E. Ulrich,
IRDC, on October 3,1984.
R-20
-------
Intersociety Committee. 1989. Methods of air sampling and analysis. 3rded. Edited by
J. Lodge. Chelsea, MI: Lewis Publishing Inc.
Iregren, A. 1982. Effects on psychological test performance of workers exposed to a
single solvent (toluene) - a comparison with effects of exposure to a mixture of organic
solvents. Neurobehavioral Toxicology and Teratology 4:695-701.
Irving, W.S., and Grumbles, T.G. 1979. Benzene exposures during gasoline loading at
bulk marketing terminals. Am. Ind. Hvg. Assoc. J. 40.
Jackson, J.B.C., Cubit, J.D., Keller, B.D., Batista, V., Bums, K., Caffey, H.M.,
Caldwell, R.L., Garrity, S.D., Getter, C.D., Gonzalez, C, Guzman, H.M.,
Kaufman, K.W., Knap, A.H., Levings, S.C., Marshall, M.J., Steger, R.,
Thompson, R.C., and Weil, E. 1989. Ecological effects of a major oil spill on
Panamanian coastal marine communities. Science 243:37-44.
Jamison, V.W., Raymond, R., and Hudson, J.O. 1975. Biodegradation of high-octane
gasoline in groundwater. Dev. Ind. Microbiol. 16:305-312.
Jenkins, L.J., Jr., Jones, R.A., and Siegel, J. 1970. Long-term inhalation screening
studies of benzene, toluene, o-xylene, and cumene on experimental animals. Toxicol.
ADD!. Pharmacol. 16:818-823.
Johnston, R.V., Pinkerton, M.N., Mensik, D.C., Swaim, L.D., Linscombe, V.A.,
Benge, M.C., Barna-Lyoyd, G., and Kilian, D.J. 1979. Hematologic and
myelogenous effects of inhaled benzene in the pig and the rat. J. Toxicol. Env. Health
5:1025-1035.
lungers, R.H., Lee, Re, Jr., von Lehmden, D.J. 1975. The EPA national fuels
surveillance network. I. Trace constituents in gasoline and commercial gasoline fuel
additives. Environ Health Perspect. 10:143-50.
Karol, M.H. 1986. Respiratory effects of inhaled isocyanates. CRC Critical Review of
Toxicology 16: 349-379.
Kauss, P.B. and Hutchinson, T.C. 1975. The effects of water-soluble petroleum
components on the growth ofChlorella vulgaris Beijerinck. Environ. Pollut. 1: 157-
174.
Kearney, C.A., and Dunham, D.B. 1986. Gasoline vapor exposures at a high volume
service station. Am. Ind. Hvg. Assoc. J. 47(9):535-9.
Keller, K.A., and Snyder, C.A. 1986. Mice exposed in utero to low concentrations of
benzene exhibit enduring changes in their colony forming hematopoietic cells.
Toxicology 42:171-181.
Kenaga, E.E., and Goring, C.A.I. 1978. Relationship between water solubility, soil
sorption, octanol-water partitioning, and concentration of chemicals in biota. In
Aquatic toxicology. Proceedings of the third annual symposium on aquatic toxicology.
New Orleans, LA, 17-18 October, 1978. Edited by J.G.Eaton, P.R. Parrish, and
A.C. Hendricks. ASTM special technical publication 707. n.p.: American Society of
Testing and Materials.
R-21
-------
Kennedy, G., Arnold, D.W., and Calandra, J.C.. 1975. Teratogenic evaluation of lead
compounds in mice and rats. Food Cosmetic Toxicology 13:629-632.
Kimball, K.D., and Levin, S.A. 1985. The limitations of laboratory bioassays: the need
for ecosystem-level testing. Bioscience 35:165-71.
Kirkhart, B. 1980. Micronucleus test on toluene. Prepared by SRI International, Menlo
Park, CA. Research Triangle Park, NC: U.S. Environmental Protection Agency.
Kirk-Othmer. 1980. Kirk-Othmer encyclopedia of chemical technology. 3rd edition.
Vol. 11, pp. 652-695. New York: John Wiley and Sons, Inc. .
Kitchen, D.N. 1984. Neoplastic renal effects of unleaded gasoline in Fischer 344 rats. In
Advances in modem environmental toxicology. Vol. VII. Renal effect of petroleum
hydrocarbons, pp. 65-70. Edited by M.A. Mehlman, C.P. Hempstreet, J.J. Thorpe,
and N.K. Weaver. Princeton, NJ: Princeton Scientific Publishers, Inc.
Klaasen, CD., Amdur, MX)., and Doull, J., eds. 1986. Casarett and Doull's toxicology-
-The basic science of poisons, third edition. New York: Macmillan.
Klaucke, D.N., Johansen, M., and Vogt, R.L. 1982. An outbreak of xylene intoxication
in a hospital. Am. J. Ind. Med. 3:173-178.
Kloss, N.W., Cox, M. G., Norton, J.A., Swenberg, and Bus, J.S. 1984. Sex
dependent differences in the disposition of (14C-5)-2,2,4-trimethylpentane in Fischer
344 rats. In Proceedings of the second international symposium on nephrotoxicitv.
Guildford, Surrey, England, August 1984. (Accepted for publication)
Knave, Bengt M.D., et al. 1976. Long-term exposure to jet fuel. An investigation on
occupationally exposed workers with special reference to the nervous system. Scand.
J. Work Environ. Health 3:152-154.
Konietzko, H., et al. 1980. Cumulative effects of daily toluene exposure. Int. Arch, of
Occup. Environ. Health 46:53-58.
Korseik, £., et al. 1982. Effects of short-term exposure to xylenes on the early cleavage
stages of cod eggs (Gadus morhva L.) Sarsia 67:299-308.
Kostas, J., and Hutchin, J. 1981. Behavioral effects of low-level perinatal exposure to
toluene in mice. Neurobehav. Toxicol. Teratol. 3:467-469.
Kucera, J. 1968. Exposure to fat solvents: A possible cause of sacral agenesis in man. i.
Pediatr. 72:857-859.
Kuna, R.A., and Kapp, R.W., Jr. 1981. The embryotoxic/teratogenic potential of
benzene vapor in rats. Toxicol. Appl. Pharmacol. 57:1-7.
Kuna, R.A., and Ulrich, C.E. 1984. Subchronic inhalation toxicity of two motor fuels.
Journal of the American College of Toxicology 3:217-229.
Kurppa, K., and Husman, K. 1982. Car painters' exposure to a mixture of organic
solvents: serum activities of liver enzymes. Scand. J. Work. Environ. Health
8(2): 137-140.
R-22
-------
Lange, A., et al. 1973a. Leukocyte agglutinins in workers exposed to benzene, toluene,
andxylene. Inter. Arch, fuer Arbeitsmedizia 31:45-50.
Lange, A., et al. 1973b. Serum immunoglobulin levels in workers exposed to benzene,
toluene, and xylene. Inter. Arch, fuer Arbeitsmedizia 31(l):37-44.
Lapss, M.W., Cox, M.G., Norton, R.M., Swenberg, J.A., and Bus, J.S. Sex-dependent
differences in the disposition of (14C-5)-2,2,4-trimethylpentane in Fischer 344 rats. In
Proceedings of the second international symposium on nepthrotoxicity. Guildford,
Surry, England. August 1984. (Accepted for publication)
Larson, D. 1983. Underground tank leaks: subsurface strategies for dealers. Super
Service Station 107:17,58-60.
Lauwerys, R. 1975. Biological criteria for selected industrial toxic chemicals: a review.
Scandinavian Journal of Work. Environment, and Health 1:139-72.
LeBlanc. G.A. 1980. Acute toxicity of priority pollutants to water flea (Daphnia magna).
Bull. Environm. Contam. Toxicol. 24:684-691.
Lebowitz, R, Brusick, D., Matheson, D., Jagannath, D., Reed, M, Goode, S., and Roy,
G. 1979. Commonly used fuels and solvents evaluated in a battery of short-term
bioassays. Environ. Mutapen. 1:172-173. (As cited in NTP, 1986.)
Legludic, H., and Turlais, C. 1914. Recheris sur la toxicite du petrole et quelque-unes de
ses actions physiologiques. Ann. Hve. Pibl. Paris 4 me ser. 21:385. (French, cited in
Browning, 1953.)
LeMesurier, S.M., Lykke, A.W.J., and Stewart, B.W. 1980. Reduced yield of
pulmonary surfactant: patterns of response following administration of chemicals to rats
by inhalation. Toxicology Letters 5:89-93.
LeMesurier, S.M., Stewart, B.W., O'Conneff, P.J., and Lukke, A.WJ. 1979.
Pulmonary responses to atmospheric pollutants. II. Effects of petrol vapor inhalation
on secretion of pulmonary surfactant Pathology 11:81-87.
Leo, A. Hansch, C., and Elkins, P. 1971. Partition coefficients and their uses. Chem.
Revs. 71(6):525-615.
Lepour, M. 1974. Role of environmental lead in increased body burdens of lead in
Hartford children. Environ. Health Perspec. 7:99.
Lewin, L. 1988. uber Allqemein- and Hautvergiftung durch Petroleum. Virchows Arch.
112:35. (German, cited in Browning, 1953.)
Lin, R.S., and Kessler, I.I. 1979. Epidemiologic finding in testicular cancer. Am. J.
Epidemiol. 110:357.
Lin, R.S., and Kessler, I.I. 1981. A multifactorial model for pancreatic cancer in man. J.
Am. Med. Assoc. 245:147-152.
Lipovskii, S.M. 1978. Serotonin content in uterine tissues and characteristics of the
contractile activity of the uterus following exposure to gasoline vapors. (Abstract)
Gig. Tr. Prof. Zabol. 7:37-40. (Russian)
R-23
-------
Litton Bionetics, Inc. 1977a. Mutagenicitv evaluation of unleaded gasoline. Final report.
Submitted to the American Petroleum Institute, March 1977. Washington, DC:
American Petroleum Institute.
. 19T7b. Rat bone marrow cytogenetic analysis. Unleaded gasoline. Final
report. Submitted to the American Petroleum Institute, November 1977. Washington,
DC: American Petroleum Institute.
. 1978a. Mutagenicitv evaluation of toluene. Final report Submitted to the
American Pea-oleum Institute, May 1978. LBI Project No. 20847. Kensington, MD:
Litton Bionetics, Inc.
. 1978b. Teratology study in rats. Toluene. Final report. Study conducted by
Litton Bionetics, Inc. Washington, DC: American Petroleum Institute.
. 1978c. Teratology study in rats. Xvlene. Final report. Submitted to the
American Petroleum Institute, January 1978. LBI Project No. 20698-4. Kensington,
MD: Litton Bionetics, Inc.
. 1978d. Teratology study in rats. Unleaded gasoline. Final report. Submitted
to the American Petroleum Institute. Washington, DC: American Petroleum Institute.
. 1980. Mutagenicitv evaluation of gasoline. API PS-6 fuel in the mouse
dominant lethal assay. Final report. Submitted to the American Petroleum Institute,
June 1980, by Litton Bionetics, Inc. Washington, DC: American Petroleum Institute.
