United States
Environmental Protection
Agency
Office of Health and
Environmental Assessment
Washington DC 20460
EPA-600/8-83-032F
December 1984
Final Report
&EPA
Research and Development
Health Assessment
Document for
Epichlorohydrin
Final
Report
-------
-------
EPA-600/8-83-032F
December 1984
Final Report
Health Assessment Document
for
Epichlorohydrin
Final Report
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Research Triangle Park, NC 27711
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
i i
-------
PREFACE
The Office of Health and Environmental Assessment has prepared this
health assessment to serve as a "source document" for Agency-wide use.
The health assessment document was originally developed at the request
of the Office of Air Quality Planning and Standards; however, the scope
of the assessment has. since been expanded to address multimedia aspects.
This assessment will help ensure consistency in the Agency's consideration
of the relevant scientific health data associated with epichlorohydrin.
In the development of the assessment document, the scientific
literature has been inventoried, key studies have been evaluated and
summary/conclusions have been .prepared so that the chemical's toxicity
and related characteristics are qualitatively identified. Observed
effect levels and other measures of dose-response relationships are
discussed, where appropriate, so that the nature of the adverse health
responses are placed in perspective with observed environmental levels.
ili
-------
-------
CONTENTS
Page
PREFACE .................... ...................................... iji
LIST OF TABLES .................. ................................. xi
LIST OF FIGURES . . . ................................................ xivl
AUTHORS, CONTRIBUTORS, AND REVIEWERS . . . . ......................... xv
1. EXECUTIVE SUMMARY .......... ................... . ............ . . 1-1
1. 1 BACKGROUND INFORMATION ..... ............... ............. 1-1
1. 1. 1 Properties ...................................... 1-1
1.1.2 Production ................... . ........... 1-1
1.1.3 Use .................................... . ....... ; 1-1
1.1.4 Environmental Release, Transport, and Fate ..... . 1-1
1.4. 5 Environmental Transformation .................... , 1-2
1. 2 UPTAKE, METABOLISM, AND EXCRETION ...................... 1-2
1.3 EFFECTS ON HUMANS ........... ...... , .................... 1-2
1.4 ANIMAL TOXICITY ......... ................... ............ 1-3
1.5 CARCINOGENICITY, MUTAGENICITY, AND REPRODUCTIVE
AND TERATOGENIC EFFECTS ........................ . ....... 1-3
1. 5. 1 Carcinogenicity ................................. 1-3
1.5.2 Mutagenicity ........ ............. ....... ........ 1-3
1.5.3 Reproductive and Teratogenic Effects ____ ........ 1-4
1. 6 SYNERGISM AND ANTAGONISM ................. ...... „ ....... 1-4
1.7 ECOSYSTEMS AND AQUATIC BIOTA ........................... 1-4
1.8 REGULATIONS AND STANDARDS .............................. 1-5
1.9 CONCLUSIONS ................. . .......................... 1-5
1. 10 RESEARCH NEEDS ......................................... 1-5
2. INTRODUCTION . . . . ........... . ................................. 2-l
3. BACKGROUND INFORMATION . .. ..................................... 3-1
3. 1 PHYSICAL AND CHEMICAL PROPERTIES ... ...... , ............. 3-1
3.1. 1 Introduction ..................... . .............. 3-1
3. 1. 2 Synonyms and Trade Names .............. .......... 3-1
-------
CONTENTS (Cont'd.)
3.1.3 Identification Numbers 3-1
3.1.4 Significance of Physical Properties with
Respect to Environmental Behavior 3-1
•3.1.5 Chemical Reactions 3-3
3.1.6 Chemical Reactions in the Environment 3-4
3.1.6.1 Hydrolysis and Related Reactions 3-4
3.1.6.2 Oxidation 3-9
3.1.6.3 Photolysis 3-10
3.2 ANALYTICAL METHODOLOGY 3-10
3.2.1 Introduction 3-10
3.2.2 Chemical Analysis in Air 3-11
3.2.3 Chemical Analysis in Water 3-12
3.3 PRODUCTION, USE, AND RELEASES TO THE ENVIRONMENT 3-13
3.3.1 Introduction 3-13
3.3.2 Production 3-13
3.3.3 Use 3-14
3.3.3.1 Synthetic Glycerine 3-15
3.3.3.2 Epoxy Resins 3-15
3.3.3.3 Textiles 3-15
3.3.3.4 Paper, Inks, and Dyes 3-16
3.3.3.5 Anion Exchange Resins 3-16
3.3.3.6 Solvents 3-16
3.3.3.7 Surface Active Agents 3-16
3.3.3.8 Epichlorohydrin-based Rubber
Elastomers 3-17
3.3.3.9 Starch Modifier 3-17
3.3.3.10 Other Current Uses 3-18
3.3.3.11 Proposed Uses 3-18
3.3.4 Substitute Chemicals/Processes 3-18
3.3.5 Environmental Release 3-19
3.3.6 Environmental Occurrence 3-20
3.4 ENVIRONMENTAL TRANSPORT AND FATE 3-20
3.4.1 Transport 3-20
3.4.1.1 Volatilization 3-20
3.4.1.2 Sorption 3-21
3.4.1.2.1 Soils 3-21
3.4.1.2.2 Sediments 3-21
3.4.2 Fate ". 3-22
3.4.2.1 Chemical and Physical Process 3-22
3.4.2.2 Biological Processes ... 3-22
3.5 SUMMARY •- 3-24
4. COMPOUND DISTRIBUTION AND RELATED PHARMACOKINETICS
IN HUMANS AND ANIMALS 4-1
VI
-------
CONTENTS (Cont'd.)
4.1 ROUTES OF EXPOSURE AND ABSORPTION 4-1
4.2 DISTRIBUTION 4_2
4,3 METABOLITE IDENTIFICATION AND PATHWAYS 4-4
4.4 EXCRETION a 4_g
4.5 SUMMARY .. 4_9
5. EFFECTS ON HUMANS 5^
5.1 EPIDEMIOLOGIC STUDIES 5-!
5.2 EFFECTS ON THE NERVOUS SYSTEM 5.3
5.3 EFFECTS ON BLOOD AND HEMATOPOIETIC TISSUE 5-3
5.3.1 Erythrocytes and Leukocytes 5-3
5.3.2 , Peripheral Lymphocytes .' 5-3
5.3.3 Immunocompetence 5-5
5.4 EFFECTS ON THE LIVER 5.5
5.5 EFFECTS ON THE SKIN 5_6
5.5.1 Case Studies 5_g
5.5.2 Sensitization !!...!...!!!!.! 5-8
5.6 SUMMARY 5.9
6. ANIMAL TOXICOLOGY g.-^
6.1 SPECIES SENSITIVITY g^
6.1.1 Acute Toxicity .... g.-^
6.1.1.1 Inhalation c_i
6.1.1.2 oral '...!!.'!!!!!."!!!!!.": ' 5-9
6,1.1.3 Subcutaneous Injection 6-10
6.1.1.4 Intraperitoneal Injection ....!" 6-12
6.1.1.5 Intraveneous Injection '" 6-12
6.1.1.6 Percutaneous Application 6-13
6.1.2 Subchronic and Chronic Toxicity .. 6-17
6.1.2.1 Inhalation 6-17
6.1.2.2 Oral .....'.'.'.'.'.'.'.'.'.'.'.'. 6-30
6.1.2.3 Intraperitoneal Injection ."" 6-31
6.1.2.4 Dermal .' g_32
6.2 EFFECTS ON THE LIVER, KIDNEYS, AND LUNGS 6-33
vi i
-------
CONTENTS (Cont'd.)
6.2.1 Liver "
6.2.2 Kidneys °-34
6.2.3 Lungs 6'34
6.3 BEHAVORIAL TOXICITY AND CENTRAL
NERVOUS SYSTEM EFFECTS 6'35
6.4 OTHER TISSUES OR ORGANS 6-37
6.4.1 Nasal Cavity 6-37
6.4.2 Eyes 6-37
6.4.3 Circulatory System b M
6.5 SUMMARY 6"38
7. CARCINOGENICITY, MUTAGENICITY, AND REPRODUCTIVE
AND TERATOGENIC EFFECTS 7-1
7.1 CARCINOGENICITY 7~l
7.1.1 Introduction 1~\
7.1.2 Animal Studies •• '"J
7.1.2.1 Inhalation Exposure: Rat 7-2
7.1.2.2 Oral Administration in Drinking
Water: Rat 7-9
7.1.2.3 Oral Administration Gavage: Rat 7-17
7.1.2.4 Dermal Exposure: Mouse 7-21
7.1.2.5 Initiation - Promotion: Mouse 7-22
7.1.2.6 Subcutaneous or Intraperitoneal
Administration: Mouse 7-23
7.1.3 Epidemiologic Studies 7-24
7.1.4 Quantitative Estimation • '-•"
7.1.4.1 Procedures for Determination of Unit
Risk 7-32
7.1.4.2 Calculation of Quantitative Estimates . 7-48
7.1.4.3 Relative Potency 7-60
7.1.5 Summary 7"°°
7.1.5.1 Qualitative Assessment 7-66
7.1.5.2 Quantitative Assessment 7-68
7.1.6 Conclusions 7"69
7.2 MUTAGENICITY 7"71
7.2.1 Introduction ., 7-71
7.2.2 Gene Mutations in Bacteria 7"7.}
7.2.2.1 Salmonella Assay /"/I
7.2.2.2 Mutations in Klebsiella 7-78
7.2.2.3 Host-Mediated Assay 7-7°
7.2.2.4 Body Fluid Analysis '.....' 7-79
vi i i
-------
CONTENTS (Cont'd.)
7.2.3 Bacterial DNA Repair Tests 7-79
Tj.2.4 Gene Mutations in Neurospora 7-79
7.2.5 Gene Mutations in Yeast 7-80
7.2.6 Gene Mutations in Mammalian Cell Cultures ....... 7-80
7.2.7 Sex-Linked Recessive Lethal Test in Drosophila .. 7-81
7.2.8 Chromosomal Aberrations in Human and Other
Mammalian Systems 7-33
7.2.8.1 Studies on Human Chromosomes in
Vitro '.- 7-83
7.2.8.2 Studies on Rodent Chromosomes in Vitro . 7-85
7.2.8.3 Studies on Human Chromosomes in Vivo ... 7-85
7.2.8.4 Studies on Rodent Chromosomes"Tn~Wvo .. 7-86
7.2.8.5 Micronueleus Assay 7777... 7-88
7.2.8.6 Dominant Lethal Assay .... 7-88
7.2.8.7 Sister-Chromatid Exchange Assay 7-89
7.2.9 Conclusions 7-99
7.3 REPRODUCTIVE AND TERATOGENIC EFFECTS 7-91
7.3.1 Reproductive Effects 7-91
7.3.1.1 Male Clinical-Epidemiologic
Investigations 7-95
7.3.2 Teratogenic Effects 7-96
7.3.3 Summary and Conclusions 7-98
8. SYNERGISM AND ANTAGONISM AT THE PHYSIOLOGICAL LEVEL 8-1
9. ECOSYSTEM CONSIDERATIONS 9_!
9.1 EFFECTS ON MICROORGANISMS AND PLANTS 9-1
9.1.1 Effects on Microorganisms and Lower Plants 9-1
9.1.2 Effects on Higher Plants [[['. 9-2
9.2. BIOCONCENTRATION, BIOACCUMULATION, AND
BIOMAGNIFICATION g-2
9.3 EFFECTS ON AQUATIC ANIMALS 9,3
9.3.1 Freshwater Fish 9-3
9.3.2 Freshwater Invertebrates 9-5
9.3.3 Saltwater Fish 9-6
9.4 SUMMARY 9_6
10. REGULATIONS AND STANDARDS 10_1
10.1 OCCUPATIONAL STANDARDS 10-i
IX
-------
CONTENTS (Cont'd.)
10.2 FOOD TOLERANCES 10"1
10.3 TRANSPORTATION REGULATIONS 10~2
10.4 WATER REGULATIONS 10~3
10.5 SOLID WASTE REGULATIONS 10'3
11. REFERENCES R~1
APPENDICES
A. Evaporation Rate of Epichlorohydrin Calculated
Accordi ng to the Method of Di111ng (1977) A-l
B. Soil Adsorption Coefficient (K ) and Soil Organic
Matter/Water Partition Coefficient (Q) B-l
C. Calculation of the Log Octanol/Water Partition
Coefficient (log P) by the Method of Hansch and
Leo (1979) L"i
D. Bioconcentration Factors Calculated for
Epichlorohydrin by Four Methods U"1
E. Comparison of Results by Various Extrapolation Models .. E-l
F. International Agency for Research on Cancer Classifica-
tion System for the Evaluation of the Carcinogenic Risk
of Chemicals to Humans F"1
-------
LIST OF TABLES
Table
3-1
3-2
3-3
3-4
3-5
3-6
3-7
4-1
4-2
5-1
5-2
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
7-1
Page
Physical and Chemical Properties of Epichlorohydrin 3-2
Typical Reactions of Epichlorohydrin 3-5
Rate Constants for Hydrolysis of Epichlorohydrin as a
Function of Temperature 3-7
Rate Constants for Epichlorohydrin Reaction with
Various Anions 3-8
Rate of Reaction of Epichlorohydrin (ECH) under
Neutral and Buffered Conditions at 37°C 3-9
Estimation of Epichlorohydrin Production, 1978-1980,
in Millions of Pounds 3-14
Domestic Consumption of Epichlorohydrin for 1977 3-15
Distribution of 14C-Radioactivity in Rat Tissue
Following a 10 mg/kg Oral Dose of 14C-Epichlorohydrin .. 4-5
Tissue Distribution of Radioactive 14C-Epichlorohydrin
and Metabolites in Rats 4-6
Illness Episodes in Epichlorohydrin Workers 5-2
Chromosomal Aberration Frequency in Lymphocytes from
Workers Exposed to Synthetic Resin ED-20 5-4
Acute Effects of Epichlorohydrin 6-2
Summary of Mortality Findings in Rats and Mice after
Acute Inhalation Exposure to Epichlorohydrin 6-8
Acute Intraperitoneal Toxicity of Epichlorohydrin 6-12
Dermal Irritation Scores for Solutions of
Epichlorohydrin in Cottonseed Oil 6-15
Subchronic Effects of Epichlorohydrin 6-19
Mortality in Mice Exposed to 2,500 ppm Epichlorohydrin . 6-25
Mortality of Mice Administered Epichlorohydrin Orally .. 6-31
Lethality Following Repeated Dermal Application of
Epichlorohydrin in Rats 6-33
Squamous Cell Carcinomas of the Nasal Cavity of Male
Wistar Rats Following Thirty 6-Hour Exposures to 100
ppm Epichlorohydrin Vapor 7-4
XI
-------
LIST OF TABLES (Cont'd.)