. 1981. Mutagenicitv evaluation of toluene: mouse dominant lethal assay.
Submitted to the American Petroleum Institute. Washington, DC: American Petroleum
Institute.
1982. Inhalation teratology study in rats: benzene. API Medical Research
Publication No. 30-30224. Washington, DC: American Petroleum Institute. (As cited
in NTP, 1986.)
Lock, E.A., Charbonneau, M., Strasser, J., and Bus, J.S. 1987a. The renal transport of
pentanoic acids derived from 2,2,4-trimethyl pentane. The Toxicologist 7:89.
Lock, E.A., Charbonneau, M., Strasser, J., and Bus, J.S. 1987b. The reversible binding
of 2,2,4-trimethyl pentane (TMP) to renal alpha-2-microglobulin in male Fischer 344
rats. The Toxicologist 7:27.
Lock, E.A., Stonard, M.D., and Elcombe, C.R. 1985. The selective induction of o>- and
B-oxidation of fatty acids in the liver and kidney of male rats adminstered 1,2',4-
trimethylpentane. The Toxicologist 5:59.
Lombardo, M., and Behrens, G. 1987. Survey of vehicle refueling. SAE Paper 871085.
Society of Automotive Engineers Inc., Government/Industry Meeting, Washington,
D.C. May 1987. n.p.: Society of Automotive Engineers.
Longacre, S.L., Kocsis, J.J., and Snyder, R. 1981a. Influence of strain differences in
mice on the metabolism and toxicity of benzene. Toxicol. Appl. Pharmacol. 60:398-
409.
R-24
-------
Longacre, S.L., Kocsis, J.J., Witner, CM, Lee, E.W., Sammett, D., and Snyder, R.
198 Ib. lexicological and biochemical effects of repeated administration of benzene in
mice. J. Toxicol. Environ. Health 7:223-237.
Loury, D J., and Butterworth, B.E. 1986a. Kidney-specific DNA repair assay: an
evaluation of unleaded gasoline. CITT Activities 6(4): 1-8.
Loury, D.J., Smith-Oliver, TM and Butterworth, B.E. 1985. Measurements of DNA
repair and cell replication in hepatocytes from rats exposed to 2,2,4-trimethyl-pentane
or unleaded gasoline. Environmental Mutagen Society. Sixteenth Annual Meeting
7(supplement 3):70.
Loury, D.J., Smith-Oliver, T., and Butterworth, B.E. 1987. Assessment of unscheduled
and replicative DNA synthesis in rat kidney cells exposed in vitro or in vivo to unleaded
gasoline. Toxicology and Applied Pharmacology 87:127-140.
Loury, D.J., Smith-Oliver, T., Strom, S., Jirtle, R., Michalopoulos, G., and Butterworth,
B.E. 1986b. Assessment of unscheduled and replicative DNA synthesis in
hepatocytes treated in vivo and in vitro with unleaded gasoline or 2,2,4-
trimethylpentane. Toxicology and Applied Pharmacology 85:11-23.
Lovchikov, V.A., Lipovskii, I.M., Strel'tsova , N.A., and Altuknov, V. V. 1978.
Dynamics of nervous systems excitability and the level of ovarian secretion in different
phases of the estrous cycle in albino rats. Abstract Fiziol. Zh. Kiev 24(1): 115-118.
(Russian)
Luke, C.A., Tice, R.R., and Drew, R.T. 1985. Duration and regimen-induced
micronuclei in the peripheral blood of mice exposed chronically to benzene. Environ.
Mutagen. 7, Suppl. 3, 29 (As cited in Dean, 1985b.)
Luster, M.I., and Rosenthal, G.J. 1986. The immunosuppressive influence of industrial
and environmental xenobiotics. TIPS. October 1986. p. 409.
Lyapkalo, A.A. 1973. Genetic activity of benzene and toluene. Gig. Tr. Prof. Azbol.
17:24-28. (In Russian with English summary, evaluation based on an English
translation provided by U.S. EPA.)
Lykke, A.W.J., and Stewart, B.J. 1978. Fibrosing alveolitis (pulmonary interstitial
fibrosis) evoked by experimental inhalation of gasoline vapours. Experientia 34:498.
Lykke, A.W.J., Stewart, B.W., O'Connel, P.J., and LeMesurier, S.M. 1979.
Pulmonary responses to atmospheric pollutants. In An ultrastructural study of
fibrosing alveolitis evoked by petrol vapour. Pathology 11:71-80.
Lyman, W.J., Reehl, W.F., and Rosenblatt, D.H. 1982. Handbook of chemical
property estimation methods. New York: McGraw-Hill.
MacFarland, H.N., Ulrich, C.E., Holdsworth, C.E., Kitchen, D.N., Halliwell, W.H.,
and Blum, S.C. 1984. A chronic inhalation study with unleaded gasoline vapor.
Journal of the American College of Toxicology 3(4):231-248.
Machle,W. 1941. Gas intoxication. J. Am. Med. Assoc. 177:1965-1971.
R-25
-------
Mackay, D., and Leinonum, P.J. 1975. Rate of evaporation of low solubility
contaminants from water bodies to atmosphere. Environ. Sci. Technol. 9(13):1178-
1180.
Mackay, D.M., Roberts, P.V., and Cherry, J.A. 1985. Transport of organic
contaminants in groundwater. Environ. Sci. Technol. 19(5):384-392.
MacMahon, B. 1985. Review of "Lifetime risks of kidney cancer estimated from animal
and epidemiologic studies" by P.E. Enterline. Prepared for Motor Vehicle
Manufacturers' Association.
Maine Department of Health and Human Services. 1986. Petroleum contamination of
Maine's drinking water wells. Augusta, ME: Maine Department of Health and Human
Services.
Maki-Paakkanen, J., Husgafvel-Pursiainen, K., Kalliomaki, P.-L., Tuominen, J., and
Sorsa, M. 1980. Toluene-exposed workers and chromosome aberrations. Journal of
Toxicology and Environmental Health 6:775-781.
Mallory, T.B., Gall, E.A., and Brickley, W.J. 1939. Chronic exposure to benzene
(benzol). IH. The pathologic results. J. Ind. Hve. Toxicol. 21:355-377.
Maltoni, C, Conti, B., Com, G., and Belpoggi, F. 1985. Experimental studies on
benzene carcinogenicity at the Bologna Institute of Oncology: Current results of
ongoing research. Am. J. Ind. Med. 7:415-446.
Marks, T.A., Ledoux, T.A., and Walbridge, C.T. 1976. Teratogenicity of a commercial
xylene mixture in the mouse. J. Toxicol. Environ. Health 9:97-105.
Marley, M.C. and Hoag, G.E. 1984. Induced soil venting for recovery/ restoration of
gasoline hydrocarbons in vadose zone. Presented at The NWWA/API conference on
petroleum hydrocarbons and organic chemicals in groundwater - Prevention, detection
and restoration. Houston, Texas.
Martin, C.N., and Campbell, J. 1985. Tests for the induction of unscheduled DNA
synthesis in HeLa cells. In Evaluation of short-term tests for carcinogens: report of the
international program on chemical safety collaborative study on in vitro assays, pp.
375-379. Edited by J. Ashby, F.J. de Serres, M. Draper, M. Ishidate, Jr., B.H.
Margolin, B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier. (As cited in Dean,
1985b.)
Matsushita, K. 1935. Pathplogische-historologische studien uber das zentral-
nervensystem bei experimenteller petroeumvergiftung. Nagasaki Igakkwai Zassi
13:967. (Zbl. Gew Hyg., 22, abstr. sec., 190) (Cited in Browning, 1953.)
Matsushita, T., Arimatsu, Y., Ueda, A., et al. 1975. Hematological and neuromuscular
response of workers exposed to low concentrations of toluene vapor. Indust. Health
13:115-121. (Cited in U.S. EPA, 1983b.)
McAuliffe, M. 1963. Solubility in water of C1-C9 hydrocarbons. Nature 200:1092-
1093.
R-26
-------
McCarrolI, N., Keech, B., and Piper, C.E. 198la. A microsuspension adaptation of the
Bacillus subtilis "rec" assay. Environ. Mutagen. 3:607-616.
McCarrolI, N., Piper, C.E. and Keech, B. 198 Ib. An E. Coli microsuspension assay for
the detection of DNA damage induced by direct-acting agents and promutagens.
Environ. Mutagen. 3:429-444.
McClain, R.M., and Becker, B.A. 1971. Effects of organolead compounds on rat
embryonic and fetal development. Toxicology and Applied Pharmacology 21:265-274.
McDermott, H., and Vos, G.A. 1979. Service station attendants' exposure to benzene
and gasoline vapors. Am. Ind. Hve. Assoc. J. 40: 315- 321.
McDermott, H.J., and Killiany, S.E. 1978. Quest for gasoline TLV. American Industrial
Hygiene Association Journal 39:110-117.
McGraw, D.S., Joyner, R.E., and Cole, P. 1985. Excess leukemia in a refinery
population. J. Qccup. Med. 27:220-222.
McGregor, D., and Ashby, J. 1985. Summary report on the performance of the cell
transformation assays. In Evaluation of short-term tests for carcinogens: report of the
international program on chemical safety collaborative study on in vitro assays, pp.
103-115. Edited by J. Ashby, F.J. de Serres, M. Draper, M. Ishidate, Jr., B.H.
Margolin, B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier. (As cited in Dean,
1985b.)
McKone, T. 1987. Human exposure to volatile organic compounds in household tap
water: the indoor inhalation pathway. ES&T21: 1194-1201.
McLaughlin, J.K. 1984. Risk factors from a population-based case-control study of renal
cancer. In Renal effects of petroleum hydrocarbons: Advances in modem
environmental toxicology. Vol. VII, pp. 227-244. Edited by M.A. Mehlman.
Princeton, NJ: Princeton Scientific Publishers, Inc.
McLaughlin, J.K., Blot, W.J., Mandel, J.S., Schuman, L.M., Mehl, E.S., and Fraumeni,
J.F., Jr. 1983. Etiology of cancer of the renal pelvis. JNCT 71:287-291.
McLaughlin, J.K., Blot, W.J., Mehl, E.S., Stewart, P.A., Venable, F.S., and Fraumeni,
J.F. 1985. Petroleum-related employment and renal cell cancer. J. OCCUD. Med. 27:9,
672-674.
McLaughlin, J.K., Mandel, J.S., Blot, W.J., Schuman, L.M., Mehl, E.S., and Fraumeni,
J.F.. 1984. A population-based case-control study of renal cell carcinoma. JNCT
72:275-284.
McLean, J.A. 1960. Blood dyscrasias after contact with petrol containing benzol. Med.
J. Austrial. 47:845-849.
McMichael, A.J., Spirts, R., Jupper, L.L., and Gamble, J.F. 1975. Solvent exposure
and leukemia among rubber workers: an epidemiologic study. J. OCCUP. Med.
17(4):234-239.