Table
7-2 Dose-response for Induction of Squamous Cell Carcinomas
in the Nasal Cavity of Male Wistar Rats Exposed to
Epichlorohydrin Vapor .................................. 7-5
7-3 Kidney Weights and Kidney/Body Weight Ratios in Male
Wistar Rats Given Epichlorohydrin in Drinking Water
for 81 Weeks ........................................... 7-14
7-4 Number of Forestomach Tumors and Incidence in Male
Wistar Rats Given Epichlorohydrin in Drinking Water
for 81 Weeks ............................ . .............. 7-15
7-5 Cumulative Mortality of Rats Given Epichlorohydrin by
Gavage ................................................. 7-18
7-6 Incidence of Rat Forestomach Macroscopic Neoplastic
Lesions ................................................ 7-19
7-7 Histopathology of Proliferative Lesions in the
Forestomach of Rats Given Epichlorohydrin (ECH) by
Gavage ................................................. 7-20
7-8 Histopathology of Papillomas and Carcinomas in the
Forestomach of Rats Given Epichlorohydrin by Gavage ---- 7-21
7-9 Comparison of Mortality in Enterline's Epichlorohydrin
Study: Updates by Cause and by Latency (1978 versus
1981) .................................................. 7-25
7-10 Observed and Expected Deaths and SMRs Among 863 Males
Exposed for More Than Three Months in the Manufacture
of Epichlorohydrin, by Time Since First Exposure
Norco, Louisiana and Deer Park, Texas 1948-1979 ........ 7-27
7-11 Comparison of Mortality in Epichlorohydrin (ECH)
Alone and Combined with Isopropyl Alcohol (IPA)
Exposure Groups in Deer Park, Texas ................ .... 7-28
7-12 Estimation of 95 Percent Upper-Limit Unit Risk
Estimates for Nasal Papillomas and Squamous Cell
Carcinomas Using the Multistage Model With and Without
Time Dependence ........................................ 7-56
7-13 Relative Carcinogenic Potencies Among 53 Chemicals
Evaluated by the Carcinogen Assessment Group as
Suspect Human Carci nogens ............. ................. 7-62
7-14 Induction of Sex-Linked Recessive Lethals in
Drosophila by Epichlorohydrin .......................... 7-82
XI I
-------
Table Page
7-15 The Effects of Epichlorohydrin on the Fertility
of Wistar Rats 7-92
8-1 Summary of Study Measurements after Exposure to
Epichlorohydrin and Subsequent Exposure to Cold 8-3
8-2 The Effect of Heat Stress on the LCen of
Epichlorohydrin in the Rat and Mouse 8-5
9-1 Percent Seedling Survival 60 Days After Sowing
Eucalyptus Seeds Treated with Epichlorohydrin
Solution. 9-2
9-2 Epichlorohydrin Toxicity to Four Fish and One Aquatic
Invertebrate 9-4
9-3 The Acute Toxicity of Epichlorohydrin to Bluegill and
Tidewater Silverside F1sh 9-5
10-1 Occupational Standards for Epichlorohydrin 10-2
E-l Estimates of Epichlorohydrin Low-Dose Risk in Male
Wistar Rats Derived from Four Different Models E-4
xi i i
-------
LIST OF FIGURES
Figure Page
4-1 Blood Concentrations of Eplchlorohydrin in Mice
after Intraperitoneal Injection of 200 mg/kg 4-3
4-2 Proposed Metabolic Pathways for Eplchlorohydrin 4-7
7-1 Mortality of Rats Following Exposure to 100 ppm of
Eplchlorohydrin ,. .,' 7-3
7-2 Mortality of Rats Following Lifetime Exposure to
Eplchlorohydrin 7-6
7-3 Growth of Rats Following Chronic Exposure to
Epichlorohydrin , , 7-7
7-4 Patterns of Eplchlorohydrin Adrainistration in Male
Wistar Rats 7-11
7-5 Intake of Epichlorohydrin in Drinking Water by Male
Wistar Rats 7-12
7-6 Effect of Epichlorohydrin Treatment of Body Weight 1n
Male Wistar Rats 7-13
7-7 Histogram Representing Frequency Distribution of the
Potency Indices of 53 Suspect Carcinogens Evaluated by
the Carcinogen Assessment Group 7-61
7-8 Mutagenicity of Aromatic Epoxy Resins and
Epichlorohydrin for S. typhimurium TA100 7-73
7-9 Mutagenicity of Epichlorohydrin, Styrene Oxide, and
DDNU-oxide at Various Concentrations 1n S.
typhiinurlum TA100 , 7-74
7-10 Dose-response Curves for Epichlorohydrin 7-75
7-11 Mutagenicity of Epichlorohydrin With and Without S-9
Mix 7-77
7-12 Dose-Response Curve for Epichlorohydrin-Treated
Cultures 7-81
xi v
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
The EPA Office of Health and Environmental Assessment (OHEA) is
responsible for the preparation of this health assessment document. The
OHEA Environmental Criteria and Assessment Office (ECAO/RTP) had overall
responsibility for coordination and direction of the document prepara-
tion and production effort. The chapters addressing physical and
chemical properties, sampling and analysis, and toxicity data were
written by Theodore Keneklis, Ph.D., Lawrence Kaufman, Ph.D.,
William McLellan, Ph.D., Nicholas Mujjar, Ph.D., Cipriano Cueto, Ph.D.,
and John Strange, Ph.D., all of Dynamac Corporation.
The OHEA Carcinogen Assessment Group (CAG) was responsible for pre-
paration of the sections on carcinogenicity. The principal authors of
the carcinogenicity material were .Larry Anderson, Ph.D.,
Steven Bayard, Ph.D, and James W. Holder, Ph.D.
The OHEA Reproductive Effects Assessment Group (REAG) was responsible
for the preparation of sections on mutagenicity (K.S. Lavappa, Ph.D.,
principal author) and teratology (Carol Sakai, Ph.D., principal author).
The following individuals provided peer review of drafts of this
document:
U.S. Environmental Protection Agency
Gregory Kew, Ph.D.
Exposure Assessment Group
Office of Research and Development
Nancy Pate, D.V.M.
Office of Air Quality Planning and Standards
W. Bruce Peirano, Ph.D.
Health Effects Research Laboratory
Office of Research and Development
Consultants and Reviewers
I.W.F. Davidson, Ph.D.
Bowman Gray Medical School
Wake Forest University
Winston-Sal em, N.C.
Derek Hodgson, Ph.D.
Chemistry Department
University of North Carolina
Chapel Hill, N.C.
XV
-------
P.O. Lotilaker, Ph.D.
Pels Research Institute
Temple University Medical Center
Philadelphia, PA
J. F. Quast, Ph.D.
Toxicology Research Laboratory
Health and Environmental Sciences
DOW Chemical USA
Midland, MI
XVI
-------
1. EXECUTIVE SUMMARY
1.1 BACKGROUND INFORMATION
1.1.1 Properties
Epichlorohydrin (l-chloro-2,3-epoxypropane) is a colorless liquid with a
characteristic chloroform-like, irritating odor. It is partially miscible
with water and soluble in benzene, alcohol, and ether. It is a bifunctional
alkylating agent that can chemically bind with many cell constituents. The
epoxy group of epichlorohydrin is highly reactive. In most reactions, the
compound behaves primarily as an epoxide, initially combining through the
epoxy group to form 3-chloro-2-hydroxypropyl derivatives. Epichlorohydrin
undergoes a variety of chemical reactions with many compounds, and thus is
widely used as a chemical intermediate.
1.1.2 Production x
Epichlorohydrin is produced commercially by high temperature chlprination
of propylene to ally! chloride, followed by chlorohydration with hypochlorous
acid to form a mixture of isomeric glycerol dichlorohydrins. The mixture is
subsequently dehydrochlorinated with alkali to yield epichlorohydrin. Epi-
chlorohydrin is produced in the United States by the Dow Chemical Company and
Shell Oil Company. U.S. production in 1977 was 276 million pounds (134 million
kilograms). In 1982, 330 million pounds were produced.
1.1.3 Use
Epichlorohydrin1s major use is as a constituent of epoxy resins and gly-
cerol. Epichlorohydrin is also used as a raw material for the manufacture of
glycerol and glycidol derivatives used as plasticizers, stabilizers, solvents,
dyestuff intermediates, surface active agents, and Pharmaceuticals. It is
also used in such products as paints, varnishes, and shellacs. In addition,
it is used directly as a stabilizer in chlorine-containing materials such as
synthetic rubber and certain insecticides.
1.1.4 Environmental Release. Transport, and Fate
The largest sources of emission of epichlorohydrin to the environment are
from its manufacture, use as an intermediate, or from accidental spills. The
.ultimate environmental fate of epichlorohydrin depends on its release, transport,
and persistence characteristics. Epichlorohydrin is known to be released into
(1) the atmosphere from manufacture and use, (2) water from industrial effluents,
and (3) the terrestrial compartment from spills and dumping. Epichlorohydrin
1-1
-------
is not expected to persist in air, water, or soil because of its tendency to
hydrolyze and otherwise degrade. If released at the water/soil interface,
epichlorohydrin's water solubility, estimated soil adsorption coefficient, and
theoretical behavior in a landfill indicate the compound will enter the water.
Epichlorohydrin released at the air/soil interface will enter the air because
of the compound's high volatility and soil mobility. At the air/water inter-
face, epichlorohydrin will partition into both media.
1.1.5 Environmental Transformation
In the environment, the major-chemical transformation of epichlorohydrin
is through hydrolysis; the half-life in distilled water at 20° C is 8 days.
Hydrolysis is expected to be faster if chloride or carbonate-bicarbonate ions
are present. The major hydrolysis product of epichlorohydrin is 3-chloro-
1,2-propanediol. Other possible transformation processes in the environment
are photolysis and oxidation, but &iese would be minor compared to hydrolysis.
1.2 UPTAKE, METABOLISM, AND EXCRETION
Epichlorohydrin is readily absorbed and rapidly distributed to various
tissues and organs. In laboratory mammals, the highest concentrations after
exposure were found in the kidney, liver, pancreas, adrenals, and spleen.
Following an oral dose of 14C-epichlorohydrin to rats, the compound was rapidly
absorbed from the gastrointestinal tract. The major routes of elimination in
rodents were via the kidneys and lungs. Approximately 40 percent of the
radioactivity, regardless of the route of administration, was excreted in the
urine within 72 hours, and about 20 percent was exhaled as 14C- carbon dioxide.
Fecal excretion amounted to about 4 percent of the dose. Epichlorohydrin is
metabolized first by hydrolysis, then by oxidation to oxalic acid or by conjuga-
tion with glutathione to form mercapturic acid derivatives.
1.3 EFFECTS ON HUMANS
Epichlorohydrin as a liquid or vapor can cause respiratory, skin, and eye
irritation in humans. Pulmonary and liver changes were detected following
exposure in one case study. Headache, nausea, and head and chest congestion
were reported following the worker's exposure to epichlorohydrin. Local skin
contact with epichlorohydrin is reported to cause severe skin irritation.
Severe skin burns as well as burning of the eyes have occurred following
accidental exposures. Allergic reactions have also been reported in workers
occupationally exposed to epichlorohydrin. In one case of a severe epichloro-
hydrin inhalation exposure, initial irritation of the eyes and throat was
1-2
-------
followed by chronic asthmatic bronchitis. In this poisoning case, liver
biopsies showed extensive fatty infiltration and degenerative changes. 1.4
1.4 ANIMAL TOXICITY
Epichlorohydrin is well absorbed and moderately toxic by oral, dermal,
and inhalation routes. The acute oral dose lethal to 50 percent of rats
exposed (LD56) to epichlorohydrin was approximately 250 mg/kg body weight.
The inhalation 6-hours LC5o in rats was 360 ppm, and the no-observed effect
level (NOEL) was 283 ppm for 6 hours. Acute exposure caused central nervous
system depression and death resulting from respiratory paralysis. A single
nonlethal dose can cause kidney and lung damage in rats. Subchronic exposure
by inhalation, oral, and intraperitoneal injection routes studies caused
severe renal toxicity. Epichlorohydrin was intensely irritating to skin,
nasal mucosa, and eyes; in addition, it can cause skin sensitization in labora-
tory animals. The target organs or tissues, listed in descending order of
sensitivity to epichlorohydrin, are the nasal mucosa (when inhaled), kidneys,
liver, and cardiovascular system. There was no unique strain or species
sensitivity indicated.
1.5 CARCINOGENICITY, MUTAGENICITY, AND REPRODUCTIVE AND TERATOGENIC EFFECTS
1.5.1 Carci nogeni ci ty
Results,of long-term animal studies provide sufficient evidence of the
carcinogenic potential of epichlorohydrin as a weak contact carcinogen which
appears to produce no metastases. It is both route-dependent and site-specific
in that tumors appear only at the site of first contact. In view of increases
in nasal carcinomas seen in rat inhalation studies, the increased forestomach
tumors in rat drinking water and gavage studies, the increased local sarcomas
produced in mice after subcutaneous injection of epichlorohydrin, and the
chromosomal aberrations found in the peripheral lymphocytes of exposed workers,
epichlorohydrin should be considered a potential human carcinogen. Although
three epidemiologic studies have not demonstrated epichlorohydrin to be car-
cinogenic to humans, they cannot be regarded as indications that epichlorohydrin
is safe because the studies had some inherent weaknesses. Considering the
above evidence and applying the International Agency for Research on Cancer
(IARC) approach (Appendix F) for classifying the weight of evidence for car-
cinogenicity in experimental animals, epichlorohydrin would be placed in the
2B category, meaning that it is probably carcinogenic to humans. The 95
percent upper-limit incremental unit risk estimate for continuous inhalation
exposure to 1 ug epichlorohydrin/m3 of air is 1.2 x 10"6 based on extrapolation
1-3
-------
from the rat Inhalation study. In terms of continuous exposure to 1 ppm, the
upper-limit unit risk is 4.8 x 10~3. For exposure via drinking water, the 95
percent upper-limit incremental unit risk estimate from water containing 1 pg
epichlorohydrin per liter is 2.8 x 10~7, based on an animal drinking water
study. In terms of relative carcinogenic potency, epichlorohydrin exposure by
either route is among the weakest of the chemicals that the EPA has evaluated
as suspect carcinogens.
1.5.2 Mutagenicity
Substantial evidence is available demonstrating that epichlorohydrin
causes gene and chromosomal mutations in several experimental systems both in
y-jtro and in animals. Cytogenetic studies of workers exposed to epichlorohy-
drin have yielded evidence for a clastogenic effect on lymphocytes. The
compound has been shown to be an active inducer of gene mutations in bacteria,
yeast, Drosophila. and cultured mammalian cells. Epichlorohydrin is also
effective in causing sister chromatid exchanges in human cells in vitro and
preferential cell killing of repair-deficient bacteria. Chromosomal effects
induced by epichlorohydrin were detected in both IQ vivo and jji vitro mammalian
cell assays.
It may be hypothesized that epichlorohydrin mutagenic action results from
its alkylating reactivity. Epichlorohydrin should be considered as potentially
hazardous to humans because of its clastogenic action in experimental systems.
1.5.3 Reproductive and Teratogenic Effects
Results from published studies indicate that epichlorohydrin (under the
conditions of the studies) was not teratogenic in mice, rats, or rabbits.
Signs of embryotoxicity were observed at doses that were toxic to the pregnant
mouse. Transient infertility was observed in male rats exposed to epichlorohy-
drin, and recovery followed termination of exposure. No detrimental effects
were observed on the fertility of male workers exposed to epichlorohydrin;
however, weaknesses in study design prevent conclusions concerning the
potential for causing infertility.
1.6 SYNERGISM AND ANTAGONISM
Synergistic and antagonistic relationships at the physiological level
between epichlorohydrin exposure and other variables (cholesterol ingestion,
cold stress, and heat stress) were limited to a few fragmentary studies.
Rabbits ingesting cholesterol and epichlorohydrin (but not those consuming
epichlorodydrin alone) showed impaired heart function and increased blood
lipid levels. Rats inhaling a single 4-hour dose of epichlorohydrin followed
1-4
-------
by a cold stress (5 C for 2 hours) showed very few physiological differences
from rats that were not cold-stressed. On the other hand, rats subjected to
heat stress (35 C for 2 hours/day for 4 weeks) and epichlorohydrin inhalation
(4 hours/day for 4 weeks) showed enhanced toxicity.
1.7 ECOSYSTEMS AND AQUATIC BIOTA
No studies were found that discussed the effects of epichlorohydrin on
ecosystems. Toxicity data available for bacteria, algae, protozoa, aquatic
invertebrates, and fish indicate that neither growth inhibition nor mortality
in aquatic biota would occur at aqueous environmental concentrations of epichlo-
rohydrin below 5 mg/1 (5 ppm). Environmental levels as high as 5 ppm have not
been shown to occur in the natural environment. Calculated estimates indicate
that low levels of epichlorohydrin, which may potentially occur in the environ-
ment, would not pose significant bioconcentration or bioaccumulation hazards
in the food chain.
1.8 REGULATIONS AND STANDARDS
Epichlorohydrin is currently controlled by U.S. and foreign regulations.
U.S. regulations provide exposure limits in the workplace, restrictions on use
in food and related industries, discharge limits into navigable waters, trans-
portation procedures, and maximum disposal limits requiring special landfills.
Epichlorohydrin is not currently regulated under the Safe Drinking Water Act
or the Clean Air Act.
1.9 CONCLUSIONS
Epichlorohydrin is a potential hazardous compound because of its alkylating
properties, its mutagenicity in a variety of systems, and its probable carcino-
genic potential for humans as determined by animal bioassay data. Moreover,
increased chromosomal aberrations have been reported in peripheral lymphocytes
of workers exposed to epichlorohydrin. Further studies are needed for a more
definitive conclusion on the possible effects of epichlorohydrin on humans.
1.10 RESEARCH NEEDS
Research needed to support or strengthen the existing data base on epi-
chlorohydrin are indicated in the following section. Particular emphasis is
placed on areas of studies needed to assess more fully the health hazards of
human exposure to epichlorohydrin.