McNellis, D.N., Earth, D.C., Khare, M., et al. 1984. Environmental Research Center,
University of Nevada at Las Vegas. Exposure assessment methodologies for
R-27
-------
hazardous waste sites. Las Vegas, NY: Office of Research and Development
Environmental Monitoring Systems Laboratory. CR810550-01.
Mehlman, M.A., Schreiner, CA., and Mackerer, C.R. 1980. Current status of benzene
teratology: a brief review. Journal of Environmental Pathology and Toxicology 4:123-
131.
Meyne,T., and Legator, M. 1980. Sex-related differences in cytogenetic effects of
benzene in the bone marrow of swiss mice. Environ. Mutagen. 2:43-50. (As cited in
NTP, 1986.)
Michon, S. 1965. Disturbance of menstruation in women working in an atmosphere
polluted with aromatic hydrocarbons. Pol. Tyg.Lek. 20:1547-1649. (Cited in U.S.
EPA, 1983b).
Mikulski, P., Wiglusz, R., Bubleska, A., and Uselis, J. 1972. Investigation of exposure
of ships' painters to organic solvents. Br. J. Ind. Med. 29(4):450-453.
Milham, S., Jr. 1983. Occupation mortality in Washington State 1950-1979. Prepared
for the Division of Surveillance, Hazard Evaluations and Field Studies, National
Institute for Occupational Safety and Health. NTIS No. PB84-199769. Cincinnati,
OH: n.p.
Miller, M.M., et al. 1985. Relationships between octanol-water partition coefficient and
aqueous solubility. Environ. Sci. Technol. 19(6):522-529.
Mirinski, V. 1979. Amenorrhea or women exposed to the chronic effect of small doses of
gasoline vapors. Probl. Akus Ginekol. 7:15-18. (Bui., Abstract.)
Mirkova, E., Zaikov, C., Anrov, G., Mikhailova, A., Khinkova, L., and Benchev, I.
1983. Prenatal toxicity of xylene. Toxicology and Immunology 27, No. 3:337-?.
Mommsen, S., Aagaard, J., and Sell, A. 1982. An epidemiological case-control study of
bladder cancer in males from a predominantly rural district. Eur. J. Cancer Clin.
Oncol. 18:1205-1210.
Mommsen, S., Aagaard, J., and Sell, A. 1983. A case-control study of female bladder
cancer. Eur. J. Cancer Clin Oncol. 19:725.
Mommsen, S., and Aagaard, J.. 1983a. A case-control study of bladder cancer. A
multivariate, stratified analysis of a low risk population. Dan. Med. Bull. 30:427.
Mommsen, S., and Aagaard, J. 1983b. Tobacco as a risk factor to bladder cancer.
Carcinogenesis 4:335.
Mommsen, S., and Aagaard, J. 1984. Occupational exposures as risk indicator of male
bladder carcinoma in a predominantly rural area. Acta Radiol. Oncol. 23:147-152.
Mommsen, S., and Sell, A. 1983. Prostatic hypertrophy and venereal disease as possible
risk factors in the development of bladder cancer. Urol. Res. 11:49.
Morgan, R.W., and Wong, O. 1985. An epidemiologic mortality study of Mobil Oil
Corporation employees at the Paulsboro, New Jersey, refinery. Final report.
R-28
-------
Submitted to Mobil Oil by Environmental Health Associates, Oakland, California.
(Unpublished)
Morimoto, K. 1974. Combined cytogenetic effects of benzene and radiation on cultured
human lymphocytes. Jpn. J. Ind. Health 17:106-107. (As cited in NTP, 1986.)
Morimoto, K. 1983. Induction of sister chromatid exchanges and cell division delays in
human lymphocytes by microsomal activation of benzene. Cancer Res. 43:1130-
1334. (As cited in NTP, 1986.)
Morimoto, K., and Wolff, S. 1980. Increase of sister chromatid exchanges and
perturbations of cell division kinetics in human lymphocytes by benzene metabolites.
Cancer Res. 40:1189-1193. (Cited in NTP, 1986.)
Morley, R., Eccleston, D.W., Douglas, C.P., Greville, W.E.J., Scott, D.J., and
Anderson, J. 1970. Xylene poisoning: a report on one fatal case and two cases of
recovery after prolonged unconsciousness. Fr. Med. J. 3:42-443.
Mortelmans, K.E., and Ricco, E.S. 1980. In vitro microbiological genotoxicity assays of
toluene. Prepared by SRI International, Menlo Park, CA. Research Triangle Park,
NC: U.S. Environmental Protection Agency.
Moszczynski, P. 1981. Organic solvents and T-lymphocytes. Lancet 1:438.
Moszczynski, P. 1982. Effect of organic solvents in the course of professional exposure
on the E rosette text and skin reactions against distreptase and tuberculin. Folia
Haematol. 109:224-229.
Moszczynski, P., and Lisiewicz, J. 1983a. Hematological indices of peripheral blood in
workers occupationally exposed to benzene, toluene and xylene. Zenbralbl Bakteriol
Mikrobiol Hyg. Abt. 1. Org. B. Hyg. Umwelthvq Krankenhaushyq. arbeitshvq. praev
Med. 178(4):329-339.
Moszczynski, P., and Lisiewicz, J. 1983b. Occupational exposure to benzene, toluene
and xylene and the T-lymphocyte functions. Haematologia 17(4):449-454.
Murray, F.J., John, J.A., Rampy, L.W., Kuna, R.A., and Schwetz, B.A. 1979.
Embryotoxicity of inhaled benzene in mice and rabbits. Am. Ind. Hvg. Assoc. J.
40:993-998.
Najem, G.R., Louria, D.B., Seebode, J.J., Thind, I.S., Prusakowski, J.M., Ambrose,
R.B., and Fernicola, A.R. 1982. Lifetime occupation, smoking, caffeine, saccharine,
hair dyes and bladder carcinogenesis. Int. J. Epidemiol. 11:212-217.
National Institute for Occupational Safety and Health (NIOSH). 1974. Criteria for a
recommended standard - Occupational exposure to benzene. Cincinnati, OH: National
Institute for Occupational Safety and Health.
. 1975. Criteria for a recommended standard - Occupational exposure to xvlene.
U.S. Department of Health, Education, and Welfare, Public Health Service, Centers
for Disease Control, Cincinnati, OH. HEW Publication No. (NIOSH) 75-168. n.p.:
National Institute for Occupational Safety and Health.
R-29
-------
National Petroleum News. 1984. How self-service appeals to motorists. National
Petroleum News 1983 Factbook 75(7A): 103.
National Research Council (NRC). 1976. Health effects of benzene. A review, n.p.:
National Research Council. NTIS PB-254388.
. 198la. The alkvlbenzenes. Washington, DC: National Academy Press.
.. 1981b. Testing for the effects of chemicals on ecosystems. Washington, DC:
National Academy Press.
. 1983. Risk assessment in the federal government: managing the process.
Washington, DC: National Academy Press.
. 1986a. Emergency and continuous exposure guidance levels for selected
airborne contaminants. Vol. 6. Benzene and ethvlene oxide. Washington, DC:
National Academy Press.
. 1986b. Knowledge and environmental problem solving. Washington, DC:
National Academy Press.
National Toxicology Program (NTP). 1986a. Technical report on the toxicology and
carcinogenesis of xvlenes (mixed) (60.2% m-xvlene. 13.6% p-xvlene. 17.0%
ethvlbenzene. and 9.1% o-xvlene (CAS No. 1330-20-7) in F344/N rats and B6C3F^
mice Ravage studies). NTP Technical Report 327. Research Triangle Park, NC: U.
S. Department of Health and Human Services, Public Health Service, National
Institutes of Health, National Toxicology Program.
. 1986b. Toxicology and carcinogenesis studies of benzene (CAS No. 71-43-2)
in F344/N rats and B6C3Fj^ mice (gavage studies). Research Triangle Park, NC.: U.
S. Department of Health and Human Services, Public Health Service, National
Institutes of Health, National Toxicology Program.
Nawrot, P.S., and Staplles, R.E. 1979. Embryofetal toxicity and teratogenicity of
benzene and toluene in the mouse. Teratology 19:41.
Neff, J.M., et al. 1976. Effects of petroleum on survival, respiration and growth of
marine animals. In Sources, effects and sinks of hydrocarbons in the aquatic
environment. Proc. of a symposium, pp. 516-532. Washington, DC: American
University, American Institute of Biological Science.
Neilsen, G.D., and Alarlie, Y. 1982. Sensory irritation, pulmonary irritation, and
respiratory stimulation by airborne benzene and alklbenzenes: prediction of safe
industrial exposure levels and correlation with their therrnodynamic properties.
Toxicol. April. Pharmacol. 65:459-477.
Nelson, N.A. 1985. Mortality in Amoco oil refinery employees. Phase I. TSCA Section
8 (e) Report, 8EHQ-0585-0557. (Unpublished)
Nestmann, E.R., Lee, G.G.-H., Matula, T.I., Douglas, G..R., and Mueller, J.C. 1980.
Mutagenicity of constituents identified in pulp and paper mill effluent using the
5a//n0/i£//a/marnrnalian-microsorne assay. Mutat. Res. 79:203-212.
R-30
-------
New Hampshire Department of Health and Human Services (NH DHHS). 1988. Private
water supply contamination data in Northwood Ridge homes. Personal
communication, March 11,1988.
New Hampshire Division of Public Health Services (NHDPHS). 1987. Indoor air
benzene concentrations found in six homes with known gasoline contamination of
drinking water. Concord, NH: New Hampshire Division of Public Health Services.
New Jersey Department of Environmental Protection (NJDEP). 1984. An exposure-risk
assessment for benzene in shower air. Paper presented at APCA Annual Meeting.
November 1984.
Nissen, N..L, and Ohlsen. A. 1953. Erythromyelosis review and report of a case in a
benzene (benzol) worker. Acta Med. Scand. 145:56-71.
Nomiyama, R., and Nomiyama, H. 1974a. Respiratory elimination of organic solvents
in man: benzene, toluene, n-hexane, trichloroethylene, acetone, ethyl acetate and ethyl
alcohol. Int. Arch. Arbeitsmed. 32:85-91.
Nomiyama, R., and Nomiyama, H. 1974b. Respiratory retention, uptake and excretion of
organic solvents in man: benzene toluene, n-hexane, trichloroethylene, acetone, ethyl
acetate and ethyl alcohol. Int. Arch. Arbeitsmed. 32:75-83.
Novikov, Y.V. 1956. Effect of small benzene concentrations on higher nervous activity
of animals in chronic experiments. Giq. Sanit. 21:20-25. Translated by B.S. Levine.
U.S. Public Health Service 2 (1960):185-191.
Nylander, P.M., Olofsson, H., Rasmusson, B., and Swaklin, H. 1978. Mutagenic
effects of petrol in Drosphilia melangaster. I. Effects of benzene and 1,2-
dichloroethane. Mutat. Res. 57:163-167. (As cited in NTP, 1986.)
Office of Science and Technology Policy (OSTP). 1985. Chemical carcinogens: Review
of the science and its associated principles. Federal Register 50:10372-10442.