Epidemiology:
Prospective and retrospective cohort studies of exposed workers
should be pursued with special attention given to quantification of
individual exposure levels versus health effects.
1-5
-------
Monitoring the lymphocytes of workers exposed to epichlorohydrin for
cytogenetic damage should be continued. These studies should include
sister-chromatid exchange analysis and a suitable assay for mutations
in somatic cells. Health monitoring should include analysis of
changes in blood count and should take into account smoking and
drinking habits. The occurrence and frequency of sperm-morphology
changes among exposed workers should be studied.
Kidney function monitoring (e.g., BUN, creatinine, protein) should
be done periodically for exposed epichlorohydrin workers to determine
whether any changes in kidney function are occurring as a result of
occupational exposure to the chemical.
The hemoglobin alkylation technique should be applied to
workers currently exposed to measured levels of epichloro-
hydrin to determine if the number of immature erythrocytes
present in the peripheral blood increase with increased
exposure.
Subchronic and Chronic Toxicity:
Chronic oral and inhalation exposure tests on mammals at several
concentrations of epichlorohydrin up to the maximum tolerated con-
centration (6. hours/day, 5 days/week for 2 years) are needed.
Since the nasal mucosa and kidneys appear to be two of the most sen-
sitive tissues to epichlorohydrin, mechanisms leading to mucosal and
renal lesions should be explored.
Since epichlorohydrin induces kidney and liver damage in mammals,
studies should be conducted to determine if hypertension is also
induced or aggravated by epichlorohydrin. Accordingly, the effects
of epichlorohydrin on the heart should be more thoroughly investi-
gated.
Genetic Toxicity:
Mammalian studies using cytogenetic analysis of mouse bone marrow
should be carried out following epichlorohydrin exposure by the
inhalation ro,ute.
A carcinogenic bioassay of epichlorohydrin in mammals exposed by the
oral and respiratory routes at several dose levels up to the maximum
tolerated dose should be conducted. The study should be designed to
serve as a model for assessment of carcinogenic risk to humans.
Reproductive and Teratogenic Research:
Since the acidity of the stomach may lead to hydrolysis of epichlo-
rohydrin, reproductive and teratogenic effects might be observed
if a mode or route of dosing other than gastric intubation were
employed.
1-6
-------
Compound Distribution and Pharmacokinetics Research:
The pathways of distribution, metabolism and elimination of epichlo-
rohydrin as a function of dose-route, dose-rate, and dose-frequency
in mammals should be investigated.
Experiments to determine the reaction of epichlorohydrin with various
nucleophiles present in biological systems should be conducted to
facilitate an understanding of the reaction of epichlorohydrin with
cells and their organelles.
An attempt should be made to correlate any results obtained using in
vivo or in vitro testing systems with molecular dosimetry, expressed
as either binding to hemoglobin or to DNA in target organs.
1-7
-------
-------
2. INTRODUCTION
The 1970 Clean Air Act as amended in 1977 requires that EPA regulate,
under Section 112, those pollutants that may reasonably be anticipated to
result in an increase in mortality or an increase in serious irreversible, or
incapacitating reversible, illness. It also states that EPA must regulate,
under Section 111 (d), those pollutants that may reasonably be anticipated to
endanger public health or welfare. This health assessment document was re-
quested by the Office of Air Quality Planning and Standards (OAQPS) as a basis
for evaluation of epichlorohydrin as a hazardous pollutant. It is envisioned
by the Office of Health and Environment Assessment to be one of several information
sources to guide regulatory strategies of the OAQPS and other EPA program offices.
In the development of this assessment document, the scientific literature
has been inventoried, the studies evaluated, and summary conclusions prepared
to identify qualitatively the chemical toxicity and related characteristics.
Observed effect levels and other measures of dose-response relationships are
discussed. In assessing the health effects of human exposure to epichlorohydrin,
few epidemiologic studies were available. The effects in humans have generally
been ascertained from either occupational or accidental exposures, and little
information has been reported on the concentrations associated with these expo-
sures. Thus, it has been necessary to rely on animal studies to derive indica-
tions of potential harmful effects in relation to dose or exposure levels.
Key animal studies are presented in a descriptive manner that includes
information on the test organism, dosage regimen and schedule of exposures,
duration of exposure, life expectancy of the animal, duration of the experi-
ment, types of effects seen with each dosage, number of test groups and con-
trols, number of animals per group, and sex and age of animals. Statistical
significance, coefficients of variation, and the purity of the test material
are specified when the data were available. Anecdotal reports are covered in
a concise form, but key studies have been expanded for discussion to reach a
"weight-of-evidence" summarization within each section.
The major topics included in this document are: physical and chemical pro-
perties, sampling and analytical methods, production and use, levels and sources
in the environment, fate and transport, and biological effects, including the
effects of epichlorohydrin on ecosystems and aquatic species. Biological
effects have been defined to include metabolism and pharmacokinetics as
well as,toxicity to organ and tissue systems, carcinogenicity, mutagenicity,
2-1
-------
teratogenicity, and reproduction. Human data on the effects of epichlorohydrin
are presented and interpreted in terms of data from animal experimentation.
This document is intended to serve as part of the basis for decision-making
in the various regulatory offices within the EPA as well as to inform the
general public of the nature and extent of information available for assessment
of health hazards resulting from environmental exposure to epichlorohydrin.
2-2
-------
3. BACKGROUND INFORMATION
3.1 PHYSICAL AND CHEMICAL PROPERTIES
3.1.1 Introduction
Epichlorohydrin is a chlorinated derivative of a, B-propylene oxide and has
the formula:
H
,C - CH -
CH2C1
It is a clear, colorless, highly reactive liquid at ambient temperature and
has a chloroform or garlic-like odor. It is both volatile and flammable (Weast
1978). Some of the relevant physical and chemical properties of epichlorohydrin
are listed in Table 3-1.
3.1.2. Synonyms and Trade Names
Epichlorohydrin has the following synonyms and trade names:
2-chloromethyl oxirane
glycidyl chloride
3-chloro-l,2-propylene oxide
or epichlorohydrin
(DL)-or epichlorohydrin
SKEKhG
l,2-epoxy-3-chloropropane
2,3-epoxypropyl chloride
glycerol epichlorohydrine
ECH
ECHH
l-chloro-2,3-epoxypropane
3-chloro-1,2-epoxypropane
(chloromethyl) ethylene oxide
2-(chloromethyl) oxirane
chloropropylene oxide
T-chloropropylene oxide
3-chloropropene 1,2-oxide
3.1.3. Identification Numbers
Epichlorohydrin has three commonly used identification numbers:
1. Chemical Abstracts Service (CAS) No. 106-89-8,
2. Registry of Toxic Effects of Chemical Substances (RTECS) No. TX 49000,
and
3. U.S.EPA No. A762-1952.
3.1.4. Significance of Physical Properties with Respect to
Environmental Behavior
Epichlorohydrin is miscible with ethanol, diethyl ether, acetone, and chlo-
rinated aliphatic hydrocarbons, and slightly soluble in petroleum hydrocarbons
3-1
-------
TABLE 3-1. PHYSICAL AND CHEMICAL PROPERTIES OF EPICHLOROHYDRIN
MOLECULAR FORMULA, MOLECULAR WEIGHT, AND ELEMENTAL COMPOSITION
Molecular Formula:
Molecular Weight:
Elemental
Composition:
C3H5OC1
92.53
C = 38.94%
H = 5.45%
Cl = 38.32%
0 = 17.29%
PHYSICAL PROPERTIES
Melting Point (Weast 1978)
Freezing Point (Shell 1969)
(Dow 1980)
Boiling Point (Shell 1969)
(Dow 1980)
Density (g/ml, 20° C) (Shell 1969)
Specific Gravity (20/20° C) (Shell 1969)
Vapor Pressure (16.6° C) (Sax 1975)
(30° C) Verschueren 1977)
Concentration in Saturated Air
(760 mmHg, 25° C) (Hine et al. 1981)
Coefficient of Expansion at 68° F
(Shell 1969)
Solubility (Shell 1969)
Water (10° C)
Water (20° C)
Pounds per Gallon (68° F) (Shell 1969)
Flash Point (Tag open cup) (Shell 1969)
(Tag closed cup) (Dow 1980)
Autoignition Temperature (Dow 1980)
Latent Heat of Vaporization (calc.)
(Shell 1969)
Odor Threshold in Air (Hine et al. 1981)
Surface Tension (20° C) (Shell 1969)
Heat of Combustion (Shell 1969)
Liquid Viscosity (25° C) (Shell 1969)
Refractive Index (25° C) (Shell 1969)
Heat Capacity (25° C) (Dow 1980)
(100° C) (Dow 1980)
Heat of Formation (25° C) (Dow 1980)
Explosive Limits (volume % in Air) (Dow 1980)
Heat of Fusion (25° C) (Dow 1980)
-48.0° C
-57.2° C
-57.1° C
116.11° C (760 mmHg)
116.07° C (760 mmHg)
20 25
d4 1.1812; d4 1.1750
1.181
10 mmHg
22 mmHg
1.7%
0.000577 per °F
6.52%
6.58%
9.58 Ibs.
41° C
31° C
416° C
9060 cal/mole at the b.p.
10 ppm
37.00 dynes/cm
4524.4 cal/gm
0.0103 poises
nQ1.4358
31.5 cal/mol0 C
40.0 cal/mol° C
-35.6 kcal/mol
3.8-21.0
2,500 cal/mol
3-2
-------
and water. Epichlorohydrin forms an azeotrope with water, distilling at 88° C
and containing 75 percent epichlorohydrin by weight (Riesser 1978).
Epichlorohydrin has weak ultraviolet absorption; the exact spectrum is
not reported. Based upon its structure, which combines alky! halide and alkyl
epoxide properties, it is reasonable to infer that the maximum absorption will
be below 300 nm, the lower cutoff for sunlight due to atmospheric absorption.
Hydrolysis of epichlorohydrin is slow at room temperature but is accelerated
by heat or traces of acid or base. Reactions with compounds containing active
hydrogen (e.g., alcohols, primary or secondary amines) normally occur initially
at the more reactive epoxide site of the molecule, although reactions involving
initial displacement of chlorine are also known to occur (Massiot and Levy
1981).
The volatility of epichlorohydrin, as indicated by its relatively high
vapor pressure, may lead to transfer from water or soil to the air phase. The
details of the environmental fate of epichlorohydrin as determined by its
physical and chemical properties are discussed in Section 3.4.
An estimated value of the log octanol/water partition coefficient, using
the method of Hansch and Leo (1979) is 0.26 ± 0.04. This indicates a low
affinity of epichlorohydrin toward fats or soil. More details are presented
in Section 9.2.
3.1.5. Chemical Reactions
Although the epichlorohydrin molecule has two available reactive sites
(the chlorine atom and the epoxy group), the epoxy group dominates the reactive
character of the compound. The three-membered epoxide ring is highly strained,
making its bonds weaker than those of linear ethers. The result is a less
stable molecule that will readily undergo acid-catalyzed reactions and cleavage
by bases. It is this high degree of reactivity to which epichlorohydrin owes
its industrial importance as a chemical intermediate (Dow 1980).
In most of its reactions, epichlorohydrin behaves as an epoxide, initially
reacting through the epoxy group with substances containing an active hydrogen
atom and with numerous other diverse compounds to form chlorohydrin derivatives.
This is true even of its behavior towards substances such as tertiary nitrogen
bases, metal alkoxides, and organic acid salts, which will normally effect
direct replacement of an active organic halogen atom. The chlorine can be
eliminated as hydrogen chloride in a subsequent step involving displacement by
the initially-generated, hydroxy group. Glycidol derivatives are thus formed
and undergo the addition reactions typical of epoxy compounds (Shell 1969).
3-3
-------
By suitable adjustment of reaction conditions, epichlorohydrin can be an
intermediate in the synthesis of a wide variety of products (see Shell Chemical
Company 1969 for more details). These transformations are illustrated by the
reaction of epichlorohydrin with alcohols. In the presence of an acidic catalyst
such as stannic chloride, 3-chloro-2-hydroxypropyl ethers are formed in high
yields (see reaction 1 in Table 3-2).
Table 3-2 lists the typical reactions that have been observed with epichlo-
rohydrin. A number of these reactions are of commercial importance and are dis-
cussed in Section 3.3.
3.1.6 Chemical Reactions in the Environment
Although available literature provides a good description of hydrolysis and
related reactions of epichlorohydrin in the laboratory, little information was
available on its photochemistry or oxidation in air, water, or soil. Epichloro-
hydrin is not persistent and appears to hydrolyze in several weeks' time under
laboratory conditions (Brb'nsted et. al. 1929) but reports of field studies on
epichlorohydrin were not found in the literature. Removal processes may be
possible to predict for epichlorohydrin in air based on molecular structure.
These processes would include reaction with hydroxyl radicals, or to a lesser
extent with ozone. The estimated rate constant for reaction of epichlorohydrin
with the hydroxyl radical is 2 x 10"12 cm3 molecule"1 sec"1 (U.S. EPA, 1980).
The atmospheric residence time was estimated to be 5.8 days; photolysis was
considered to be possible but not probable (see section 3.1.6.3).
3.1.6.1 Hydrolysis and Related Reactions—Epichlorohydrin hydrolyzes by a
complex scheme. Many papers (e.g., Ross 1950; Addy and Parks 1965) delineate
mechanisms and rates of epoxide hydrolysis, but none of the studies has investi-
gated hydrolysis under environmental conditions.
The chlorine atom does not react or directly participate in the initial
hydrolysis, but it does affect the initial hydrolysis rate by its inductive
effects (Ross 1950 and 1962 as cited in NIOSH 1976a; Pritchard and Long 1956;
Pritchard and Siddiqui 1973; Kwart and Goodman 1960). Epichlorohydrin can
hydrolyze by two general mechanisms: uncatalyzed and acid-catalyzed (Brb'nsted
et al. 1929; Ross 1950). In the uncatalyzed reactions, the rate-determining step
involves opening of the epoxide ring by the attack of water, an anion, or other
nucleophile at the C-l carbon (Bronsted et al. 1929; Kwart and Goodman 1960; Long
and Pritchard 1956; Addy and Parker 1965):
3-4
-------
TABLE 3-2. TYPICAL REACTIONS OF EPICHLOROHYDRIN
Monqhydric
Alcohols
Organic Acids
Acyl Chlorides
-CH-CHsCl +
Catalyst
ROH -> R-0-CHS-CHOH-CH2C1
A
CH2-CH-CH2d + RCOOH
A
CH2-CH-CH2C1 + RCOC1
-» RCOO-CHa-CHOH-CH^Cl
+
HO-CH2-CHOOCR-CH2C1
-» CH2C1-CH(OOCR)-CH2C1
Aldehydes
CH»-CH-
CH2-CH-CH2C1 + RCHO
R H
C
0 0
I I
-» ClCH^-CH-CHj
Amines
0
/\
+ RHNH
-» RHNCH^-CHOH-CH2C1
Grignard Reagents
A
RMgBr
CH^R-CHOH-CH2C1 ^
+
CHsOH-CHR-CH2Cl +
-» CHaR-CHOMgBr-CHaCl
CH20MgBr-CHR-CH2Cl
Water
A
CH2-CH-CH2C1 + HOH
-* CH2OH-CHOH-CH2C1
Inorganic Acids
C<-W
PT + Upl
-5- rHiri-rnnH-rHnPi
Source: DOW (1980)
3-5
-------
0
/\
C1CH2CH-CH2 + A
OH
C1CH2CH-CH2A » C1CH2CHCH2A + OH
The acid-catalyzed reactions have been identified as having an A-2 type mech-
anism. This mechanism may depend on the strength of the acid. Epichlorohydrin
is first protonated reversibly, and the protonated compound reacts with water
or an anion. The rate-determining step can involve ring opening at either the
C-l or C-2 carbon; in the acid-catalyzed process, opening at C-2 may be pre-
ferred owing to the stability of the secondary carbonium ion (Bronsted et al.
1929; Long and Pritchard 1956; le Noble and Duffy 1964):
/
C1CH2CH-CH2 +
0
l\
C1CH2CH-CH2 +
A
OH
C1CHCH-CH
or
C1CH2CHCH2OH
A
OH
/-.
C1CH2CH-CH2V
OH
A
I
C1CH2CHCH2
A
Most of the information necessary for the product and half-life calcula-
tions for environmental hydrolysis of epichlorohydrin either was experimentally
measured or could be estimated from available data. Table 3-3 summarizes the
epichlorohydrin hydrolysis rate constants, k.^ and k2 (k.^ is for the uncatalyzed
addition of water, and k« is for the acid-catalyzed addition of water). Table
3-4 lists the experimentally derived rate constants, k~ and k., for anion
reactions with epichlorohydrin (k3 is for the uncatalyzed addition of an
anion, and k, is the acid-catalyzed addition of an anion).
3-6
-------
Table 3-3.
Rate Constants for Hydrolysis of Epichlorohydrin as a
Function of Temperature
10
(liter mol~ s~ )
Reference
0.0
tU. U
-------
Table 3-4. Rate Constants for Epichlorohydrin Reaction with Various Anions
Anion
Chloride, Cl"
Iodide, I"
Thiosulfate, $203
Formate, HC02
Benzoate, CgHgC02
Acetate, CH3C02"
Nitrate, N03"
Bicarbonate, HC03
Carbonate, C03
°C
20
40
20
20
20
20
20
37
20
65
75
80
85
35
45
50
60
ioV 10 V
_1 -1 2 -2 -1
(liter mol s ) (liter mol s )
1.15C, 0.99d 0.45C
6.3e 6.8e
10. Ob
6.3b
0.47b
0.52b
0.62b
3.33C
0.022f
0.179
0.3Q9
0.529
0.68s
0.429
0.839
1.42.9
2.5s
a Rate for uncatalyzed addition of anion.
bRate for acid-catalyzed addition of anion.
cBronsted et al. (1929).
dRoss (1950)
eAddy and Parker (1965).
fPetty and Nichols (1954).
9Shvets and Aleksanyan (1973).
3-8
-------
The hydrolytic half-life for epichlorohydrin at 20°C in distilled water was
determined to be 8 days (Bronsted et al. 1929). Ross (1950) reported the rates
of first-order hydrolytic reactions of epichlorohydrin at 37°C under acidic and
neutral conditions (Table 3-5).
Epichlorohydrin is expected to hydrolyze faster if the water has a high
chloride or high carbonate-bicarbonate content. Although the hydrolysis pro-
duct is 3-chloro-l,2-propanediol, significant concentrations of other products,
including l,3-dichloro-2-propanol, can be formed from further reactions with
aqueous anions.
Table 3-5. Rate of Reaction of Epichlorohydrin (ECH) under Neutral
and Buffered Conditions at 37°C
Duration of
Reaction (h)
24
48
72
100
169
Neutral Condition9
% ECH n
Reacting ^ (h x)
36.5 0.0190
59.0 0.0185
72.5 0.0180
84.5 0.0185
96.5 0.0200
Buffered
% ECH
Reacti ng
52.0
76.6
88.0
96.0
100.0
Condition
4 (h"1)
0.0185
0.0185
0.0180
0.0185
—
Relates to reaction in water.
Relates to reaction in water containing 0.1 M sodium acetate and 0.1 N
acetic acid.
Source: Ross (1950).
3.1.6.2 Oxidation ~ Shell (1969) lists several oxidation and reduction reactions
of epichlorohydrin, none of which is environmentally significant, for example:
0
/\
CH2-CHCH2C1
HO
C1CH2C-OH
u
/\
CH2CHCH2C1
HI
3-9
-» C1CH2CH2CH3
-------
Epichlorohydrin can be oxidized by free radical processes in liquid (Dobbs
et al. 1976; Beckwith 1982) or gas phases (Oilling et al. 1976); these reactions
may occur as photochemically initiated atmospheric reactions (Gay and Bufalini;
1971; Bufalini 1971). The liquid phase, free radical oxidations discussed here
are probably not important in the environment, but are possible mechanisms by which
epichlorohydrin could be oxidized by atmospheric free radical initiators.
Available literature evaluates the mechanisms of liquid phase reactions with
a few free radical initiators. The structure of the free radical produced from
epichlorohydrin depends in part upon the radical initiator. Dobbs et al. (1976)
suggested that the t-butoxyl radical, (CHOgCO, preferentially abstracts a hydrogen
atom from an alicyclic carbon, whereas the hydroxyl radical, HO, preferentially
abstracts one from an acyclic carbon. Their experimental work with epichlorohydrin
was limited to the identification of species formed by reactions with hydroxyl
radicals produced by the titanium (III) ion-hydrogen peroxide system.
3.1.6.3 Photolysis — No ultraviolet absorption data were available for epichlo-
rohydrin. Neither the alkyl halide nor the epoxide portion are expected to have
strong absorption in the sunlight region (wavelengths above 300 nm) (Calvert and
Pitts 1966). Epichlorohydrin's absorption maximum is probably below 250 nm and,
at most, is expected to have only minimal absorption above 300 nm. No significant
direct photochemical reactions are expected under environmental conditions. In-
direct processes involving reactions other than photochemically generated species
are of course possible.
No reports on the photodegradation of epichlorohydrin in the environment were
found in the literature. In a laboratory study, Dilling et al. (1976) determined
the decomposition rate at 27 ± 1° C of 10 ppm epichlorohydrin in an atmosphere
containing 5 ppm nitric oxide (NO). Two 275-W reflector sunlamps (which had a
short wavelength cutoff of 290 nm) were used as UV sources. The intensity was
stated to be about 2.6 times that of natural sunlight at noon on a summer day in
Freeport, Texas. The rate of disappearance of epichlorohydrin was determined by
a gas chromatograph with a flame ionization detector. The half-life reported from
these studies was 16.0 hours, and no products were identified. Extrapolation from
these results to the rate of photolysis in the environment is not justified.
3.2 ANALYTICAL METHODOLOGY
3.2.1 Introduction
Several approaches have been developed for determining the quantity of epi-
chlorohydrin present in environmental samples. Four useful analytical methods that
3-10
-------
have been described include: (1) volumetric determination (Swan 1954); (2)
oxidation and colorimetric determination at 412 nm (Jaraczenska and Kaszper
1967); (3) extraction with carbon tetrachloride and spectrophotometric deter-
mination of absorbence at 1,274 cm'1 (Adamek and Peterka 1971); and (4) ring
opening by halogen acids and quantification of the hydrogen halide (Dobinson
et al. 1969).
There are several other methods that have been developed for epichlorohy-
drin and other epoxides (Dobinson et al. 1969) that may have specific applica-
tions. These methods vary widely in specificity and sensitivity. Methods for
analysis of air and water are described below.
3.2.2 Chemical Analysis in Air
Several methods exist for measuring epichlorohydrin concentrations in
air. Daniel and Gage (1956) described a sensitive colorimetric method for
measuring epichlorohydrin vapor. This method is based on the oxidation of
epichlorohydrin with periodic acid, followed by reaction of the formaldehyde
with ammonia and acetylacetone to give a yellow-colored solution. The method
is capable of giving a reasonably accurate result with as little as 20 ug of
epichlorohydrin and is therefore capable of analyzing atmospheric concentra-
tions of 10 mg/cm using a 2-liter sample. The analytical error is estimated
at about 2 percent.
The analysis of epichlorohydrin in workroom air is best achieved using an
adsorption technique and gas chromatography. NIOSH (1976a) recommended sampling
using activated charcoal as the adsorbent. The determination of epichlorohy-
drin at the level of a few parts per billion has been performed using gas
chromatography-mass spectrometry, which provides the highest sensitivity with
high specificity (van Lierop 1978).
A standard sampling and analytical method for epichlorohydrin has been
developed by NIOSH (1976a). The method involves trapping epichlorohydrin
yapor from a known volume of air on charcoal and then desorbing it with carbon
disulfide. An aliquot of the desorbed sample is injected into a gas chromato-
graph; the area of the resulting peak is determined and compared with standards.
This method was developed to analyze epichlorohydrin over the range of 11.7 to
43.1 mg/m at an atmospheric temperature of 23° C and a pressure of 765 mmHg.
For a 20-liter sample, the useful range of this method is 2 to 60 mg/m3 at a
detector sensitivity that gives nearly full deflection on a strip chart recorder
for a 1 mg sample. The method is capable of measuring levels as low as 50 ppb
3-11
-------
(NIOSH 1976a). Any compound that has about the same retention time as epichloro-
hydrin under the gas-chromatographic conditions used in this method will interfere
with the analysis.
A portable, battery-operated gas analyzer and a detector tube are available
from at least one instrument supplier for the detection of epichlorohydrin in air
(AIHA 1961).
Anderson et al. (1981) reported the results of a comparison of activated
charcoal, Amberlite XAD-2, and Amberlite XAD-7 for sampling of epichlorohydrin
in workroom air. Amberlite XAD-7 was observed to be an excellent adsorbent for
epichlorohydrin, giving high recoveries and no decomposition. Percent recovery
of 8, 40, and 400 ug samples with a dichloromethane eluent was between 99 and
100 percent, with a range of estimated standard deviations between 1.2 and 1.7.
3.2.3 Chemical Analysis in Water
Some of the methods used for measuring epichlorohydrin levels in water are
essentially the same as those used for levels in air. Many of the methods for
analysis in air involve first trapping the epichlorohydrin in an aqueous medium.
Daniel and Gage (1956) described a sensitive colorimetric method for deter-
mining levels of epichlorohydrin in water based on oxidation with periodic acid to
form formaldehyde and then reaction with ammonia and acetylacetone to form
3,5-diacetyl-l,4-dehydrolutidine, which has a yellow color. The color is allowed
to develop, and the optical density of the sample is measured with a spectro-
photometer at 412 mm. This method is accurate to 20 jjg in a 15 ml water sample
(1.3 ppm).
In another method (Dobinson et al. 1969), aqueous samples at concentrations
ranging between 10 and 1,000 ppm may be analyzed by treatment with alcoholic
potassium hydroxide and then determining the generated chloride ions potentio-
metrically using aqueous silver nitrate solution.
Epichlorohydrin may also be determined in water samples by extracting the
samples with carbon tetrachloride (Adamek and Peterka 1971). The carbon tetra-
chloride is then analyzed for epichlorohydrin using infrared spectrophotometry.
The analytical band used for quantification is at 1,274 cm , using a 1-mm cell
thickness and a 10-ml water sample. The limit of quantification was approximately
0.03 percent (300 ppm). Infrared spectroscopy is a convenient and rapid analy-
tical technique; however, in this assay, the sensitivity was low. It may be
increased by extracting larger water samples and using a greater cell thickness.
3-12
-------
3.3 PRODUCTION, USE, AND RELEASES TO THE ENVIRONMENT
3.3.1 Introduction
The purpose of this document is to present available information relevant
to human health effects of epichlorohydrin. Available information regarding
sources, emissions, and ambient air concentrations, has been included only to
give the reader a preliminary indication of the potential presence of this sub-
stance in the ambient air. While the available information is presented as
accurately as possible, this discussion is acknowledged to be based on limited
data and is not intended to be used alone to make regulatory conclusions regarding
risks to public health.
If a review of the health information indicates that the Agency should con-
sider regulatory action for epichlorohydrin, a considerable effort will be under-
taken to obtain more extensive information regarding sources, emissions, and am-
bient air concentrations. Such additional data will provide information for
drawing regulatory conclusions regarding the extent and significance of public
exposure to epichlorohydrin.
3.3.2 Production
Epichlorohydrin is produced commercially in the U.S. by the chlorination
process. The chlorination process is a three-step series of reactions. The first
step is the production of allyl chloride from propylene and chlorine. The allyl
chloride is used to provide epichlorohydrin by hypochlorination and subsequent
neutralization. Currently, all the ally! chloride produced in the U.S. is used
in the production of epichlorohydrin. Crude epichlorohydrin may be transferred
directly to the glycerine production step. Refined epichlorohydrin is sold for
other uses (Blackford, 1978).
In the U.S., production of epichlorohydrin started in about 1937 and expanded
in 1949 in connection with the first synthetic glycerine plant. Epichlorohydrin
is currently produced by two companies, Shell Chemical Company (Deer Park, Texas;
Norco, Louisiana) and Dow Chemical U.S.A. (Freeport, Texas) by the chlorohydri-
nation of allyl chloride. Ciba-Geigy Corp. (Toms River, New Jersey) and Union
Carbide Corp. (South Charleston, West Virginia) have in the past produced, or
have the capacity to produce, epichlorohydrin. Production of crude epichloro-
hydrin by Shell and Dow in 1977 was about 296 million pounds (135 million kg).
Estimated production of refined epichlorohydrin by Shell and Dow in 1977 was
203 million pounds (92 million kg).
3-13
-------
The production of epichlorohydrin for the years 1978 through 1980, based
on estimates of the Chemical Information Service (U.S. EPA 1983) are shown in
Table 3-6. Estimated capacity for 1982 was 640 million pounds (Shell: 220
million pounds, Dow: 420 million pounds).
3.3.3 USE
The estimated U.S. consumption of epichlorohydrin in 1977 was as follows:
synthetic glycerine, 25 percent; unmodified epoxy resins, 53 percent; epichloro-
hydrin elastomers, 2 percent; other products, 15 percent; and exports, 5
percent. Uses included in the 15 percent consumption for other products are
glycidol ethers, some modified epoxy resins, wet strength resins for paper,
water treatment resins, surfactants, and ion exchange resins. Domestic consump-
tion data for 1977 are summarized in Table 3-7. In 1978, the United States
accounted for 49 percent of the world's total epichlorohydrin use (Blackford
1978).
Table 3-6 Estimation of Epichlorohydrin Production,
1978-1980, in Millions of Pounds.
Glycerine Feed
Refined Feed
Total
1978
—6U
265
325
1979
310
350
1980
300
Source: U.S. EPA 1983
Table 3-7 Domestic Consumption of Epichlorohydrin for 1977
Use
Crude epichlorohydrin
For synthetic glycerine
For refined epichlorohydrin
Refined Epichlorohydrin
For unmodified epoxy resins
For epichlorohydrin elastomers
For miscellaneous
million Ibs
291
75
216
203
152
7
44
(million kg)
(132)
(34)
(98)
(92)
(69)
(3)
(20)
Source: Blackford 1978
3-14
-------
3.3.3.1 Synthetic Glycerine—1977 132 million pounds (60 million kg) of synthetic
glycerine were produced. Of that, 66 million pounds (30 million kg) were pro-
duced from epichlorohydrin. Approximately 25 percent of crude epichlorohydrin
production in 1981 was estimated to be used to produce glycerine (U.S. EPA 1983).
About 65.5 percent percent of the synthetic glycerine produced is expected to be
derived from epichlorohydrin in 1982 (Blackford 1978).
Glycerine is produced from epichlorohydrin .by one company in the U.S., Dow
Chemical, Freeport, Texas (SRI 1982). The capacity of Dow's Freeport plant is
115 million pounds (52 million kg) per year. (Chemical Marketing Reporter
1981).
3.3.3.2 Epoxy Resins—The principal application of epichlorohydrin is in the
manufacture of epoxy resins. The term "epoxy resin" is assigned to polymeric
materials containing epoxide groups. Epoxy resins are commercially used in
protective coatings, bondings, adhesives, reinforced plastics, and other
products. The consumption of unmodified epoxy resins in the U.S.
in 1980 was 317 million Ibs. (144 million kgs.)
About 90 percent of commercially produced epoxy resins are made by reaction
of epichlorohydrin with 2,2-di(4-hydroxypheny1) propane. There are many indus-
trial users of epoxy resins; 25 or more generally use epoxy resins as starting
materials for their products (Osterhof 1981).
3.3.3.3 Textijes—Epichlorohydrin has been used to esterify the carboxyl
groups of wool. The resulting product has both increased life and improved
resistance to moths. Epichlorohydrin has been used to prepare protein-modified,
wool-like fiber, which has an affinity for acid-dyes and exhibits resistance
to both molds and insects. In addition, epichlorohydrin has been used in the
preparation of polyacrylonitrile, polyvinyl chloride, polyvinyl alcohol, and
other fibers. It has also been used to impart wrinkle resistance and to pre-
pare antistatic agents and textile sizings, and derivatives of epichlorohydrin
have shown utility as leveling, dispersing, softening, emulsifying, and washing
agents (Dow 1980).
3.3.3.4 Paper, Inks, and Dyes—Wet-strength paper sizing may be prepared from
either polyamides modified with epichlorohydrin or from the reaction product
of epichlorohydrin and an alkylene amine. In the paper industry, epichloro-
hydrin adducts are also useful as filler retention aids, paper coatings,
flocculants, and antistatic agents. Paper and paperboard products with improved
3-15
-------
printability, pigment retention, folding endurance, and gloss also have been
prepared with epichlorohydrin reaction products (Dow 1980).