Ogata, M., Tomokumi, K., and Takatsuka, Y. 1970. Urinary excretion of hippuric acid
and m- or p-methylhippuric acid in the urine of persons exposed to vapours of toluene
and m- or p-xylene as a test of exposure. Br. J. Ind. Med. 27:43-50.
Ogata, M., and Miyake, Y. 1979. Disappearance of aromatic hydrocarbons and organic
sulfur compounds from fish flesh reared in crude oil suspension. Water Res. 13:75-
78.
OH, M.B., Townsend, J.C., Fisbeck, W.A., and Langner, R.A. 1978. Mortality among
individuals occupationally exposed to benzene. Arch. Environ. Health 33:3-10.
Olson C.T., K., Yu., O., Hobson, D.W., and Serve, M. P. 1985. Identification of
urinary metabolites of the nephrotoxic hydrocarbon 2,2,4-trimethylpentane in male rats.
Biochemical and Biophysical Research Communications 130:331-316.
Osgood, J.O. 1974. Hydrocarbon dispersion in groundwater: significance and
characteristics. Groundwater 12(6).
Ott, M.G., Townsend, J.C., Fishbeck, W.A., and Langner, R.A. 1978. Mortality among
individuals occupationally exposed to benzene. Arch. Environ. Health 33:3-9.
R-31
-------
Overman, D.O. 1981. Testing for percutaneous embryotoxicity of laboratory reagents in
the hamster. Teratology. 23:56A.
Ovrum, P., Hultengren, M, and Lindqvist, T. 1978. Exposure to toluene in a
photogravure printing plant: concentration in ambient air and uptake in the body.
Scand. J. of Work. Env.. and Health 4:237-45.
PACE. 1983. Seminar on groundwater and petroleum hydrocarbons. Protection,
detection, restoration. In Proceedings: Petroleum Association for Conservation of the
Canadian Environment. Toronto, Ontario. June 1983.
Paksy, K.A., Molnar, J., Naray, M.M., and Barcza, G. 1982. Comparative study on the
acute effects of benzene, toluene, and m-xylene in the rat Acta Phvsiol. Acad. Sci.
Hung. 59:317-324.
Palitti, F., Fiore, M., DeSilva, R., Tanzarella, G, Ricordi, R., Forster, R., Mosesso, P.,
Astolfi, S., and Loprieno, N. 1985. Chromosome aberration assays of 5 chemicals in
Chinese hamster cells in vitro. In Evaluation of short-term tests for carcinogens:
Report of the international program on chemical safety collaborative study on in vitro
assays, pp. 443-450. Edited by J. Ashby, F.J. de Series, M, Draper, M. Ishidate, Jr.,
B.H. Margolin, B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier Publishing
Company. (As cited in Dean, 1985b.)
Pandya, K.P., Rao, G.S., Dhasmana, A., and Zaidi, S.H. 1975. Occupational exposure
of petrol pump workers. Ann. Occup.Hvg. 18:363-364.
Parry, J.M. 1985. Summary report on the performance of the yeast and Asperqillus
assays. In Evaluation of short-term tests for carcinogens: report of the international
program on chemical safety collaborative study on in vitro assays, pp. 25-46. Edited
by J. Ashby, F. J. de Serres, M. Draper, M. Ishidate, Jr., B.H. Morgolin, B.E.
Matter, and Shelby. Amsterdam: Elsevier. (As cited in Dean, 1985b)
Parry, J.M., and Eckardt, F. 1985. The induction of mitotic aneuploidy, point mutation
and mitotic crossing-over in the yeast Saccharomyces cerevisiae strains D61-M and D6.
In Evaluation of short-term tests for carcinogens: Report of the international program
on chemical safety collaborative study on in vitro assays, pp. 261-269. Edited by J.
Ashby, F.J. de Serres, M. Draper, M. Ishidate, Jr., B.H. Margolin, B.E. Matter, and
M.D. Shelby. Amsterdam: Elsevier.
Pasquill, F. 1961. The estimation of the dispersion of windborne material. Meteorol.
Maq. 90, 1063:33-49.
Patty's industrial hygiene and toxicology. 1981. 3rd rev. ed. Volume 23: Toxicology.
Edited by George D. Clayton and Florence E. Clayton. New York: John Wiley and
Sons.
Patty's industrial hygiene and toxicology. 1983. 3rd rev. ed. Volume 2B: Toxicology.
Edited by George D. Clayton and Florence E. Clayton. New York: John Wiley and
Sons.
Phillips, C.F., et. al. 1978. Gasoline vapor exposure during bulk handling operations.
American Industrial Hygiene Association Journal 39:118-128.
R-32
-------
Phillips Petroleum Company. 1984. Toxicity study summary for unleaded gasoline.
Submitted to The American Petroleum Institute on November 12,1984.
Pickle, L.W., and Gottlieb, M.S. 1980. Pancreatic cancer mortality in Louisiana. Am. J.
Public Health 70:256-259.
Plotnikov, Y.K. 1978. Characteristics of occupational reference of patients with leukemia
and lymphogranulomatosis (according to the data of the Kaibyshev region). Probl.
Gematol. Pereliv. Krovi 23:18-21.
Poincare, L. 1885. Recherches experimentales sur les effects d'unair charge de vapeours
depetrole. Ann. Hvq. Publ. (Paris. 3me ser.)13:312. (French, cited in Browning
1953.)
Poklis, A., and Burkett, C.D. 1977. Gasoline sniffing: a review. Clinical Toxicology
11(1): 35-41.
Poison, C.J., and Tattersall, R.N. 1969. Clinical toxicolgv. 2nd ed.. pp. 430-431.
Philadelphia: LippincotL
Pound, A.W. 1970. Induced cell proliferation and the intiation of skin tumour formation
in mice by ultraviolet light. Pathology 2:269-275.
Price, K.S. et al. 1974. Brine shrimp bioassay and sea water BOD (biochemical oxygen
demands of petrochemicals). J. Water Pollut. Control Fed. 46(l):63-77.
Probst, G.S., McMaho, R.E., Holl, L.E., Thompson, C.Z., Epp, J.K., and Neal, S.B.
1981. Chemically-induced DNA synthesis in primary rat hepatocyte cultures: a
comparison with bacterial mutagenicity using 218 compounds. Environ.
Mutaeen.3:11-32.
Przybylowski, J. 1979. Histopathological and histochemical changes in the liver and lung
and behavior of certain biochemical parameters of blood serum in chronic experimental
benzine and ethyline poisoning. Patol. Pol. 30(3):387-399. (Polish)
Przybylowski, J., Kowalki, WM and Podolocki, A. 1976a. Effect of chronic
experimental poisoning with leaded and straight gasoline vapors on the circulatory
system. (Abstract). Oatik. Oik. 27(2): 149-156.
Przybylowski, J., Matuszewski, W., Podolecki, A., and Kaminski, K. 1977. Blood
serum and tissue monoamine oxidase and tissue glyoxalase 1. Activities in
experimental chronic intoxication with gasoline and lead gasoline. (Abstract).
Bromatol. Chem. Toksvkol. 10(l):75-78. (Polish)
Przybylowski, J., Wysocki, J.M., and Podolecki, A. 1974. Study on blood serum
proteins in rabbits chronically intoxicated with ethyline 78 and benzine. Bromotol.
Chem. Toksvkol. 7(2^:151-156. (Polish)
Przybylowski, J., Wysocki, J., Szczepanski, Z., Sychlowy A., and Podolecki, A.
1976b. Phagocytic activity of granulocytes, blood serum proteins, and properdine
level in workers chronically exposed to lead gasoline. (Abstract). Bromatol. Chem.
Toksvkol. 9:33-39. (Polish)
R-33
-------
Przybylowski, J., Zych, F., and Podolecki, A. 1978a. Effects of chronic experimental
intoxication with benzine and ethyline vapours on the pituitary-adrenal system.
(Abstract). Endokrvnol. Pol. 29(5):399-406. (Polish)
Przybylowski, J., Zych, F., Podolecki. A., et al. 1978b. Health status in workers
chronically exposed to ethyl gasoline and other petroleum products. (Abstract). Med.
Pr. 29:539-544. (Polish)
Pyykko, K. 1980. Effects of methylbenzenes in microsomal enzymes in rat liver, kidney
and lungs. Biochem Biophvs. Acta 633:1-9.
Pyykko, K. 1984. Effects of pretreatment with toluene, phenobarbital, and 3-
methylcholanthrene on the in vivo metabolism of toluene and on the excretion of
hippuric acid in the rat Pharmacol. Res. Commun. 16:217-25.
Ransley, D.L. 1984. Xylenes and ethylbenzene. In Kirk-Othmer encyclopedia of
chemical technology. 3rd ed., Vol. 24, pp. 709-744. New York: John Wiley and
Sons, Inc.
Rao, G. S., and Pandya, K.K. 1978. Toxicity of petroleum products: effects on alkaline
phosphatase and lipid peroxidation. Environ. Res. 16:174-178.
Rao, G.S., and Pandya, K.P. 1980. Hepatic metabolism of heme in rats after exposure to
benzene, gasoline and kerosene. Archives of Toxicology 46:313-317.
Rappaport, S., Selvin, S., and Walters, M. 1987. Exposures to hydrocarbon components
of gasoline in the petroleum industry. Am. Ind. Hyg. Assoc. J. 2 (4): 148.
Reeve, G.R., Thomas, T.L., Kelly, V.F., Waxweiler, R.J., and Itaya, S. 1982. A
proportionate mortality study of an Oil, Chemical, and Atomic Workers local in Texas
City, Texas. An. N.Y. Acad. Sci. 381:54-61.
Registry of Toxic Effects of Chemical Substances (RTECS). 1983,1984,1987. Registry
of Toxic Effects of Chemical Substances. Washington, DC: U.S. Dept. of Health and
Human Services, National Institute of Occupational Safety and Health (NIOSH).
Reynolds, S.H., Stowers, S.J., Patterson, R.M., Maronpot, R.R., Aaronson, S.A., and
Anderson, M.W. 1987. Activated oncogenes in B6C3Fi mouse liver tumors:
implications for risk assessment. Science 237:1,309-16.
Rhudy, R.L., et al. 1978. Ninety-day subacute inhalation study with toluene in albino
rats. Toxicol. Appl. Pharmacol. 45:284-285.
Richardson, K.A., Wilmer, J.L., Smith-Simpson, D., and S kopek, T.R. 1986.
Assessment of the genotoxic potential of unleaded gasoline and 2,2,4-trimethylpentane
in human lymphoblasts in vitro. Toxicology and Applied Pharmacology 82:315-322.
Riihimaki, V., Savolainen, K., Pfaffli, K., Sippel, H., and Laine, A. 1982. Metabolic
interaction between m-xylene and ethanol. Arch. Toxicol. 49:253-63.
Rinsky, R.A., Smith, A.B., Hornung, R., Filloon, G., Young, J., Okun, H., and
Landrigan, J. 1987. Benzene and leukemia: an epidemiological risk assessment. New
England J. Med. 316(17):1044-1050.
R-34
-------
Rinsky, R.A., Young, R.J., and Smith, A.B. 1981. Leukemia in benzene workers.
Amer. J. Ind. Med. 2:217-245.
Rooney, S.A. 1984. Lung surfactant. Environmental Health Perspectives 55:205-226.
Rosenkranz, H., and Leifer, Z. 1980. Determining the DNA-modifying activity of
chemicals using DNA-polymerase-deficient Escherichia coli. In Chemical mutagens:
Principles and methods for their detection. Vol. 6, pp. 109-147. Edited by F. de
Series and A. Hallaender. New York: Plenum Publishers. (As cited in NTP, 1986.)
Rothman, D., and Johnson, R. 1985. Refueling emissions from uncontrolled vehicles.
August 1985. Ann Arbor, MI: U.S. Environmental Protection Agency, Mobile Source
Laboratory.
Rotunno, R.M., Tarrantino, M., and Bonsignore, D. 1976. Atmospheric lead pollution.
An epidemiological study in gasoline station workers. Lav. Lem. 28:65-72.
Rozen, M.G., and Snyder, C.A. 1985. Protracted exposure of C57BL/6 mice to 300 ppm
benzene depresses B~ and Tlymphocyte numbers and mitogen responses. Evidence
for thymic and bone marrow proliferation in response to the exposures. Toxicology
37:13-26.
Rozen, M.G., Snyder, C.A., and Albert, R.E. 1984. Depressions in B' and T
lymphocyte mitogen-induced blastogenesis in mice exposed to low concentratons of
benzene. Toxicol. Lett. 20:343-349.
Rozman, C, Woessner, S., and Salz-Serrania, L. 1968. Acute erythromyelosis after
benzene poisoning. Acta Haematol. 40:234-237.
Runion, H.E. 1975. Benzene in gasoline. Am. Ind. Hyg. Assoc. J. 36:338-350.
Rushmore, T., Snyder, R., and Kalf, G. 1984. Covalent binding of benzene and its
metabolites to DNA in rabbit bone marrow mitochondria in vitro. Chem. Biol. Interact.
49:133-154.
Rushton, L., and Alderson, M.R. 1981a. An epidemiological survey of eight oil
refineries in Britain. Br. J. Ind. Med. 38:225-234.
Rushton, L., and Alderson, M.R. 198 Ib. A case-control study to investigate the
association between exposure to benzene and deaths from leukemia in oil refinery
workers. Br. J. Cancer 43:77-84.
Rushton, L., and Alderson, M.R. 1983. An epidemiologic survey of oil distribution
centers. Br. J. Tnd. Med. 40:330-339.
Rykowski, R. A.. 1986. Research: analyzing gasoline volatility. Presented at The 1986
Washington Conference on Ozone Control Strategy. September 1986, Arlington, VA.
Saito, K. 1973. Electroencephalographic studies on petrol intoxication: comparison
between nonlead and leaded white petrol. Brit. J. of Indust. Med. 30:352-358.
R-35
-------
Salamanca-Gomez, F., Palma, V., Navarrete, C, Garcia, T., Moreta, G., Hernandez, S.,
Buentello, L. 1989. Chromosome abnormalities and sister chromatid exchanges in
children with acute intoxication due to inhalation of volatile substances. Archives of
Environmental Health 44, No. 1:49-53.
Sanders, W.N., and Maynard, J.B. 1968. Capillary gas chromatographic method for
determining the C3 - C\2 hydrocarbons in full-range motor gasolines. Analytical
Chemistry 40(4).
Sanitz, D.A., and Moure, R. 1984. Cancer risk among oil refinery workers. L
Occupational Med. 26:662-670.
Sanner, T., and Rivedal, E. 1985. Tests with the Syrian hamster embryo (SHE) cell
transformation assay. Prog. Mutat. Res. 5:665-671.
Santesson, C.G. 1987. Chronic poisoning with coal tar benzene: four deaths. Clinical
and pathological anatomical observations of several colleges and illustrating animal
experiments. (German). Arch. Hyg. (Munich) 31:336-376.
Sarto, F., Cominato, I., Pinto, A.M., Brovedani, P.O., Merler, E., Peruzzi, V., Bianchi,
V., and Levis, A.G. 1984. A cytogenic study on workers exposed to low
concentrations of benzene. Carcinogenesis 5:827-832. (As cited in Dean, 1985.)
Sato, A. etal. 1974. Pharmacokinetics of benzene and toluene. Int. Arch. Arbeitsmed.
33:169-182.
Savolainen, H., Vainio, H., Helojoki, M., and Elovaara, E. 1978. Biochemical and
lexicological effects of short-term, intermittent xylene inhalation exposure and
combined ethanol intake. Arch. Toxicol. 41:195-205.
Savolainen, K., Kekoni, J., Riihimaki, V., and Laine, A. 1984. Immediate effects of m-
xylene on the human central nervous system. Disease. Metabolism and Reproduction
in the Toxic Response to Drugs and Other Chemicals Arch. Toxicol.. Suppl. 7:412-
417.
Sax, N.I. 1984. Dangerous properties of industrial materials. 6th ed. New York: Van
Nostrand Reinhold Company.
Schairer, L., and Sautkulis, R. 1982. Detection of ambient levels of mutagenic
atmospheric pollutants with the higher plant Tradescantia. In Environmental
mutagenesis. carcinogenesis. and plant biology. Vol. II, pp. 154-194. Edited by
Klekowski. New York: Praeger. (As cited in NTP, 1986.)
Schairer, L., Van't Hpff, J., Haynes, C., Burton, R., and deSerres, F. 1978.
Measurement of biological activity of ambient air mixtures using a mobile laboratory for
in situ exposures: preliminary results from the Tradescanitia plant test system. In
Application of short-term bioassays in the fractionation and analysis of complex
environmental mixtures. Edited by M.L.Waters, et al. n.p.: U.S. Environmental
Protection Agency. (As cited in NTP, 1986.)
Schaum J. n.d. Applied techniques for conducting exposure assessments, health and
environmental risk analysis. In press.
R-36
-------
Schell, R.M. 1984. Estimation of the public health risks associated with exposure of
ambient concentrations of 87 substances. Revised February 1985. Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards.
Schiestl, R.H. 1989. Nonmutagenic carcinogens induce intrachromasornal recombination
in yeast. Nature 337:285-288.
Schmidt-Bleek, F.f Peichl, L., Reiml, D., Behling, G., and Muller, K.W. 1988. A
concept for detecting unexpected changes in the environment early. Regulatory
Toxicology and Pharmacology 8:308-327.
Schottenfeld, p., Warschauer, M.E., Zauber, A.G., Meikle, J.G., and Han, B.T. 1981.
A prospective study of morbidity and mortality in petroleum industry employees in the
United States - A preliminary report In Quantification of occupational cancer, pp.
247-265. Banbury Report 9. Edited by R. Peto and M. Schneiderman. Cold Spring
Harbor, NY: Cold Spring Harbor Laboratory.
Schumacher, H., and Grandjean, E. 1960. Comparative investigations on the anesthetic
effect and acute toxicity of nine solvents. Arch. Gewerbepathol. Gewerbehve. 18:109-
119.
Schwartz, E. 1987. Proportionate mortality ratio analysis of automobile mechanics and
gasoline service station workers in New Hampshire. New Hampshire Department of
Health and Human Services, Division of Health Services. American Journal of
Industrial Medicine 12:91-99
Scudder, J., and Anderson, N. 1986. Petroleum contamination of Maine's drinking water
wells. June 1986. Augusta: Maine Bureau of Health, Department of Human Services.
Selling, L. 1916. Benzol as a leucotoxin. Studies on the degeneration and regeneration of
the blood and hematopoietic organs. Johns Hopkins Hosp. Rep. 17:83-142.
Seppalainen, A.M., Husman, K., and M&rtenson, C. 1978. Neurophysiological effects
of long-term exposure to a mixture or organic solvents. Scand. J. Work Environ. &
Heahh 4:304-314.
Seppalainen, A.M., Savolainen, K., and Kovala, T. 1980. Effects of xylene and alcohol
on human visual evoked potentials. 2nd. Int. Congress on Toxicology, Brussels,
Belgium, July 6-11, 1980. Toxicol. Lett. 0:73. (Special issue 1).
Shabecoff, P. 1983. Gasoline releases from storage tanks threaten water supply EPA
says. New York Times. November 30, 1983.
Shah, J., and Singh, H. 1988. Distribution of volatile organic chemicals in outdoor and
indoor air. ES&T22(12):1381.
Shalin, M., and Fournier, F. 1978. Suppression of mutation induction and failure to
detect mutagenic activity with Athabasca tar sand fractions. Mutat. Res. 58:29-34.
(As cited in NTP, 1986)
Shehata, A.T. 1985. A multi-route exposure assessment of chemically contaminated
drinking water. An international symposium on advances in health risk assessment for
systemic toxicants and chemical mixtures. Toxicol. Ind. Health l(4):277-298.
R-37
-------
Shen, T. 1981. Estimating hazardous air emissions from disposal sites. Pollution
Engineering 13(8):31-34.
Shigeta, S., Aikawa, S.H., Misowa, T., and Suzuki, K. 1983. Fetotoxicity of inhaled
xylene in mice. Teratology 28(1):22A.
Shiller, J. 1987. Gasoline vapor exposure parameters during refueb'ng oprations.
Presented at the 80th Annual APCA Meeting, New York, June 1987. Paper 87-31.3.
Silverman, D.T., Hoover, R.N., Albert, S., and Graff, K.M. 1983. Occupation and
cancer of the lower urinary tract in Detroit J. Natl. Cancer Inst. 70:237-245.
Simmon, V., Kauhanen, K., and Tardiff, R. 1977. Mutagenic activity of chemicals
identified in drinking water. Dev. Toxicol. Environ. Sci. 2:249-258. (As cited in
NTP, 1986.)
Siou, G., Conan, L., and El Haitem, M. 1981. Evaluation of the clastogenic action of
benzene by oral administration with 2 cytogenetic techniques in mouse and Chinese
hamster. Mutat. Res. 90:273-278. (As cited in NTP, 1986.)
Sloof, W. 1979. Detection limits of a biological monitoring system based on fish
respiration. Bull. Environ. Contain. Toxicol. 23(4-5^:517-523.
Sloof, W., et al. 1983. Detection limits of a biological monitoring system for chemical
water pollution based on mussel activity. Bull. Environ. Contam. Toxicol. 30:400-
405.
Slooff, W. and Baerselman, R. 1980. Comparison of the usefulness of the Mexian
axoloti (Ambystoma mexicanum) and the clawed toad (xenopus laevis) in lexicological
bioassays. Bull. Environ. Contam. Toxicol. 24:439-443.
Smith, B., Plummer, J., Wolf, C, Philpot, R., and Bend, J. 1982. P-xylene metabolism
by rabbit lung and liver and its relationship to the selective destruction of pulmonary
cytochrome P-450. J. Pharmacol. EXP. Ther. 223:736-42.
Smith, K.H. 1983. Determination of the reproductive effects in mice of nine selected
chemicals. Prepared for the National Institute for Occupational Safety and Health.
Cincinatti, OH: National Institute for Occupational Safety and Health.