Epichlorohydrin polyhydroxy compounds and their esters are useful in the
production of special printing inks and textile print pastes. These products
yield flexible films that are chemically inert to sodium hydroxide and other
chemical solutions (Dow 1980).
3.3.3.5 Anion Exchange Resins—Water-insoluble, anion-exchange resins having
good stability may be prepared by reacting epichlorohydrin with ethylenediamine
or a high molecular weight homolog. Strong-base, anion-exchange resins can be
produced by reacting epichlorohydrin with polymeric tertiary amines. Epi-
chlorohydrin-based anion exchangers have been used successfully to clean
polluted air and water. Cation-exchange resins may be produced by the con-
densation of epichlorohydrin with polyhydroxy phenols followed by sulfonation
of the product (Dow 1980).
3.3.3.6 Solvents—Epichlorohydrin is a good solvent for cellulose acetate,
rosin, and ester gum (Dow 1980). The reaction of epichlorohydrin with alcohols,
alcoholates, and the sodium salts of stearic, oleic, palmitic, myristic, and
other fatty acids yields products used as vinyl polymer plasticizers, solvents
for food and tobacco flavorings, and as plasticizers for polyurethanes (Dow
1980).
3.3.3.7 Surface Active Agents—A number of epichlorohydrin-based, surface-
active agents have been synthesized by condensing epichlorohydrin with a
polyamine such as tetraethylene-pentamine plus a fatty acid such as stearic
acid. A sulfonated epichlorohydrin derivative has occasionally been sub-
stituted for epichlorohydrin. Such products have been found useful in cos-
metics and shampoos, and as detergents, sudsing agents, water softeners, and
demulsifiers (Dow 1980).
3.3.3.8 Epi chl orohydri n-Based Rubber Elastomers—Copolymers of epichloro-
hydrin with ethylene oxide are members of a new family of specialty polyether
rubbers. These elastomers possess desirable properties over a wide range of
temperatures and are resistant to gasoline, oil, and ozone. Other advantages
are "good aging properties," high resiliency, and flexibility at low tempera-
tures (Dow 1980).
Only about 3 percent of refined epicholorhydrin consumption in the United
States in 1982 is expected to be used for the production of epichlorohydrin
3-16
-------
elastomers. It is estimated that the consumption of epichlorohydrin in manu-
facturing the elastomers was about 7 million pounds (3.2 million kg) in 1977
(Blackford 1978).
Applications for epichlorohydrin-based rubber include automotive and
aircraft parts, seals, gaskets, wire and cable jackets, adhesives, packings,
hose and belting, rubber-coated fabrics, and energy-absorbing units (Dow
1980).
3.3.3.9 Starch Modifier—Food starch may be modified by epichlorohydrin to
produce stable canned food products. According to Rutledge and Islam (1973),
treating rice with epichlorohydrin cross-links the starch granules and pro-
duces a stable rice which retains favorable properties after canning.
The U.S. Food and Drug Administration permits food starch to be treated
with epichlorohydrin alone and with combinations of epichlorohydrin and pro-
pylene oxide, acetic anhydride, and succinic anhydride. Provision has also
been made for sequential treatment of starch with epichlorohydrin followed by
propylene oxide. The use of these reagents is subject to limitations con-
cerning maximum concentrations of the treatment reagents and maximum allowable
concentrations of chemical residues in the treated food. (21 CFR 172.892).
Food starch may be treated with epichlorohydrin not to exceed 0.3 percent and
propylene oxide not to exceed 10 per cent, residual propylene chlorohydrin in
the modified starch not to exceed 5 ppm.
3.3.3.10 Other Current Uses—A variety of other products are produced from
epichlorohydrin, most of them in relatively small volumes. Among them are
glycidyl ethers, some types of modified epoxy resins, intermediates for plas-
ticizers, dyestuffs, Pharmaceuticals, oil emulsifiers, and lubricants (Riesser
1978). It is also used as a stabilizer in chlorine-containing materials such
as chlorinated rubber and chlorinated insecticides (Shell 1969; Abdel Sayed et
al. 1974).
3.3.3.11 Proposed Uses—Epichlorohydrin has been recommended as a good solvent
for cellulose acetate, rosin, and ester gums (Shell 1969), although its toxi-
city may preclude such use. Dow (1980) has recommended the following addi-
tional possible applications of epichlorohydrin or its derived products:
- asphalt improvers
- corrosion inhibitors
- electrical insulation for wire
3-17
-------
- filament sizing
- fire-retardant urethanes
- liners for polyethylene bottles
- linoleum and linoleum cements
- lubricant additives
- petroleum production aids
- photographic film bases
- rubber latex coagulation aids
- waterproofing compounds
- zinc electroplating compounds
Epichlorohydrin in conjunction with copper ions has been proposed as a
possible spermicidal agent by Kalla and Bansal, 1977. The safety of this
combination as a contraceptive agent would have to be demonstrated before this
proposed use could be considered seriously.
3.3.4 Substitute Chemicals/Processes
Glycerine has been manufactured by at least three processes other than
epichlorohydrin hydrolysis; none require chlorinated hydrocarbons as inter-
mediates. Furthermore, the use of epichlorohydrin in glycerine production has
diminished (Blackford 1978).
The unique properties of epoxy resins and epichlorohydrin elastomers are
difficult to replace, especially if the use of closely related chemicals such
as epibromohydrin or halogenated 1,2-epoxybutanes are also prohibited. Other
compounds containing an epoxy ring could be used to make epoxy resins, but the
properties of the resins, as well as manufacturing costs, might be adversely
affected by substituting them for epichlorohydrin.
3.3.5 Environmental Release
Epichlorohydrin may be released into the environment as a result of its
manufacture, use, storage, transport, and disposal. It has been estimated that
epichlorohydrin emissions to the atmosphere from the three major production
facilities in the United States totaled about 1.47 x 105 pounds (6.7 x 104 kg)
in 1978 (Anderson et al. 1980). Releases from these facilities occurred
mainly through condenser vents of the distillation columns, although smaller
amounts of emissions also came from storage tanks and loading and handling
facilities and from plant equipment leaks.
Epichlorohydrin is also released during its use in the production of epoxy
resins, elastomers, and miscellaneous products (Anderson, et al. 1980).
During 1978, epichlorohydrin was released in at least 11 locations in the
3-18
-------
United States during the production of epoxy resins; these emissions totaled
about 2.5 x 10 pounds/year (1.1 x 105 kg/year). An additional 8.1 x 104
pounds/year (3.7 x 10 kg/year) of epichlorohydrin was estimated to be released
during its use in the production of chemicals other than glycerine (Anderson,
et al. 1980),
Epichlorohydrin may also be released as a component of industrial effluents
and other wastes. No information was found concerning actual modes of waste
disposal in the United States.
Epichlorohydrin has been reported to have been released in accidental
spills on at least two occasions. In a train accident in January 1963, about
5,000 gallons of epichlorohydrin was spilled into the New River at South
Fayette, West Virginia (Gillenwater 1965). In another train accident in
January 1978, (also in West Virginia), more than 20,000 gallons of epichloro-
hydrin were spilled near the center of the town of Point Pleasant, about 150
feet from the Ohio River (Chemical Week 1978). Apparently only in the second
case was a cleanup attempted. The chemical was reported not to have contami-
nated the Ohio River. Local officials ordered the removal of about 1 acre of
soil (to several feet deep). The soil was eventually removed to a Dow Chemical
Company facility in Texas. The level of epichlorohydrin in water from wells
closest to the spill area at the time was 75 ppm. After estimation of the rate
of subsurface movement, the city's wells were closed and since then water has
been obtained from a radial collector several miles from the city (EPA, 1978).
No empirical information on epichlorohydrin release rates from landfills
or lagoons was found. In a theoretical discussion, Falco, et al. (1980) deve-
loped a model to predict the transport, sorption, and degradation properties of
epichlorohydrin and other organic chemicals from waste disposal sites. This
predictive model indicated that sorption of epichlorohydrin onto soil from
groundwater is unlikely, and that "approximately 100%" of the compound released
from unconfined landfills and lagoons would reach surface waters. Additional
results of this model are discussed in Section 3.4.1.2.2.
3.3.6 Environmental Occurrence
Hushon, et al. (1980) indicated that epichlorohydrin had been identified
in water samples from an oil refinery, in industrial effluents, and in surface
water. However, no sample concentrations were reported and no information was
provided concerning the location from which the samples were collected.
Levels of epichlorohydrin in the ambient air have not been determined.
3-19
-------
3.4 ENVIRONMENTAL TRANSPORT AND FATE
The ultimate environmental fate of epichlorohydrin depends on its release,
transport, and persistence characteristics. Epichlorohydrin is known to be
released into (1) the atmospheric compartment as a result of manufacture and
use, (2) the aquatic compartment with industrial effluents, and (3) the terrestrial
compartment due to spills and dumping. Upon release, epichlorohydrin is not
expected to persist. If released at the water/soil interface, epichlorohydrin's
water solubility, estimated soil adsorption coefficient, and predicted behavior
when released from a landfill indicate the compound will move into the aquatic
compartment. Should epichlorohydrin be released at the air/soil interface,
high volatility of the compound and soil mobility suggest it will favor the
atmosphere. At the air/water interface, the volatility, solubility, and other
data indicate epichlorohydrin will partition into both media. In the discussion
below, the environmental fate of epichlorohydrin is assessed in light of its
transport and persistence properties.
3.4.1 Transport
3.4.1.1 Volatilization—Epichlorohydrin is a volatile liquid with a latent
heat of vaporization of 9,060 cal/mole at the boiling point (Hawley 1977;
Riesser 1978). Based on its vapor pressure, reported to be 10 mmHg at 16.6° C
(Sax 1975) and 22 mmHg at 30° C (Verschueren 1977), it is expected to volatilize
under normal environmental conditions. Although its evaporation half-life has
not been experimentally determined, it may be predicted using the model system
proposed by Dill ing et al. (1976) and Dill ing (1977). Based on a 1 mg/1
aqueous solution in a total volume of 250 ml or water, 6.5 cm deep and stirred
at 200 rpm at 20° C, the predicted evaporation half-life of epichlorohydrin is
0.15 days (2.37 days if the depth is 100 cm). The calculations are shown in
Appendix A. These experimental conditions will not be encountered in natural
aquatic environments; therefore, the actual half-life in the environment may
differ from these data.
3.4.1.2 Sorption
3.4.1.2.1 Soils. The soil-water partition coefficient per unit organic
matter (K ) for epichlorohydrin was estimated using the regression equation
of Kenaga and Goring (1980): log KQc = 3.64-0.55 (log water solubility) ±
1.23 orders of magnitude (see Appendix B). The calculated KQC values range
from 10.28 x lO"1'23 (0.61) to 10.28 x 101'23 (174) at a solubility of 60,000
3-20
-------
mg/1 and 9.76 x lO"1'23 (0.57) to 9.76 x 101'23 (166) at 66,000 mg/1. Although
actual experimental determination of KQC could yield a value different from
those estimated, epichlorohydrin does have a low estimated K (i.e., 100).
Thus, it would have a low potential for soil adsorption.
An alternative method of estimation developed by Briggs (1973) indicates
epichlorohydrin to be "mobile" in soils. The soil organic matter/water parti-
tion coefficient, Q, was calculated (see Appendix. B) by the equation:
log Q = 0.524 (log P) + 0.618
where P is the octanol/water partition coefficient. Log P was estimated
according to the method of Hansch and Leo (1979) (see Appendix C). The calcu-
lated Q value of 5.68 indicates the compound will be "mobile" in soils, according
to Briggs1 (1973) rating system, and prefer the aqueous medium to soil.
3.4.1.2.2 Sediments. Epichlorohydrin, with its estimated K of 100, is pre-
oc * ^
dieted to have a low potential for sorption onto stream sediments (Kenaga and
Goring 1980). Using another mathematical approach, the EXAMS Model, Falco et
al. (1980) estimated the transport, sorption, and degradation properties of
various organic chemicals from waste disposal sites. The model's results for
epichlorohydrin indicated that sorption onto soil and sediment particles would
not be important processes. The authors stated that approximately 100 percent
of the compound released from unconfined landfills and lagoons would reach
surface waters. Falco et al. (1980) also reported that the potential for
contamination of bottom sediments in water bodies would be low because the
model predicted that concentrations in the sediments of a river, a pond, and a
lake would each be only 20 percent of the overlying water concentrations.
Sorption of epichlorohydrin onto sediments suspended within the water column
would also be very low; it was predicted that only 0.001 percent would be
sorbed onto suspended sediments in a river reach traversed in 5 days (50-250
miles), 0.01 percent onto suspended sediments in a pond with a 100-day retention
time, and 0.001 percent onto suspended sediments in a reservoir or lake with a
365-day retention time (Falco et al. 1980).
3.4.2 Fate
3.4.2.1 Chemical and Physical Process—The major chemical processes affecting
the environmental fate of epichlorohydrin are hydrolysis and oxidation. These
are discussed in more detail in Sections 3.1.6.1 and 3.1.6.2.
3-21
-------
Physical processes which may remove epichlorohydrin from the atmosphere
include adsorption on aerosol particles and subsequent removal, or adsorption
by soil and water at the earth's surface. Photodegradation is not expected to
be a significant potential environmental process. It is discussed in Section
3.1.6.3.
3.4.2.2 Biological Processes—Experimental data on bioaccumulation of epi-
chlorohydrin were not found in the literature. The aqueous reactivity, log P
value, and hydrolytic properties of epichlorohydrin suggest there is a low
potential for bioaccumulation or bioconcentration in aquatic and terrestrial
food chains. Published bioconcentration factors are based on either water
solubility or octanol/ water partition coefficients. Four different equations
for estimating bioconcentration factors (BCF) are presented in Appendix D.
Two equations based on water solubility are those of Chiou et al. (1977) and
Lu and Metcalf (1975). Two other equations based on log P values were reported
by Neely et al. (1974) and Veith et al. (1980). The log BCF values for epi-
chlorohydrin calculated by these four equations range from -0.032 to 0.968.
Log BCF values of less than 2 indicate low bioconcentration potential (Kenaga
1980).
Information on microbial biodegradation of epichlorohydrin is limited.
Epichlorohydrin was identified as an intermediate in the enzymatic hydrolysis
of 2,3-dibromo-l-propanol when chloride ions were present (Castro and Bartinicki
1968). Unspecified cultures of Flavobacteriurn were isolated from soil in an
alfalfa field and grown in a medium containing 0.005 M dibromopropanol. A
crude enzyme extract was obtained from the bacteria by centrifugation of
sonicated cells. The activity of the crude extract was reported to be equiv-
alent to the activity of the cell suspension. The crude enzyme extract was
partially purified, precipitated, and then passed through a Sephadex G-200
column. Dibromopropanol, epichlorohydrin, and epibromohydrin were each metabo-
lized by both the crude enzyme extract and the purified extract at pH 7.
Castro and Bartinicki (1968) observed that 2,3-dibromo-l-propanol was first
converted into epibromohydrin. The next step depended on the presence of
bromide or chloride ions. If bromide ions were present, the epoxide group
opened to yield 1,3-dibromopropanol. However, if chloride ions were present,
epibromohydrin was converted to epichlorohydrin.
3-22
-------
The biochemical degradation of epichlorohydrin was studied by Bridie et
al. (1979a) using unidentified seed cultures. They reported the theoretical
oxygen demand (TOD), biochemical oxygen demand (BOD), and the chemical oxygen
demand (COD). Five-day BOD's were measured using the American Public Health
Association's Standard Method No. 219 published in 1971. The method was modi-
fied by adding 0.5 mg/1 allylthiourea to prevent nitrification. The seed cul-
tures were obtained from a biological sanitary waste treatment plant, and dup-
licates were checked for activity using a mixture of glucose and glutamic acid.
COD was obtained by the standard potassium dichromate method ASTM D 1252-67
published by the American Society for Testing and Materials in 1974. The BOD's
for epichlorohydrin using unadapted and adapted seed culture were 3 and 14 per-
cent of the TOD, respectively. In comparison, the BOD's with unadapted seed
were 82 percent for glycerine and 1 percent for dichloropropanol (Bridie et al.
1979a).
3.5 SUMMARY
In 1980, 300 million pounds (136 million kg) of epichlorohydrin were pro-
duced in the U.S. by Shell Chemical Company and Dow Chemical U.S.A. (U.S. EPA
1983). Epichlorohydrin is used to produce epoxy resins, synthetic glycerine,
elastomers, and other products.