Smith, M. 1972. An investigation of passenger car refueling losses. SAE Paper 720931.
n.p.: Society of Automotive Engineers.
Smolik, R., Grzybet Hrynectwics, K., Lange, A. and Zatonski, W. 1973. Serum
complement level in workers exposed to benzene, toluene and xylene. Int. Arch.
Arbeitsmed. 31:243-247.
Smyth, H.F., and Smyth, H.F., Jr. 1928. Inhalation experiments with certain lacquer
solvents. J. Ind. Hvg. 10:261-271.
Smyth, H.F., Jr., Carpenter, C.P., Weil, C.S., Pozzani, U.C., and Striegel, J.A. 1962.
Rangefmding toxicity data: list VI. Am. Ind. Hvg. Assoc. J. 23:95-107.
R-38
-------
Snider, E.A., and Manning, F.S. 1982. A survey of pollutant emission levels in
wastewaters and residuals from the petroleum refining industry. Environment
International 7:237-258.
Snyder, C.A., Goldstein, B.D., Sellakumar, A.R., and Albert, R.E. 1984. Evidence for
hematoxicity and tumorigenesis in rats exposed to 100 ppm benzene. Amer. J. Ind.
Med. 5:429-434.
Snyder, C.A., Goldstein, B.D., Sellakumar, A.R., Bromberg, I., Laskin, S., and Albert,
R.E. 1980. The inhalation toxicology of benzene: incidence of hematopoietic
neoplasms and hematoxicity in AKR/J and C57BL/6J mice. Toxicol. Appl.
Pharmacol. 54:323-331.
Snyder, C.A., Goldstein, B.D., Sellakumar, A.R., Bromberg, I., Laskin, S., and Albert,
R.E. 1982a. Toxicity of chronic benzene inhalation: CD-I mice exposed to 300 ppm.
Bull. Env. Contam. Toxicol. 29:385-391.
Snyder, C.A., Goldstein, B.D., Sellakumar, A., Wolman, S.R., Bromber, I.,
Ehrlichman, M.N., and Laskin, S. 1978. Hematotoxicity of inhaled benzene to
Sprague-Dawley rats and AKR mice at 300 ppm. Bull. Env. Contam. Toxicol.
29:385-391.
Snyder, R., and Kocsis, J.J. 1975. Current concepts of chronic benzene toxicity. CRC
Grit. Rev. Toxicol. 3:265-288.
Snyder, R., Longacre, S.L., Witner, CM., and Kocsis, J.J. 1982b. Metabolic correlates
of benzene toxicity. Adv. Exp. Med. Biol. 136A:245-256.
Snyder, R., Longacre, S.L., Witner, C.M., Kocsis, J.J., Andrews, L.S., and Lee, E.W.
1981. Biochemical toxicology of benzene. Rev. Biochem. Toxicol. 3:123-153.
Snyder, W., Cook, M., Nasset, E., Karhausen, L., Howells, G., and Tipton, I. 1975.
Report of the task group on reference man. In International commission on radiological
protection papers. New York: Pergamon Press.
Spencer, P. 1982. Histopathological analysis of spinal cord of animals exposed to
gasoline phase 1: Single sections study performed at the Institute of Neurotoxicology,
New York. Submitted to the American Petroleum Institute, October 5,1982.
Spencer, P.S. 1983. Experimental evaluations of selected petrochemicals for subchronic
neurotoxic properties. In Advances in modern environmental toxicology. Vol. VI.
Applied toxicology of petroleum hydrocarbon, pp. 199-211. Edited by M.A.
Mehlman. Princeton, NJ: Princeton Scientific Publishers, Inc.
Srbova, J.J., Teisinger, and Skramovsky. 1950. Absorption and elimination of inhaled
benzene in man. Arch. Ind. Hvg. 2:1-8.
SRI. 1982. SRI International. Chemical economics handbook. Section 583:04001.
Menlo Park, CA: SRI International Journal.
Sridharan, B.N., Batra, B.K., and Savkur, L.D. 1963. The effect of a carcinogen on the
progeny of treated mice. Indian J. Pathol. Bacteriol. 6:26-33. (As cited in IARC,
1982.)
R-39
-------
Steele, R., and Whilhelm, D. 1966. The inflammatory reaction in chemical injury. I.
Increased vascular permeability and erythema induced by various chemicals. British J.
EXD. Pathol. 47:612-23.
Stemhagan, A., Slade, J., Altman, R., and Bill, J. 1983. Occupational risk factors and
liver cancer. A retrospective case-control study of primary liver cancer in New Jersey.
Am. J. Epidemiol. 117:443-454.
Sterner, J.H. 1941. Study of hazards in spray painting with gasoline as a dilutent. J. Ind.
Hve. Toxicol. 23:437-448.
Stevens, J.B., and Callaway, J.L. 1940. Mixed epithelioma of the back arising from daily
application of a phenol and ergot ointment. Am J Cancer 38:364-366. (As cited in
NIOSH, 1976.)
Struhsaker, J.W., et al. 1974. Effects of benzene (a water soluble component of crude
oil) on eggs and larvae of Pacific herring and northern anchovy. Poll, and Phvs. of
Marine Organisms.
Styles, J.A., and C.R. Richardson. 1984. Cytogenetic effects of benzene: dosimetric
studies on rats exposed to benzene vapour. Mutation Res. 135:203-209. (As cited in
Dean, 1985b.)
Suffolk Country Department of Health Services. 1989. Effects of pas spills on indoor air
quality in neighboring homes. (Draft Report) Farmingville, New York: Suffolk
Country Department of Health Services.
Sukhanova, V.A., Makareva, L.M., and Bocko, V.S. 1969. Investigation of functional
properties of leukocytes of workers engaged in the manufacture of xylene. Hyg. Sanit.
34:448-450.
Susten, A., Dames, B., Burg, J., and Niemeier, R. 1985. Percutaneous penetration of
benzene in hairless mice: an estimate of dermal absorption during tire-building
operations. American Journal of Industrial Medicine 7:323-35.
Swinyard, E.A. 1975. Noxious gases and vapors. Carbon monoxide, hydrocyanic acid,
benzene, gasoline, kerosene, carbon tetrachloride, and miscellaneous organic solvents.
In Parmocol. Basis Ther.. 5th ed.: pp. 900-911.
Syjrovadko, O.N. 1977. Working conditions and health status of women handling
organosiliceous varnishes containing toluene. Gig. Tr. Prof. Zabol. 12:15-19. (Cited
in U.S. EPA, 1983b.)
Symms, K.G. 1986. Approximation of the inhalation exposure to volatile organic
chemicals from showering with contaminated household water. Presented at The 1986
ASCE Annual Convention and Geotech IV Conference. Boston, Massachusetts.
Tabershaw, I.R., and Lamm, S.H. 1977. Benzene and leukemia. Lancet 33:867-869.
Tangley, L. 1984. Ground water contamination: local problems become national issues.
Bioscience 34(3):142-146,148.
R-40
-------
Tanooka, H. 1977. Development and application of Bacillus subtilis test systems for
mutagens, involving DNA-repair deficiency and supprcssible auxitrophic mutations.
Mutat. Res. 42:19-32. (As cited in NTP, 1986.)
Tatrai, E., Hudak, A.,Barcza, G., and Ungvary, G. 1979. Embryotoxic effect of m-
xylene. Eeeszlsegtudomanv 23:147-151.
Tatrai, E., Rodics, K., and Ungvary, G. 1980. Embryotoxic effects of simultaneously
applied exposure of benzene and toluene. Folia Morphol. 28:286-289.
Tatrai, E., Ungvary, G., and Cseh, I.R. 1980. The effect of long-term inhalation of o-
xylene on the liver. Acta. Med. Acad. Sci. Hung. 37:211-216.
Tatrai, E., Ungvary, G., and Cseh, I.R. 1981. The effect of long-term inhalation of o-
xylene on the liver. In Ind. Environ. Xenobiotics, Proc. Int. Conf., Meeting Date
1981. pp. 293-300.
Tewari, Y.B., Miller, M.M., Wasik, S.P., and Martire, D.E. 1982. Aqueous solubility
and octanol-water partition coefficient of organic compounds at 25.0°C. J. Chem. Eng.
Data 27:451-454.
Theriault, G., and Goulet, L. 1979. A mortality study of oil refinery workers. J. Occup.
Med. 21:367-370.
Thibodeaux, L.J. 1981. Estimating the air emissions of chemicals from hazardous waste
landfills. Journal of Hazardous Materials (4).
Thibodeaux, L.J., and Hwang, S.T. 1982. Landfarming of petroleum wastes - modeling
the air emission problem. Environmental Progress 1(1).
Thomas, F.B., Haider, C.A., Holdsworth, C.E., and Cockrell, B.Y. 1984a.
Hydrocarbon nephropathy in male rats. Pan II. Temporal and morphological
characterization of the renal lesions. Abstract of work performed at the Chemical
Industry Institute of Toxicology (CIIT), submitted to C.E. Holdsworth (American
Petroleum Institute) by M.W. Kloss (CIIT) on October 26, 1984.
Thomas T.L., Decoufle, P., and Moure-Eraso, R. 1980. Mortality among workers
employed in petroleum refining and petrochemical plants. J. Occup. Med. 22:97-103.
Thomas, T.L., Waxweiler, R.J., Crandall, M.S., White, D.W., Moure-Eraso, R., and
Fraumeni, J.F., Jr. 1984b. Cancer mortality patterns by work category in three Texas
oil refineries. Am. J. Ind. Med. 6:3-6.
Thomas, T.L., Waxweiler, R.J., Crandall, M.S., White, D.W., Moure-Eraso, R., Itaya,
S., and Fraumeni, J.F., Jr. 1982a. Brain cancer among OCAW members in three
Texas oil refineries. Ann. N.Y. Acad. Sci. 381:120-129.
Thomas, T.L., Waxweiler, R.J., Moure-Eraso, R., Itaya, S., and Freumeni, J.F., Jr.
1982b. Mortality patterns among workers in thre Texas oil refineries. J. OCCUP. Med.
24:135-141.
Tice, R., Costa, D., and Drew, R. 1980. Cytogenetic effects of inhaled benzene in
murine bone marrow. Induction of sister chromatid exchanges, chromosomal
R-41
-------
aberrations, and cellular proliferation inhibition in DBA/2 mice. Proc. Nat'l. Acad.
Sci. USA 77:2148-2152. (As cited in NTP, 1986.)
Tice, R., Fogt, T., and Costa, D. 1982. Cytpgenetic effects of inhaled benzene in murine
bone marrow. In Genotoxic effects of airborne agents, pp. 257-275. Edited by R.
Tice, D. Costa, K. Schaich. New York: Plenum Press.
Tice, R.R., Sawey, M.J., Drew, R.T, and Cronkite, E.P. 1984. Benzene-induced
micronuclei in the peripheral blood of mice: a retrospective analysis. Environ.
Mutagen. 6:421. (As cited in Dean, 1985b.)
Tironi, G., Nebel, G., and Williams, R. 1986. Measurement of vapor exposure during
gasoline refueling. SAE Paper 860087. Presented at Society of Automotive Engineers
Inc., International Congress. Detroit, Michigan. February 1986.