Emissions of epichlorohydrin to the atmosphere from production in 1978
were estimated to be 1.47 x 105 pounds (6.7 x 104 kg). Emissions from epoxy
resin production in 1978 were estimated to be 2.5 x 105 pounds (1.1 x 105 kg).
An additional 8.1 x 10 pounds (3.7 x 104 kg) of epichlorohydrin were estimated
to be released during its use in the production of chemicals other than glycerine
(Anderson 1980). No data were found regarding release rates into water bodies,
landfills, or lagoons; although several accidental spills have been documented.
Levels of epichlorohydrin present in the atmosphere were not available.
Epichlorohydrin released into the environment is not expected to persist.
Its behavior relative to air, soil, and water have been assessed based on
solubility, volatization, and other properties. It is predicted to have low
potential for soil and sediment adsorption, and that 100 percent of epi-
chlorohydrin released from a lagoon or landfill would reach surface waters.
Epichlorohydrin released to the atmosphere would be removed by chemical and
physical processes; its atmospheric half-life is calculated to be 5.8 days.
There is low potential for bioaccumulation of epichlorohydrin. Little is
known about the biodegradation of epichlorohydrin.
3-23
-------
-------
4. COMPOUND DISTRIBUTION AND RELATED PHARMACOKINETICS
The major route of exposure to epichlorohydrin in humans is through the
respiratory tract. There is also potential for dermal exposure. Exposure via
the oral route is expected to be slight; however, the possibility exists that
exposure by this route could occur as a result of ingesting contaminated water
or the leaching or unpolymerized epichlorohydrin from plastic wrap or plastic
containers into food.
4.1 ROUTES OF EXPOSURE AND ABSORPTION
No studies on the exposure and absorption of epichlorohydrin by humans
have been reported. However, there are studies that indicate rats absorb
epichlorohydrin following oral or inhalation exposure.
Weigel et al. (1978) administered a single 10 mg/kg dose of 14C-epichloro-
hydrin by oral gavage to 21 male (mean weight 250 g) and 21 female (mean
weight 208 g) Charles River CD rats. The epichlorohydrin was radiolabeled in
both the carbon 1 and 3 positions and had a specific activity of 1.66 mCi/mmole.
Animals were killed at 2, 4, 8, 12, 24, 48, and 72 hours after dosing, and the
concentrations of radioactivity in tissues, fluids, and excreta were measured.
Eight hours after treatment, less than 10 percent of the administered dose was
recovered from the gastrointestinal tract. Peak tissue concentrations of
radioactivity were reached 2 hours after dosing in males and after 4 hours in
females. Following absorption, 14C-labelled metabolites were released from
the body via the urine, exhaled air, and feces (see Section 4.4).
Smith et al. (1979) administered epichlorohydrin to rats by the oral or
inhalation routes and studied the pharmacokinetics of absorption, distribution,
and excretion. Single oral doses of 1 mg/kg or 100 mg/kg l,3-14C-epich1oro-
hydrin were administered to groups of four male Fischer 344 rats (weighing
190-220 g). Addi-tional groups of rats were exposed for 6 hours to air con-
taining 1 ppm or 100 ppm l,3-14C-epichlorohydrin. The total uptake, calculated
by summing all recovered radiolabel during and after the 6-hour exposure, was
15.5 ug/hour for exposure at 1 ppm and 1,394 ug/hour for exposure at 100 ppm
epichlorohydrin. The doses absorbed were 0.37 and 33 mg/kg, respectively.
Thus, a 100-fold difference in exposure concentration produced a 90-fold
difference in the absorbed doses. At 72 hours, regardless of dose level or
route, 46-54 percent of the radiolabel was excreted in the urine and 25-42
4-1
-------
percent was exhal'ed as carbon dioxide. These experiments indicate that epi-
chlorohydrin is absorbed well from the gut or the lungs, is rapidly distributed
to other tissues, and much of the administered epichlorohydrin is metabolized
and excreted within 72 hours.
Mice have been exposed to epichlorohydrin by the intraperitoneal and dermal
routes of exposure. In a study by De Petrocellis, et al. (1982), male mice were
given epichlorohydrin dissolved in corn oil, by a single intraperitoneal injection.
Groups of ten mice were killed by decapitation at 1, 3, 5, 7, 10, 15, 20 and 30
minutes after injection and blood samples were collected. Levels of epichloro-
hydrin in the samples were determined by a gas chromatograph with a flame ioni-
zation detector; these are plotted against time in Figure 4-1. As can be
seen from the graph, the jn vivo half-life of epichlorohydrin is extremely short,
being only just detectable after 15 minutes.
10.00 pr
=1 I I I I I I I I I I I I I I I I I I I
LIMIT OF ASSAY SENSITIVITY-
nn-il I I I
I I I I I I I I I I I I I I I I I I I
10 15
TIME, min
20
25
30
Figure 4-1. Blood concentrations of epichlorohydrin in
mice after intraperitoneal injection of 200 mg/kg.
Source: De Petrocellis, et al. 1982.
4-2
-------
4.2 DISTRIBUTION
The tissue distribution of 14C in rats receiving a single oral dose of
14C-epichlorohydrin (10 mg/kg) was studied by Weigel et al. (1978). Details of
the experiment have been outlined in Section 4.1. With the exception of kidney
levels in females at 48 hours, the highest tissue concentration at all time inter-
vals studied was in the kidneys, followed by the liver, pancreas, adrenals, and
spleen. In other organs studied (lungs, heart, brain, fat, muscle, skin, ovaries,
and/or testes), the levels were essentially at or below whole blood levels.
Peak tissue levels were reached in male rats at 2 hours and in female rats at 4
hours. Table 4-1 shows the peak tissue levels and the 72-hour tissue levels in
male and female rats. The tissue distribution patterns were similar in both sexes.
The authors did not estimate rates of clearance from tissues, although data were
collected at various time points. The chemical form of the radioactivity in
tissues was not determined. The principal route of elimination was via the kidneys.
TABLE 4-1. DISTRIBUTION OF 14C-RADIOACTIVITY IN RAT TISSUES FOLLOWING
A 10 mg/kg ORAL DOSE OF 14C-EPICHLOROHYDRIN
Tissue
Kidneys
Liver
Pancreas
Adrenal s
Spleen
Others f
Peak Level
Maleb
22.02
11.29
10.26
7.94e
6.87
2.57-4.82
(ug/g)a
Female0
22.73
7.77d
8.74
10.55d
4.81
2.61-5.13
72-Hour
Male
4.17
2.48
1.31
0.69
1.30
0.71-1.66
Level (ug/g)a
Female
8.68
3.40
1.52
4.81
1.56
0.74-1.39
Epichlorohydrin equivalents.
Two-hour sample.
cFour-hour sample.
Peak value was at 8 hours.
Peak value was at 4 hours.
Lungs, heart, brain, fat, muscle, skin, ovaries, and/or testes.
Source: Weigel et al. (1978).
4-3
-------
Smith et al. (1979) studied the distribution of radioactivity into 27 differ-
ent tissues in Fischer 344 rats 3 hours after a single oral dose of 100 mg/kg
14C-epichlorohydrin or at the end of a 6-hour exposure to 100 ppm 14C-epichloro-
hydrin in air. After the oral dose, the highest tissue levels were found in the
stomach, small intestine, kidneys, and large intestine. After inhalation, the
highest levels were found in the nasal turbinates, lacrimal glands, kidneys, large
intestine, and liver. The data are summarized in Table 4-2.
In this study, Smith et al. (1979) found high local concentrations in the
nasal turbinates after inhalation and in the stomach after ingestion. The dis-
tribution of radioactivity in the other organs examined was similar to that found
by Weigel et al. (1978).
4.3 METABOLITE IDENTIFICATION AND PATHWAYS
Smith et al. (1979) chromatographed by ion-exclusion the urinary metabolites
produced after inhalation exposure or oral administration of l,3-14C-epichlorohy-
drin to rats. After inhalation, six peaks of radioactivity appeared in the urine,
whereas after oral administration, seven peaks of radioactivity appeared in the
urine (72 hours). Two major peaks separated from the urine after oral administra-
tion accounted for 23 and 10 percent of the administered radioactivity, respectively.
After inhalation exposure, three major peaks in the urine accounted for about 36
percent of the radioactivity. Although there appears to be a difference in meta-
bolism depending on route of administration, no definitive conclusions can be
made, since the chemical identities of the urinary metabolites were not reported.
Fakhouri and Jones (1979) dosed male Sprague-Dawley rats orally for 5 con-
secutive days with 50 mg/kg epichlorohydrin and collected urine for 7 days.
Ether extracts of urine were chromatographed on thin-layer plates, and the
metabolites recovered were identified by gas chromatography-mass spectroscopy.
The N-acetyl derivatives (mercapturic acids) of l,3-(bis-cysteiny)propan-3-ol
and S-(2,3-dihydroxypropyl)cysteine were identified as major components and
beta-chlorolactic acid as a minor metabolite. No quantification of metabolites
was reported. The author's proposed metabolic scheme for epichlorohydrin is
shown in Figure 4-2.
Epichlorohydrin has two reactive electrophilic sites, the C-l carbon in the
epoxide ring and C-3, the chlorine-bearing carbon. These carbons can behave as
alkylating agents and hence can react nonenzymatically with glutathione or
protein sulfhydryl groups. However, the enzymatic reaction of epichlorohydrin
4-4
-------
t f\lt
^** ^^
I-H r-H
3B 0*
OCD.
_l>-
— i nn
o o
1 1 nx
O
J~~* T"
I— IO
^^> H-i
I-H Q_
t— LU
O 1
eco
CD1*
1— liH
CD
^E
f"* f^s
1 r-
O
CO
1 —
«=c
LL. Q£
0
u_
zo
o
I-H LU
=5 ID
CO CO
t— i O
t—>C
CO LU
1— 1
CD Z
O
LU 1— 1
_•? h—
co •*
co —i
I-H
O
1C
1
CO
o
I-H
K—
O O-i-
So_t—
Q.
OL^.
CD 0
ox:
rH 1
CM
0)
3
u to
4-» to
soP
O.CL.
a.
i.
O 3
CD O
1
co
U O)
4-> 3
-Q tO IO
E O to
Q.O.T-
O. (—
s-
CD 3
CD O
rH *""
^
a>
to
u to
.a -t->-i-
CO tO|— '
JM£ O
O)
E S-
CD O
O-C
rH 1
CO
*X3
O)
•a u
CO O)
•i— O)
I— co
g
to
, —
^^.
cn
o-
LU
cn
3
CO
S
CO
ro
a.
Cn
O"
LU
CO
3
CO
E
to
CO
r— •
CL-
OT
x^
cr
1 1 1
co
3
CO
to
CO
a.
CO
LU
CO
3
CO
O)
3
to
to
r—
—
co tr> co i—i cn CD co I-H CD oo cn co
co CD co CD cn co i— 1 1 — CD r^. co oo
CD CM CO r-H CM CO i— 1 CD CD rH i— 1 CD
|H«4
\ /\ <\ ^\ *^ JK^ ^\^ ^^ ^ *s^p/Sfcrf^ XM^XM^*/
CD«*COrHCDCnCDCOLf)COCD COCM
CD CD i-H CO CO CM CO CM CO ^- i-H CO rH
LncocDr^Ln^•cOLO^>.^•co coco
r-HCM rH CnCM i-H i-H
X*^S^"^/-1*NX-^S^P^/*^X*^^'^S ^*N/**S ^™NX" ^
CM cn CM co oo CM CM i—i «d- co CD P-.
cni — rH uo r~- i— i co CD coco cntn
CD CM «tf" CO CO «*• rH i-t O rH rH CD
cn cn co ^- ^*- CM cn oo cn CM co cn i— i
tn cn r^ CM ^ co co ^co (^- o r*~ cn
cn co CM r-. CD CM co CD CD oo «^- cn cn
CM «*• CO r-. ^d- i-H i-H i-H i-H rH
^^/•^^-^^^X—NX'^N/^^X"^ X*^X*^i^™^iX*^N/i^Ny1*^
•!*• CD CD i-H CO rH CD I"- rH in CO ^f **• *d"
CO rH CO i— 1 CO CO i— 1 CD Cn f^. Cn rH CD CM
CD rH CO ^- CO CM i— 1 i-H CD |— 1 CD cn CD CD
»~ -<*- *•*- -l*~ --*%-_^*^_ JW_J**-_J ^M^N^^S^^^^S^^X^^
f-?l fj^ t~l f*^ f^| ^J. <^ ^- Qf*( Q*J Q») ^s^ JHJ p*^ t f^
CMrH^-rHCncOCOcnCM LnCDcnCDCDCO
COOOLnOrH!-HCnCOCQCMI--.«st-r-l«3-
cMcor~.t~^LncMi— ii— i I-H co r- 1 cn I-H
OTO TD -a -a -a
x*"S/"""s/^">x""*N/**%**-tN/^^/'B"^ X'-N/^^X'^^^X'^^NX^S
rHcotnu^cococDco cocncoocDcn
CD CM CD CO «3- CO OO CD Cn CO CO 00 CO CM
Cn sj- CO CM CM CM rH rH CD CD CD CD CD CD
CM
cnLnrHcocDCDCOi— it— icDCDcncnr^co
r-- «3~ o LO cn «^- «^- coco «^" CM CD co co CD
^•Lni—rcncooocDcooo co co co CM CM i— i
O rH i-H
r-l
CO 0) CO
c c a>
•r- «r- "O 4->
4-> 4-> c ro
to to co c
0) 0) i— T-
4-> -u cn .a
CO CO i_
c: c i — to 3
-C «r- to ••— ro i— 4J
o >» E ro to co a>
COi— O) O)-i- S- CO C E O) C-)-»i — i —
E i — c cn s- a> s- ro u
OrOT3S-CJ>C:rOrO to&-rOtotO4->
4->E'i-rOCO'i-3S-i — O) T3 O) r0 3 ro
co co i«£ —i —i — i — i OQO_ (— «a; ic z: s LL.
ai
+*
4-
O
r-
CJ
ro
ai
o
4-
O)
O)
.c
T3
CO
O)
s_
3
CO
o a>
f°l C_
UJ» >-•
O) to
O
' — O.
CO . X
i- CO O)
O i-
3 C.
i. O O
O -C -I-
CD CO
tO r—
4-> S- CO
(O O -C
i- O
co ^:
cn 5— i
4-> CD
0 C
CM a> t-
CM O. O)
i c: •+->
CD 0 <4-
cn u ro
rH
c. t-
cn o o
•r- +J O)
«^« fo vjj
CO i— CO
•r- CO CO
a> -co
S <= TD
CO t~ »^N
ro o -P r~-
•a re I-H
^" Q) ^-^
**• t- -o
co a) a)
S- co i — ro
O) •!- -I-
0 to -i— ai
co a> E a)
•r- s- -a s- . -C
U_ 3 CO , 4->
CO S ' — 'i—
a> o i — c E
i — CL a) to o co
CO X > 4->
E O) 0) ro 4->
i — S- ro ..
-p s- i. a»
r— oj a) o) o
3 _c co E a> s-
•a -P o -r- c 3
-------
«
!*A* 1
-8-e
"5 p n
i 5 § 8
I ^ I X
i < $,-*
if
u
i
b
. 1
i i
I *
i 1
i!
I I
b* A
I I
3 1
§ i
LJJ
ii
4,6 !
-------
with glutathione is much more rapid (Fjellstedt et al. 1973; Hayakawa et al.
1975). An enzyme, glutathione-S-epoxide transferase, isolated from rat liver,
conjugates various epoxides to glutathione. Epichlorohydrin was conjugated to
glutathione at 26 percent of the rate of the standard assay substrate (1,2-
epoxy-3-(p-nitrophenoxy)propane) (Fjellstedt et al. 1973). The products of
the enzymatic reaction were not identified; therefore, the site or extent of
conjugation of epichlorohydrin glutathione was not established. (Fjellstedt et
al. 1973).
Epichlorohydrin may be enzymatically converted to 3-chloro-l,2-propanediol
by epoxide hydratase. Jones et al. (1969) observed that epichlorohydrin had
the same antifertility effects as 3-chloro-l,2-propanediol and that both
compounds resulted in the same urinary metabolite in rats, S-(2,3-dihydroxy-
propyl)cysteine (Fakhouri and Jones 1979; Jones and O'Brien 1980).