Toft, I.K., Olofsson, T., Tunek, A., and Berlin. M. 1982. Toxic effects on mouse bone
marrow caused by inhalation of benzene. Arch. Toxicol. 51:295-302.
Toftgard, R., and Gustafsson, J.-A. 1980. Biotransformation of organic solvents: a
review. Scand. J. of Work. Env.. and Health 6:1-18.
Toftgard, R., and Nilsen, O. 1982. Effects of xylene and xylene isomers on cytochrome
P-450 and in vitro enzymatic activities in rat liver, kidney and lung. Toxicology
23:197-212.
Toftgard, R., Nilsen, O., Glaumann, H., and Gustasson, J.-A. 1983. Induction of
cytochrome P-450 in the rat liver after exposure to xylenes-dose-response relationship
and dependence on endocrine factors. Toxicology 27:119-37.
Topham, J. 1980. Do induced sperm-head abnormalities in mice specifically identify
mammalian mutagens than carcinogens? Mutat. Res. 74:379-387. (As cited in NTP,
1986.)
Travis, C, et al. 1987. Cancer risk management Environmental Science &Technologv
21 (5): 415.
Trump, B.F., Lipsley, M.M., Jones, T.W., Heatfield, B. M., Higginson, J., Endicott,
K., and Hess, H.B. 1984. An evaluation of the significance of experimental
hydrocarbon toxicity to man. In Advances in modem environmental toxicology. Vol.
VII. n.p.: n.p.
Tsai, S.P., et al. 1983. Retrospective mortality and medical surveillance studies of
workers in benzene areas of refineries. J. Occup. Med. 25:685-592.
Tsuruta, H. 1982. Percutaneous absorption or organic solvents in, on the penetration
rates of hydrophobia solvents through the excised rat skin. Industrial Health 20:335-
45.
Tunek, A., Hogstedt, B., and Olofsson, T. 1982. Mechanism of benzene toxicity.
Effects of benzene and benzene metabolites on bone marrow cellularity, number of
granulopoietic stem cells and frequency of micronuclei in mice. Chem. Biol. Interact.
39:139-148.
R-42
-------
Tunek, A., Schelin, G, and Jergil, B. 1979. Microsomal target proteins of metabolically
activated aromatic hydrocarbons. Chem-Biol. Interactions 27:133-144.
Turner, D.B. 1970. Workbook of atmospheric dispersion estimates. Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Air Programs.
Ungvary, G., et al. 1983. Chemical-induced changes in hepatic vascular innervation.
2.Mikrosk Anat. Forsch 97:823-828. (Abstract)
Ungvary, G., Tatrai, E., Hudak, A., Barcza, G., and Lorincz, M. 1980. Studies on the
embryotoxic effects of ortho-, meta-, and para-xylene. Toxicology 18:61-74.
Ungvary, G., Varga, B., Horvath, E., Tatrai, E., and Folly, G. 1981. Study on the role
of maternal sex steriod production and metabolism in the embryo-toxicity of p-xylene.
Toxicology 19(3):263-268.
U.S. Department of Health and Human Services (U.S. DHHS). 1983. Third annual
report on carcinogens. Summary. September 1983. Research Triangle Park, NC:
Public Health Service, U.S. Department of Health and Human Services.
U.S. Department of Energy (U.S. DOE). 1986. Citations from the U.S. Department of
Energy Database: Gasoline fuel additives. Washington, DC: Energy Information
Administration.
U.S. Environmental Protection Agency (U.S. EPA). 1970. Investigation of passenger car
refueling losses. U.S. Environmental Protection Agency, Office of Air and Water
Programs.
. 1972. Investigation of passenger car refueling losses, second year proeram.
U.S. Environmental Protection Agency, Office of Air and Water Programs.
. 1975a. Expansion of investigation of passenger car refueling losses: third
year program. U.S. Environmental Protection Agency, Office of Air and Water
Programs.
. 1975b. Sampling and analysis of selected toxic. Toxic substances, task II-
ethvlene dibromide. U.S. Environmental Protection Agency, Office of Toxic
Substances.
. 1978a. Assessment of health effects of benzene germane to low-level
exposure. U.S. Environmental Protection Agency.
. 1978b. Assessment of human exposures to atmospheric benzene. U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards.
. 1978c. Bioassays on aquatic organism to 93 chemical compounds. U.S.
Environmental Protection Agency.
.. 1979a. Ambient water quality criteria document for benzene. Washington,
D.C.: U.S. Environmental Protection Agency.
. 1979b. Carcinogen assessment group's final report on population risk to
ambient benzene exposures. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards.
R-43
-------
.. 1979c. Transportation and marketing of petroleum liquids. In Compilation of
air pollutant emmission factors. Research Triangle Park, NC: U.S. Environmental
Protection Agency.
. 1980a. Ambient water criteria for benzene. Washington, DC: Environmental
Protection Agency.
_. 1980b Benzene health and environmental effects. Washington, DC: U.S.
Environmental Protection Agency.
. 1980c. Northeast corridor regional modeling project - Detemination of organic
species profies for gasoline liquids and vapors. Abstract. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards.
.. 1980d. Water quality criteria documents. Federal Register. Part V. Nov. 28,
1980. 79319-79379.
_. 1980e. Water quality criteria for toluene. Washington, DC: Office of Drinking
Water.
. 1981a. Handbook for performing exposure assessments. Draft Report. U.S.
Environmental Protection Agency, Office of Health and Environmental Assessment
. 1981b. Health effects advisory for benzene. U.S. Environmental Protection
Agency, Office of Drinking Water.
. 1983a. Hazard profile for benzene. Cincinnati, OH: U.S. Environmental
Protection Agency, Environmental Criteria and Assessment Office.
. 1983b. Health assessment document for toluene. Washington, DC: U.S.
Environmental Protection Agency, Office of Health and Environmental Assessment
. 1984a. Estimation of the public health risk form exposure to gasoline vapor
via the gasoline marketing system. U.S. Environmental Protection Agency, Office of
Air Quality Planning and Standards.
. 1984b. Evaluation and selection of models for estimating air emissions from
hazardous waste treatment, storage, and disposal facilities. U.S. Environmental
Protection Agency.
. 1984c. Evaluation of air pollution regulatory strategies for gasoline marketing
industry. Washington, DC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards and Office of Mobile Sources.
. 1984d. Health effects assessment for benzene. Cincinnati, OH: U.S.
Environmental Protection Agency, Environmental Criteria and Assessment Office.
. 1984e. Regulatory strategies for the gasoline marketing industry. Notice of
availability of a regulatory strategies analysis document for public comment.
49FR31706. Federal Register. August 8.1984.
R-44
-------
. 1985a. Development of statistical distributions or ranges of standards factors
used in exposure assessments. U.S. Environmental Protection Agency, Office of
Health and Environmental Assessment.
. 1985b. Drinking water criteria document for benzene. Final draft.
Washington, DC: U.S. Environmental Protection Agency, Office of Drinking Water.
. 1985c. Drinking water criteria document for ethvlbenzene. Washington, DC:
U.S. Environmental Protection Agency, Office of Drinking Water.
. 1985d. Drinking water criteria document for gasoline. Draft report
Washington, DC: U.S. Environmental Protection Agency, Environmental Criteria and
Assessment Office.
. 1985e. Hazardous waste management system. Identification and listing of
hazardous waste. 40 CRF part 261, Appendix I. Federal Register 50:48886-48966.
. 1985f. Study of gasoline volatility and hydrocarbon emissions from motor
vehicles. Ann Arbor, MI: U.S. Environmental Protection Agency, Office of Mobile
Sources.
. 1985g. The air toxics problem in the United States: An analysis of cancer
risks for selected pollutants. U.S. Environmental Protection Agency, Office of Air and
Radiation and Office of Policy, Planning and Evaluation.
. 1985h. The endangarment assessment handbook. U.S. Environmental
Protection Agency, Office of Waste Programs Enforcement
. 1985L Toxicology handbook. Principals related to hazardous waste site
investigations. U.S. Environmental Protection Agency, Office of Waste Programs
Enforcement.
. 1985J. Underground motor fuel storage tanks: a national survey. Volume 1 -
Technical report. U.S. Environmental Protection Agency, Office of Pesticides and
Toxic Substances.
. 1986a. Compilation of air pollutant emission factors. Volume I: Stationary
point and area sources (4th Edition with Supplement A). Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
. 1986b. Draft final report of the Philadelphia Integrated Environmental
Management Project. U.S. Environmental Protection Agency, Office of Policy
Analysis.
. 1986c. Evaluation of air pollution regulatory strategies for gasoline marketing
industry. Response to public comments. July 30, 1986. Washington, DC: U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards.
1986d. Hazardous waste tank failure model: description of methodology.
Draft report Submitted by Pope-Reid Associates, Inc. U.S. Environmental Protection
Agency, Office of Solid Waste.
. 1986e. Hazardous waste tank risk analysis. Draft report. U.S. Environmental
Protection Agency, Office of Solid Waste.
R-45
-------
. 1986f. Health and environmental effects profile for xylenes (o-. m-. p-) (Final
draft). U.S. Environmental Protection Agency, Office of Health and Environmental
Assessment.
. 1986g. Summary of state reports on releases from underground storage tanks.
U.S. Environmental Protection Agency, Office of Underground Storage Tanks.
. 1986h. Superfund exposure assessment manual. Draft. U.S. Environmental
Protection Agency, Office of Emergency and Remedial Response.
. 1987a. The risk assessment guidelines of 1986. Washington, DC: U.S.
Environmental Protection Agency, Office of Health and Environmental Assessment.
.. 1987b. Toxicolopical profile. Benzene, (first draft). No.2-78. April 29,
1987. Washington, DC: U.S. Environmental Protection Agency.
. 1987c. Underground storage tanks: technical requirements. Proposed rule.
74 Federal Register 12662. April 17,1987.
.. 1987d. Evaluation of the carcinogenicity of unleaded gasoline. Washington,
DC: U.S. Environmental Protection Agency.
. 1988. Drinking water criteria document for gasoline. Final draft. Cincinnati,
OH: U.S. Environmental Protection Agency, Environmental Criteria and Assessment
Office, Office of Health and Environmental Assessment
Uyeki, E.M., Ashkar, A.E., Shoeman, D.W., and Bisel, T.U. 1977. Acute toxicity of
benzene inhalation to hematopoietic precursor cells. Toxicol. Appl. Pharmacol. 40:49-
Valpey, R., Sumi, S.M., Copass, M.K., and Coble, G.J. 1978. Brief communications.
Acute and chronic progressive encephalopathy due to gasoline sniffing. Neurology
507-510.
Van Gelder-Othway, S. 1976. Some physical and biological effects of oil films floating
on water. In Marine ecology and oil pollution, pp.255-268. Edited by J.M. Baker.
New York: John Wiley & Sons.
Van Raalte, H.G.S., and Grasso, P. 1982. Hematological, myelotoxic, clastogenic,
carcinogenic, and leukemogenic effects of benzene. Reel. Toxicol. Pharmacol. 2:153-
176.
Veith, G.D., Foe, D.L., and Bergstedt, B.S. 1979. Measuring and estimating the
bioconcentration factor of chemicals in fish. J. Fish. Res. Board Can. 36:1040-1048.
Venit, S. 1985. Summary report of the performance of the bacterial mutation assays. In
Evaluation of short-term tests for carcinogens: report of the international program on
chemical safety collaborative study on in vitro assays, pp. 11-23. Edited by J. Ashby,
F.J. de Series, M. Draper, M. Ishidate, Jr., B.H. Margolin, B.E. Matter, and M.D.
Shelby. Amsterdam: Elsevier Publishing Company. (As cited in Dean, 1985b.)
R-46
-------
Vermont Department of Water Resources and Environmental Engineering. 1987.
Summary of Hardwick Water Supply/Kwik Shop Investigation. Montpelier, VT:
Department of Water Resources and Environmental Engineering.
Versar American Management Systems, Inc. 1984. Hazardous air pollutants: An
exposure and risk assessment for 35 counties, n.p.: Versar American Management
Systems, Inc.
Verschueren, K. 1983. Handbook of environmental data on organic chemicals. 2nd ed.,
pp. 1188-1193. New York: Van Nostrand Reinhold Co.
Vigliani, E.G. 1976. Leukemia associated with benzene exposure. Ann. NY. Acad. Sci.
271:143-151.
Vigliani, E.C., and Fomi, A. 1976. Benzene and leukemia. Environ. Res. 11:122-127.
Vigliani, E.G., and Saita, G. 1964. Benzene and leukemia. New Ene. J. Med. 271:872-
876.
Von Burg, R. 1982. Toxicology updates. Xylene. J. Appl. Toxicol. 2(5):269-271.
Von Euler, G., Fuxe, K., Hansson, T., Eneroth, P., and Gustafsson, J.A. 1989.
Persistent effects of neonatal toluene exposure on regional brain catecholamine levels
and turnover in the adult male rat. Toxicology 54:1-16.
Von Euler, G., Fuxe, K. Hansson, T., Ogren, S.-O. Agnati, L.F., Eneroth, P.,
Harfstrand, A., and Gustafsson, J.-A. 1988. Effects of chronic toluene exposure on
central monoamine and peptide receptors and their interactions in the adult male rat
Toxicology 52:103-126.
Von Oettingen, W.F., Neal, P.A., Donahue, D.D., Svirbely, J.L., Baernstein, H.D.,
Monaco, A.R., Valaer, P.J., and Mitchell, J.L. 1942. The toxicity and potential
dangers of toluene with special reference to its maximal permissible concentration.
U.S. Public Health Serv. Pul. Health Bull. No. 279.
Vozovaya, M. A. 1974. Gynecological illnesses in workers of major industrial rubber
products plant occupations. Gig. Tr. Sostov. Spetsif. Funkts. Rab. Neftekhim. Khim.
Promsti. 1974:561. (Russian, Abstract.)
Vozovaya, M. A. 1975a. Effect of low concentrations of gasoline, dichloroethane, and
their combination on the generative function of animals and development of the progeny
of two generations. (Abstract). Gig. Tr. Prof. Zabol. (6):20-23.
Vozovoya, M.A. 1975b. The effect of small concentration of benzene, dichloroethane
alone and combined, on the reproductive function of animals and the development of
the progeny. Giqiena Truda I Professionarnve Zabolevaniva 7:20-23.
Vozovoya, M.A. 1976a. The effect of low concentrations of benzene and dichloroethane
separately and combined on the resproductive function of animals. Fiaiena i Sanitariva
7:100-102.
Vozovaya, M.A. 1976b. The effect of low concentrations of gasoline, dichloroethane,
and their combination on the reproductive fucntion of animals. (Abstract). Giq. Sanit.
(6): 100-102. (Russian).
R-47
-------
Wahlberg, J.E. 1984. Erythema-inducing effects of solvents following epicutaneous
administration to man - studied by laser Doppler flowmetty. Scand. J. Work Environ.
Health 10: 159-162.
Walton, W.C. 1985. Practical aspects of groundwater modeling. Worthington, Ohio:
National Water Well Association.
Wambsgas, D.E. 1986. State programs for controlling ozone-II. Presented at The 1986
Washington conference on ozone control strategy. Arlington. Va.
Wang, C.C., and Irons, G.V. 1961. Case reports. Acute gasoline intoxication. Archives
of Environmental Health 2:114-116.
Ward, C.O., Kuna, R.A., Synder, N.K., Alasker, R.D., Coate, W.B., and Craig, P.H.
1985. Cytogenetics and cytokinetics of cultured lymphocytes from benzene-exposed
workers. Int. Arch. Occup. Environ. Health 46:31-41. (As cited in Dean, 1985b.)
Watanabe, T., Endo, A., Kato, Y., Shima, S., Watanabe, T., and Ikeda, M. 1980.
Cytogenetics and cytokinetics of cultured lymphocytes from benzene-exposed workers.
Int. Arch. Occup. Environ. Health 46:31-41. (As cited in Dean, 1985b.)
Wen, C.P. 1983. Epidemiological studies of kidney cancer with particular reference to
gasoline exposure. In Proceedings of the workshop on kidnev effects of
hydrocarbons. July 18-20.1983. pp. 377-392. Published by the American Petroleum
Institute, December 1983, and submitted to EPA as FYI-AX-0184-028054.
Wen, C.P., Tsai, S.P., and Gibson, R.L. 1982. A report on brain tumors from a
retrospective cohort study of refinery workers. Ann. N.Y. Acad. Sci. 381:130-138.
Wen, C.P., Tsai, S.P., Gibson, R.L., and McClellan, W.A. 1984a. Long-term mortality
of oil refinery workers. II. Comparison of experience of active, terminated, and
retired workers. J. Occup. Med. 26:118-127.
Wen, C.P., Tsai, S.P., Moffitt, K.B., Bondy, M., and Gibson, R.L. 1984b. Summary.
Epidemiologic studies of the role of gasoline hydrocarbon exposure in kidney cancer
risk. In Renal effects of petroleum hydrocarbons: Advances in modern environmental
toxicology. Vol. VII, pp. 245-257. Edited by M.A. Mehlman. Princeton, NJ:
Princeton Scientific Publishers, Inc.
Wen, C.P., Tsai, S.P., Orkon, R.L., Weiss, N.S., Wong, O.W., and McClellan, W.A.
1984c. Long-term mortality study of oil refinery workers. J. North. Cancer. Inst.
(Citation unavailable).
Wen, C.P., Tsai, S.P., Weiss, N.S., McClellan, W.A., and Gibson, R.L. 1981. A
population-based cohort study of brain tumor mortality among oil refinery workers
with a discussion of methodological issues of SMR and PMR. In Quantification of
occupational cancer, pp. 413-432. Edited by R. Peto and M. Schneiderman. Cold
Spring Harbor, NY: Cold Spring Harbor Laboratory.
Westpath. 1980. Histopathologic evaluation of tissues and organs from B6C3Fj mice in a
two-year motor fuel chronic inhalation study, 12 month serial sacrifice and 6-12 month
R-48
-------
data. Conducted by Westpath Laboratories, Inc. Co. for International Research and
Development Corporation, MI.
Whittemore, I.M. 1979. High resolution gas chromatography of the gasolines and
naphthas. In Chromatoeraphv in petroleum analysis, pp. 41-74. Edited by K.H.
Altgelt and T.H. Gouw. New York: Marcel Dekker, Inc.
Wigle, D.T. 1977. The distribution of lung cancer in two Canadian cities. Can. J. Public
Health 68:463-468.
Williams, G.W., Ton, C, and Brat, S.V. 1985. Tests with the rat hepatocyte primary
culture/DMA repair test In Evaluation of short-term tests for carcinogens: report of the
international programs on chemical safety collaborative study on in vitro assays, pp.
341-345. Edited by J. Ashby, F.J. de Serres, M. Draper, M. Ishidate, Jr., B.H.
Margolin, B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier. (As cited in Dean,
1985b.)
Williams, R. 1987. Breathing zone levels of gasoline vapors during automobile refueling.
GMR-5873. June 1987. Warren, MI: General Motors Research Laboratories.
Wilson, C.B. 1983. Immunopathologic studies in mice and rats chronically exposed to
hydrocarbon vapors. Draft report, pp. 1-40. Supported in part by the American
Petroleum Institute.
Wilson, R. 1985. Harvard University review of "Lifetime risks of kidney cancer
estimated from animal and epidemiologic studies" by P.E. Enterline. Prepared for the
American Petroleum Institute.
Wilson, R.H. 1943. Toluene poisoning. J. Amer. Med. Assoc. 123:1106.
Windholz, M. 1983. The Merck index. 10th ed. pp. 151,1364,1447-1448. Rahway,
NJ: Merck & Co., Inc.
Winek, C.L., and Collom, W.D. 1971. Benzene and toluene fatalities. J. Occ. Med.
13:259-261.
Woiwode, W., and Drysch, K. 1981. Experimental exposure to toluene: further
consideration of cresol formation in man. British J. of Ind. Med. 38:194-7.
Wol, M.A., Rowe, V.K., McCollister, D.D., Hollingsworth, R., and Oyen, F. 1956.
Toxicological studies of certain alkylated benzenes and benzene experiments on
laboratory animals. AMA Arch. Ind. Health 14:387-398.
Wolf, M.A., Rowe, V.K., McCollister, D.D., Hollingsworth, R.L., and Oyen, F. 1956.
Toxicological studies of certain alkylated benzenes and benzene. Experiments on
laboratory animals. AMA Arch. Ind. Health 14:387-398.
Wong, O. 1985. Environmental Health Associates review of "Lifetime risks of kidney
cancer estimated form animal and epidemiologic studies" by P.E. Entraline. Prepared
for Mobil Oil Corporation.
Wong, O., Margan, R.W., and Whorton, M.D. 1983. Comments on the NIOSH study of
leukemia in benzene workers. Technical report submitted to Gulf Canada, Ltd. by
Environmental Health Associates, August 31,1983.
R-49
-------
World Health Organization (WHO). 1981. Recommended health based limits in
occupational exposures to selected organic solvents. Geneva, Switzerland: United
Nations World Health Organization. WHO Technical Report Series 664:7-24.
Wurger, F.E., Graf, U., and Frei, H. 1985. Somatic mutation and recombination test in
wings of Drosophila melanogaster. In Evaluation of short-term tests for carcinogens:
report of the international program on chemical safety collaborative study on in vitro
assays, pp. 325-340. Edited by J. Ash by, F.J. de Serres, M. Draper, M. Ishidate,
Fr., Margolin, B.E. Matter, and M.D. Shelby. Amsterdam: Elsevier Publishing
Company. (As cited in Dean, 1985b.)
Zhibura, L.P. 1974. Effect of gasoline vapors on the state of reproductive functions in
young female workers. (Abstract) Giq. Tr. Sostov. Spetsif. Funkts. Rab. Neftekhim.
Khim. Promsti. p.62-66. (Russian)
R-50
------- |