Fakhouri and Jones (1979) proposed that glycidol (2,3-epoxypropanol) was
an intermediate in epichlorohydrin metabolism in rats. This intermediate
would be formed by dehydrochlorinative cyclization of 3-chloro-l,2-propanediol.
This epoxide enzymatically couples with glutathione and is converted to the
mercapturic acid, N-acetyl-S-(2,3-dihydroxypropanol)cysteine, which is found
in the urine of epichlorohydrin-dosed rats.
The work of Jones and O'Brien (1980), however, weakened the case for gly-
cidol being an intermediate and suggested rather that glycidol would react
with chloride ion to form 3-chloro-l,2-propanediol. This reaction would not
be favored in the reverse direction (formation of glycidol). 3-Chloro-l,2-
propanediol can be oxidized to chlorolactic acid. This metabolite was identi-
fied by gas-liquid chromatography of urinary methyl ester derivatives after
36Cl-labeled 3-chloro-l,2-propanediol was administered to rats. Conversion of
beta-chlorolactic acid to oxalic acid may occur and could result in renal
toxicity due to deposition of oxalic acid crystals in the kidneys.
Jones and O'Brien (1980) also proposed that in rats 3-chloro-l,2-propanediol
could be phosphorylated to 3-chloroglycerophosphate and that this compound
might account for the antifertility effects of epichlorohydrin or 3-chloro-l,2-
propanediol. Mashford and Jones (1978) found that 3-chloroglycerophosphate
inhibited rat sperm enzyme activities (glyceraldehyde-3-phosphate dehydrogenase
and triosephosphate isomerase) and hence glycolysis. Only the S(-) isomer and
not the R(+) isomer of 3-chloro-l,2-propanediol produced antifertility or
antiglycolytic effects. Since epichlorohydrin has not been shown to have
enzyme inhibitory effects, it may be that it is metabolized in vivo to S(-)
alpha-chlorohydrin phosphate, to exert its antifertility effect.
4-7
-------
4.4 EXCRETION
The major routes of excretion of epichlorohydrin metabolites are through
the urine and via the respiratory tract. Weigel et al. (1978) found that
between 38 and 40 percent of the radioactivity from an oral dose of 14C-epichlor-
ohydrin was excreted in the urine of rats in 72 hours. Much of this radio-
label was excreted during the first 4 hours, 17.2 percent in male rats and
28.6 percent in female rats. Expired carbon dioxide accounted for 21 percent
of the radioactivity excreted by males and 18 percent by females. The rate of
conversion of label to 14C02 was initially rapid; by 4 hours, 8 percent of the
dose and by 8 hours, 14 percent of the dose appeared as expired 14C02. Less
than 4 percent of the radioactivity was excreted in the feces.
Smith et al. (1979) reported that 25-42 percent of the radioactivity of
l,3-14C-epichlorohydrin administered orally or by inhalation was excreted as
14C02, while 46-56 percent of the radioactivity was excreted in the urine by
72 hours. The epichlorohydrin was administered as a single oral dose of 1
mg/kg or 100 mg/kg of l,3-14C-epichlorohydrin or as a 6-hour exposure to 1 ppm
or 100 ppm in air. Approximately 2-6 percent of the radioactivity administered
was recovered in the feces. The urinary metabolites were fractionated into six
components by ion-exchange chromatography. None of these radiolabeled products
was unchanged epichlorohydrin, and no unchanged epichlorohydrin was present in
expired air.
The rate of excretion was calculated from plasma concentrations of radio-
activity. After inhalation exposure to air containing 1 ppm and 100 ppm epi-
chlorohydrin, the rate constants of excretion were 0.155 and 0.159 per hour,
respectively. A semi logarithmic plot of the time course of combined excretion
of radiolabel in urine and in exhaled air gives a biphasic curve, the slower
phase dominating after 24 hours. The half-lives of the fast and slow phases
of elimination were 1.5 and 26.4 hours, respectively. The calculated rate con-
stants were 0.55 per hour for the fast phase and 0.26 per hour for the slow
phase (Smith et al. 1979).
4.5 SUMMARY
Epichlorohydrin is well absorbed following oral, inhalation, or dermal expo-
sure and is rapidly distributed to the various tissues and organs. The highest
tissue concentrations were found in the kidneys, followed by the liver, pancreas,
adrenals, and spleen. Immediately following oral administration of high doses,
high tissue levels were found in the stomach, small intestine, kidneys, and large
intestine; and immediately following inhalation exposure to high concentrations,
4-8
-------
high tissue levels were found in the nasal turbinates, lacrimal glands, kidneys,
large intestine, and liver. The major routes of excretion of metabolites of
epichlorohydrin were via the urine and respiratory tract. Approximately 40
percent of the radioactivity, regardless of the route of exposure, was excreted
in the urine of rats within 72 hours. Exhaled radiolabeled carbon dioxide ac-
counted for 18 and 21 percent of the radioactive dose in female and male rats,
respectively. Fecal excretion was minor and accounted for less than 4 percent
of the dose.
When epichlorohydrin was administered orally to rats, it underwent hydrolysis
to produce 3-chloro-l,2-propanediol, which might be metabolized by two possible
pathways (see Figure 4-2). The first was by epoxidation to glycidol, then hydro-
lysis and conjugation with glutathione to produce S-(2,3-dihydroxypropyl)cysteine,
which was identified in rat urine.
The second pathway involved oxidation, first to chlorolactic acid and then
to oxalic acid, a substance known to be toxic to the kidneys. Epichlorohydrin
might also be conjugated directly with glutathione to produce S-(2,3-dihyroxypro-
pyl)cysteine or might undergo a second conjugation with glutathione to produce
l,2-(bis-cysteinyl)propan-2-ol, which also has been identified in rat urine.
It can be concluded that epichlorohydrin in rats and mice is readily absorbed,
rapidly distributed to tissues and organs, and eliminated primarily via the urine
and the lungs.
4-9
-------
-------
5. EFFECTS ON HUMANS
Several reports of human exposure to epichlorohydrin have been found in
the literature. Most were occupational exposures by inhalation or skin con-
tact. A few experimental studies were also identified. According to one
unpublished study, human exposure to epichlorohydrin has led to increased
susceptibility to respiratory tract infection, altered cerebral electrical
activity, cytogem'c changes in hematopoietic tissues, and liver dysfunction.
Epichlorohydrin is known to cause delayed skin burns, and it may cause skin
sensitization reactions.
5.1 EPIDEMIOLOGIC STUDIES
Only one report of human epidemiology was identified which was unpublished
in the open scientific literature. A single retropective mortality study
conducted on Dow Chemical Company employees has been reported by Kilian (written
communication, April 1976, as cited in NIOSH 1976a). The medical examination
records of 507 employees who had been occupationally exposed to epichloro-
hydrin for up to 16 years were examined. The majority of these employees,
however, was exposed to epichlorohydrin for 5 years or less. No control group
was included in the study, and no environmental monitoring data were reported.
The employees were classified as either having minimal or moderate exposures
based on their job titles and work histories. An attempt was made to correlate
any abnormal clinical findings with the degree of exposure. An independent
consulting firm analyzed the results of the study.
The employees' records were examined for illness (work absence for 7 or
more days) and for changes in electrocardiograms (ECG), chest X-rays, pulmonary
function tests, and clinical chemistries including urinalysis, hemograms, and
blood chemistries. Hemograms included hematocrit, leukocyte, lymphocyte, and
eosinophil cell counts. Blood chemistries included creatinine and blood urea
nitrogen (BUN) levels, albumin-to-globulin ratio, and lactate dehydrogenase
(LDH), alkaline phosphatase (AP), serum glutamic-oxaloacetic transaminase
(SCOT), and serum glutamic-pyruvic transaminase (SGPT) activities.
The illness episodes for the minimal and moderate exposure groups are
shown in Table 5-1. Respiratory illness accounted for 30 percent of the
illnesses reported for the moderate exposure group while employed in the
epichlorohydrin exposure area, but only 12 percent of the illnesses were
respiratory while workers were employed in other areas. The consulting firm
5-1
-------
concluded that the employees working in epichlorohydrin exposure areas were
more likely to experience respiratory illnesses than employees working else-
where.
Table 5-1. Illness Episodes in Epichlorohydrin Workers
No. of Employees
Total Episodes of Illness
Episodes/Employee
Respiratory Illnesses3
In Exposure Areas
In Nonexposure Areas
Minimal Exposure
213
1,343
6.3
254 (19%)
231 (17%)
Moderate Exposure
49
193
4.0
57 (30%)
24 (12%)
aRespiratory illnesses were tabulated while workers worked in epichlorohydrin
exposure areas and in other areas as well.
Source: Adapted from NIOSH (1976a).
The electrocardiographic (ECG) and chest X-ray findings were within
normal ranges as were all other clinical analyses, except for white blood cell
counts at the 4th, 8th, and 12th years of employment, and eosinophil cell
counts, which were slightly elevated during the 2nd and 5th years of epichlo-
rohydrin exposure in the moderate exposure group. The LDH activities were
elevated above normal levels in both exposure groups and the albumin-to-globulin
ratio was significantly .lower (p<0.05) in the moderate exposure group. The
consulting firm concluded from these results that, except for the increased
incidence of respiratory illness, no association could be established between
epichlorohydrin exposure and pulmonary, kidney, liver, and blood effects.
This study provides useful information concerning employee exposure to
epichlorohydrin and possible toxic effects; however, it is inadequate to
assess properly human health hazards associated with epichlorohydrin exposure.
Because of the lack of controls, the consultants compared the minimal and
moderate exposure groups. No quantitative exposure data were provided, so
dose-response relationships could not be developed. Also, no evaluative
consideration was presented for those individuals dropped from the study
because of illness, retirement, or death.
5-2
-------
5.2 EFFECTS ON THE NERVOUS SYSTEM
Fomin (1966) determined the olfactory threshold for epichlorohydrin and
examined the effects of low-level epichlorohydrin exposure on light sensitiv-
ity and cerebral electrical activity in humans. The olfactory threshold was
determined in 18 subjects (sex unspecified) who were between 17 and 33 years
old. The experimental details were not provided. The olfactory threshold was
0.3 mg/m (0.08 ppm) in the most sensitive subjects, whereas a concentration
of 0.2 mg/m (0.05 ppm) was undetected by the subjects. Light sensitivity was
investigated in four subjects exposed to 0.2, 0.3, 0.5, and 0.75 mg/m3 epi-
chlorohydrin. The experimental procedure was not described; however, for 92
dark adaptation curves obtained, no statistically significant changes in light
sensitivity were observed. Fomin then examined the effects of epichlorohydrin
on alpha-rhythm bursts. Two subjects were exposed to 0.2 and 0.3 mg/m3 epi-
chlorohydrin, and the cerebral biopotentials were recorded using an electro-
encephalograph (EEG). An epichlorohydrin concentration of 0.3 mg/m3 caused
significant changes in the voltage of the alpha-rhythm; in four subjects the
activity increased and in one subject it decreased. No changes were observed
in the subjects exposed to 0.2 mg/m epichlorohydrin. The psychological and
physiological significance of such alpha-rhythm changes is unclear.
5.3 EFFECTS ON BLOOD AND HEMATOPOIETIC TISSUE
5.3.1 Erythrocytes And Leukocytes
Sram et al. (1980) examined a group of 28 workers occupationally exposed
to epichlorohydrin for 4 years and found decreased erythrocyte counts (3.7-4.1
12
x 10 /I) in five workers, decreased hemoglobin concentrations (10.8-13.2
g/100 ml) in 16 workers, and decreased leukocyte counts (3.4-4.4 x 109/1) in
five workers.
5.3.2 Peripheral Lymphocytes
A few studies were found where chromosomal abnormalities were examined in
blood lymphocytes from workers occupationally exposed to epichlorohydrin.
These studies are described in Section 7.2.8.3.
Kucerova et al. (1977) conducted a cytogenetic study in 35 workers occu-
pational^ exposed to epichlorohydrin for 2 years (estimated air concentration
o
0.5-5 mg/m ). The percentage of cells with chromosomal aberrations was 1.37
before exposure, 1.91 after the 1st year, and 2.69 after the 2nd year of
exposure. The aberrations were mostly in the form of chromatid and chromosomal
breaks.
5-3
-------
A cytogenetic evaluation of peripheral lymphocytes from 93 workers exposed
to epichlorohydrin in the United States currently revealed an increase in
aberration rates in comparison with a 75-person group seen for preemployment
examination (Picciano 1979). Statistically significant differences were found
in the distribution of individuals with chromatid breaks, chromosomal breaks,
severely damaged cells, and total abnormal cells. The ratio of chromatid to
chromosomal breaks for the exposed group was 4:1. These findings were consis-
tent with the observations reported by Kucerova et al. (1977).
Recently, a cytogenetic analysis of cultured lymphocytes from 146 persons
occupationally exposed to synthetic epoxy resin revealed an increase in the
average frequency of cells with chromosomal aberrations (Suskov and Sazonova
1982). The synthetic epoxy resin ED-20 has epichlorohydrin as its original
monomer. Individuals of both sexes were examined, and 74 healthy individuals
having no occupational contacts with synthetic resins served as controls. The
average age of resin-exposed workers was 39.1 years, and the period of their
working with the resins ranged from 4 months to 30 years; the average age of
controls was 34 years. Controls and exposed workers were matched for sex,
smoking, alcohol consumption, and medications. The average concentration of
epichlorohydrin in the air of work areas was determined to be 1 mg/m . The
results of the analysis are shown in Table 5-2. The average frequency of
cells with chromosomal aberrations and the number of aberrant chromosomes per
cell in the exposed workers were significantly different (p<0.001) from those
in the control workers, whereas the average frequency of breaks per aberrant
chromosome did not differ significantly between the two groups.
Table 5-2. Chromosomal Aberration Frequency in Lymphocytes from
Workers Exposed to Synthetic Resin ED-20
Breaks per
Group Aberrant
Chromosomes
Average Chromosomal
Aberration Frequency
Aberrant Chromosomes
per Cell
Exposed workers
Control workers
5.5%
2.4%
0.054
0.024
1.23
1.26
Source: Suskov and Sazomova (1982).
5-4
-------
5.3.3 Immunocompetence
Thurman et al. (1978) studied the In vitro effects of epichlorohydrin on
human lymphocytes. Both T-cell and B-ce1i responses were studied. Human
peripheral blood lymphocytes were separated by Ficoll-Hypaque gradient centri-
fugation, suspended in medium plus serum, and distributed in wells of microtiter
plates. Mitogenic response was measured by incorporation of tritiated thymidine
([ H]TdR) into DNA. A variety of mitogens were used: phytohemagglutinin
(PGA-P, 0.05%) stimulated mainly mature T-cells; Cdncanavalin A (Con A, 5
ug/well) stimulated both immature and mature T-cells; pokeweed mitogen (PWM
1%) stimulated both T and B cells; and E. coli lipopolysaccharide (IPS, 10
ug/well) was shown to be a B-cell stimulant. Stimulated human lymphocyte
cultures were exposed to 0.6, 3, 6, or 60 ug epichlorohydrin per well. At 60
ug/well, there was nonspecific cytotoxicity. At the lower levels, there was a
dose-response related inhibition of the mitogenic response elicited by Con A
and PWM, but not by PGA-P. No data on human lymphocytes stimulated by IPS were
given. The data indicate that epichlorohydrin affects the immune function of
immature lymphocytes. The jn vivo effects on immune resistance have not been
studied.
5.4 EFFECTS ON THE LIVER
Schultz (1964) reported the case of a 39-year-old worker who was acciden-
tally exposed to epichlorohydrin gas from a tank with a defective closure. He
felt paralyzed for a moment and then fled outside to the fresh air. The
initial symptoms were burning of the eyes and throat that intensified after an
hour. These symptoms were followed by swelling of his face, nausea, repeated
vomiting, and severe headache.' During the night after the exposure, he became
short of breath and the next morning was admitted to the hospital. Upon examina-
tion, the mucosal lining of the upper respiratory tract was found to be irritated
and the liver was enlarged. Two dayj following the accident, he had a significantly
enlarged liver and jaundice. The serum bilirubin was 3.44 mg/100 ml, which is
almost three times the upper limit of the normal range, and urobilinogen was pre-
sent in the urine. After 18 days of hospitalization, the jaundice had subsided,
and the patient was discharged with a slightly enlarged liver. Five months later,
the patient was found to have bronchitis, elevated blood pressure, and liver dys-
function. Liver function continued to be abnormal 8 months later, when serum
bilirubin was 2.6 mg/100 ml and there were abnormal amounts of urobilin and uro-
bilinogen in the urine. The patient was examined 2 years after the exposure and
5-5
-------
found to have delayed sulfobromophthalein elimination, increased galactose
excretion, and biliary pigments in the urine. Urobilin, bilirubin, and
urobilinogen were positive. The liver pathology on biopsy was described as
diffuse, severe, fatty degeneration. Other possible causes of liver damage
and prior liver disease were explored and ruled out. It was concluded that
the liver damage was caused by epichlorohydrin exposure. Chronic asthmatiform
bronchitis was present and was also attributed to epichlorohydrin. In addi-
tion, the patient had hypertension that was considered to be unrelated to the
exposure. No other reports of liver damage in humans exposed to epichlorohydrin
were found in the literature.
5.5 EFFECTS ON THE SKIN
5.5.1 Case Studies
Ippen and Mathies (1970) described five male workers with burns resulting
from exposure to epichlorohydrin or a mixture of epichlorohydrin and methanol.
Two of the subjects were exposed twice.
A 25-year-old chemical worker spilled a mixture of epichlorohydrin-methanol
on both hands. Two days later he noticed redness and burning of his hands. On
the third evening, it had intensified so he went to an outpatient clinic.
There was severe reddening and swelling of the hands to the wrists and several
blisters a few millimeters in diameter. The patient was treated with corticoid
ointment and Ronicol tablets (3-hydroxymethyl pyridene tartrate, a vasodilator).
He returned to work 22 days later; his hands were still red 43 days after the
exposure, and he had red blisters on his wrists.
A 29-year-old male accidentally spilled epichlorohydrin on his right
trouser leg. Ten minutes later, he felt a burning sensation on the upper
thigh of his right leg and observed mild reddening. He treated himself with
an anesthetic ointment and continued to work, but 62 hours after the accident
he went to an outpatient clinic because the reddening and burning continued to
increase. There were two areas of deep redness the size of the palm of the
hand and several smaller spots on the anterior surface of the thigh. The skin
over the areas appeared dry and tanned. The worker was treated as an outpatient
with antibacterial salves. He experienced no serious pain and although he
still had moderate residual redness, he was able to return to work 9 days
after the exposure.
A third case involved a 19-year-old male chemical worker who spilled pure
epichlorohydrin on his left shoe. Six hours later, he noticed red spots on
5-6
-------
the dorsum of his*foot. Burning, itching, blisters, and skin erosion developed;
the blisters were opened by a physician, and the man was treated by application
of a topical anesthetic and a corticoid salve. The reddening intensified,
and the man was admitted to the hospital 10 days after exposure. A severe
skin erosion 5 cm in diameter was observed on the dorsum of the foot. Lymph
nodes in the left groin were painful and enlarged. Temperature was slightly
elevated and antistreptolysin titer negative. Staphylococcus aureus was
cultured from the skin erosion. He was effectively treated .with penicillin
injections, compresses, and corticoid salves. The patient was discharged from
the hospital 1 month after exposure. Nearly 2 years later, this same subject
worked for 3 days with epichlorohydrin while wearing protective rubber gloves,
onto which he spilled the chemical. During the night of the third day, he
noticed burning, swelling, reddening, and blister formation on several fingers
of both hands. The patient was admitted to the hospital the next day. After
treatment with metal foil bandages and bland salves, the lesions lessened. The
patient was discharged from the hospital 10 days after admittance and returned
to work 20 days later. In a followup examination 8 days after discharge, his
fingers still showed persistent redness.
A fourth case involved a 32-year-old male chemical worker who accidentally
poured an unspecified amount of epichlorohydrin into his right safety shoe.
Even though he removed the shoe immediately and rinsed his foot with lukewarm
water, a spotty redness developed over the ball and base joint of the large
toe. He was admitted to the hospital within 2 hours and treated with saline
solution compresses. The symptoms lessened and he was discharged after 5
days. Eight days later, he spilled epichlorohydrin into his left shoe.
Despite being aware of a slight burning sensation in his foot during the
night, he worked the following day. On the 3rd day after the accident, a
blister developed. He was admitted to the hospital with reddening and swelling
of the left foot and a blister of 2 x 1 cm filled with yellow fluid. During
his hospital stay, peripheral arteriosclerosis with hyperlipidemia was diagnosed.
The values for total lipids, e'sterified fatty acids, and triglycerides were
elevated two to threefold above normal. It was concluded that no causative
relationship between exposure to epichlorohydrin and the arteriosclerosis with
hyperlipidemia could be determined. The symptoms regarding the affected foot
subsided after local treatment with saline solution and ointments. The patient
was released from the hospital after 14 days. He returned to work 4 weeks
after the accident,
5-7
-------
The fifth case involved a 21-year-old male worker who wiped up approximately
1.5 liters of an epichlorohydrin-methanol mixture (4:6) from a laboratory
floor. Even though he washed his hands with soap and water, he experienced
redness and itching on the palms. Four days after the incidenti he began to"
apply a corticoid salve. He was admitted to the hospital 2 days later complain-
ing of intense itching on both hands and red and swollen fingers. During his
11-day stay, he was treated by intravenous injections of a saponin mixture
(Reparil) and by application of a heparinoid salve. The redness and swelling
gradually diminished, but the surfaces of his hands were hard and rough. He
did not return for followup examinations.
In the two cases in which the patients were involved in two accidental
exposures, Ippen and Mathies (1970) stated that there were no signs of sensiti-
zation and referred to the skin effects as protracted chemical burns that did
not develop as quickly as acid or base burns. Since the burns had a latent
period of several minutes to several hours, they were more similar to burns
produced by X-ray or ethylene oxide. Since epichlorohydrin can penetrate
rubber or leather, specific work precautions are necessary. The severity of
the burns depend on the duration and extent of exposure; therefore it would
appear that there is a longer latent period for appearance of symptoms when
methanol-epichlorohydrin mixtures are the causative agent (2-4 days) rather
than epichlorohydrin alone.
5.5.2 Sensitization
Ippen and Mathies (1970) did not find sensitization in patients who had
two exposures to epichlorohydrin. Only one human sensitization experiment
(Fregert and Gruvberger 1970> as cited in NIOSH 1976a) was found in the litera-
ture. This study involved only one subject and experimental details were
lacking; thus, no general conclusions could be drawn concerning skin sensitiza-
tion in humans. However, some studies examining occupational eczema indicate
that sensitization reactions may occur after chronic exposure to plastics and
solvents containing epichlorohydrin causes skin sensitization reactions (see
Section 6.1.1.6).
Jirasck and Kalensky (1960) studied patients with occupational eczema.
All of the 57 patients studied were found to be hypersensitive to epoxide
resins (a 20 percent solution in acetone); 23 of the 57 had weak-to-moderate
reactions in skin tests with a 1 percent solution of epichlorohydrin.
5-8
-------
Fregert and Gruvberger (1970, as cited in NIOSH 1976a) studied sensitiza-
tion to epichlorohydrin and cross-sensitization to propylene oxide in one
subject. Patch testing with 0.1, 0.5, and 1.0 percent epichlorohydrin in
ethanol gave positive reactions after 8-11 days. Solutions of 0.1 and 0.01
percent epichlorohydrin gave positive patch tests after 2 days. Propylene
oxide was positive at 0.2 percent; negative cross-sensitization results were
found with chloropropane, l-chloro-2 propanol, and ethylene oxide.
Lambert et al. (1978) presented four classes of occupational eczema where
there was an allergic skin reaction to epichlorohydrin. The first case involved
a subject who had worked in a chemistry laboratory for 16 years and was exposed
to several chemicals including resins and epichlorohydrin. He had eczematous
lesions on his hands that spread to his forearms and legs. The condition sub-
sided when he was on vacation and was aggravated when he returned to work.
Epicutaneous testing with 1 percent epichlorohydrin gave a strong-positive
reaction; 0.5 percent epichlorohydrin gave a weak reaction. A second worker
who molded epoxy resins developed eczema and had a moderate skin reaction to 1
percent epichlorohydrin. A third worker who had eczema on both hands had
strong-positive allergic skin reactions to furan resin and epichlorohydrin. A
fourth worker who had eczema on the fingers and backs of his hands manufactured
fiberboard. He was not allergic to the material alone or to the solvent alone
but had a positive reaction to both. The solvent was found to contain epichlo-
rohydrin (extracted from the material). A patch test for epichlorohydrin was
also positive.
5.6 SUMMARY
Epichlorohydrin has been shown to cause respiratory, skin, and eye irrita-
tion in humans. Most human exposures reported in the literature were employment
related. In a single retrospective mortality study by Kilian (written communica-
tion April 1976, as cited in NIOSH 1976a), medical records for 507 employees
who had been occupationally exposed to epichlorohydrin for up to 16 years were
examined. Although the available information in many aspects of this study
was limited, the study indicated an increase in acute respiratory illnesses in
employees working in epichlorohydrin exposure areas; no relationship was noted
between epichlorohydrin exposure and pulmonary, kidney, liver, and blood
effects. High-level accidental exposures have produced pulmonary and liver
changes in humans. In a severe epichlorohydrin inhalation poisoning, initial
irritation of the eyes and throat was followed by chronic asthmatic bronchitis
and extensive fatty infiltration and degenerative changes in the liver.
5-9
-------
Headache, nausea, and head and chest congestion have been reported following
worker exposure to epichlorohydrin. Local dermal contact has been shown to
cause severe skin irritation. Skin burns have been reported from accidental
exposures, and a few cases of skin sensitization reactions have also been
reported.
Cytogenetic studies of workers exposed to epichlorohydrin have produced
evidence for clastogenic effects on lymphocytes. In a recent study by Sushov
and Sazonova (1982), where cultured lymphocytes from 146 workers occupationally
exposed to epichlorohydrin resin were examined, the average frequency of cells
with chromosomal aberrations and the number of aberrant chromosomes per cell
increased significantly over controls. Epichlorohydrin should be considered
as potentially hazardous to humans as a result of its mutagenic action in
experimental systems and its potential to induce chromosomal effects in humans.
Limited epidemiologic studies have not demonstrated epichlorohydrin to be
carcinogenic to humans; however, in long-term animal studies the compound has
been shown to induce local sarcomas in mice receiving subcutaneous injections
and to induce squamous cell papillomas and carcinomas of the nasal epithelium
in rats exposed by inhalation. Further study of the potential carcinogencity
of epichlorohydrin to mammalian species is basic to any analytical health
assessment of this compound.
5-10
-------
6. ANIMAL TOXICOLOGY
6.1 SPECIES SENSITIVITY
Most of the toxicologic information on epichlorohydrin concerns acute
exposure either by inhalation or by the oral route. Only a few investigators
have examined the subchronic and chronic effects of epichlorohydrin exposure
in laboratory animals. No individual strain or species differences in
sensitivity have been observed in the studies examined. Epichlorohydrin
has been found to be extremely irritating to skin, nasal mucosa, and eyes
upon acute exposure. The target organs or tissues to the subchronic effects
of epichlorohydrin, in descending order of sensitivity are the nasal mucosa
(when administered by inhalation route), kidneys, liver, cardiovascular
system, skin, and muscle.
6.1.1. Acute Toxicity
Acute exposure to epichlorohydrin causes systemic toxicity, and,
regardless of the route of exposure, results in a similar sequence of
symptoms. Animals exposed to high doses of epichlorohydrin show central
nervous system depression with death occurring due to paralysis of the
respiratory center. A summary of the acute toxicity data is given in Table
6-1. ,
6.1.1.1 Inhalation—Carpenter et al. (1949) exposed groups of six male or
female Sherman rats weighing 100-150 g to epichlorohydrin at a concentra-
tion of 250 ppm (950 mg/m3 for 4 hours). The authors found that from two to
four rats died in each group during exposure. The animal deaths in this
study were listed as ranges rather than as specific numbers of deaths.
Smyth and Carpenter (1948) and Weil et al. (1963) (apparently reporting the
same data) indicated that four of six Sherman rats died after exposure to
epichlorohydrin at a concentration of 250 ppm for 8 hours (time of death
not specified). In these range-finding studies, the epichlorohydrin vapor
was produced by injecting liquid at a metered rate into a heated Pyrex
evaporator tube supplied with metere'd, forced air. The vapor was then
cooled. The animals were exposed in a desiccator connected to the evapo-
rator. The concentrations of epichlorohydrin were calculated on the basis
of the rates of liquid delivery and airflow. In-these studies, no quanti-
tative analyses were performed on the vapor in the exposure chambers.
6-1
-------
I
1
CJ>
o
UJ
U,
tu
o
«»-
cu
C£
•P
O
It 1
cu
to
o
a
cu
'u
cu
cu
JJ
i
en
t-i
•
<0
0)
s-
1
CO
e-
(0
o
(0
CO
TJ
1
CM
JC
«*•
t-
s.
a.
0
in
CM
5
O
•r~
2
c
1— 1
^••^
CO
v^
•
I-M
•P
CU
r— •
•1—
T3
(0
-§
*e
«?
JC
cp
1
a.
o.
o
CM
CO
C£
C
o
'•§
"to
1— I
o
CD
cn
rH
N^g/
(0
CU
1
•p
•p
o
JC
CM
S.
£
1
3
CO
1
CO
CM
&
C,
o
1
JC
c
t-i
t~\
o
CO
en
rH
^
(d
c'
15
(A
•J™*
(O
o
€
n
in
j—
CO
S-
g.
o.
CD
CO
(0
OS
o
's-
C3
5*
(0
£
o
E
H
in
JC
o
1
CO
CO
to
o
•p" •
1o
(0
I— 1
o
r-l
^«^
CO
CU
cu
1?
to
S-
cl
CO
-
•
s-
M-
E
Q.
0.
S
r-T
i
co
CD
•V
i-H
CO
as
c
o
CO
co
c
1—4
CD
CD
i"H.
^^
CO
CO
E
1
C
o
"cO
j_
1
0
y
'(O
-P
cu
-J
CM
S-
o
E
a.
o.
s
i
i— i
' *
r-l
CO
C£
C
O
CO
to
c
>— 1
CD
T^
^^
CO
CU
1
JC-
cu
!—•»
•p
0
z:
JC
<0
U
o
U
as
•P
at
CO
CO C
s- "-
•P
c 5?
CU -P
CO O CO
JC U -P
•P S-
CU r— O
r- tO E
J-
CM
a.
a.
en
•a
CD
CD
3
•c
o
cu
3
c
o
•r"
ts
6-2
-------
TJ
CU
C
'•P
C.
o
u
r-5
1
to
UJ
1
CO
o
re i—
•r- JC
s- u
0
cu re
P C
Q *r™
O. 3
re cu
•i- O
S- 3
3 -a
*o
a_
O)
I*.
to
to
re
*re
0
re
•r~
E
i.
flU
i^
TJ
re
re
m en
Er-l
JZ
to
cu"
TJ
"s-
« o cu
rei— c
•r™ ad »f-™
S- 0 C
3 •«-
<= >vP
cu re cu
•PCt.
O'r- O
(- S-
Q. 3 TJ
CU
"TJ V)
re cu re
&•• Z3 S—
3 TJ 0
>>CU C
1 S- •!-
o
o.
3
^
LO
CM
rH
re
2
o
re £?
C x^> /~s «^-
•r- CM CM <7>
'"^ £ CT> CJ> t^^/
S £ si vi CO
cn re ^
r™i i£? • /**s • /*"> re
V^ r— O r-^ O CU
•o re to re to —i
C OV
«— re P T-I +j t-t TJ
re cu **^> cu •>>> c
re re
^-> >>/-> cu re cu re
cu re vo o > o > s-
JZ V> C7V " CU £Z CU C "O
P> EfH ^ E S- S 3
E -C re i- re i- s-
co co — i :*: — i :*: u_
CO
c
re" +J
*r- re
c re b
5*E c
o 3 re
O.TJ W
CO CO i— i—
« w T3 re re
re re *.r- ' j^ _c
•r- CU S- +J 4J
•p >» c: x: o o o
o o o a a o o
0)
•^4.
"X^
O)
O) O5 O3 O) ,O) O)
S S E E E E°
CO O O CO t£> O r-t
«:»• m
• «»^*
/•^SCn
O r— 1
tOX^
cr>
r-l CO
"^ re
re cu
s_
•o
o
Q
O)
cn
s
,— ,
CD
LO
T3
C
re
0"
m
in
CM
co
cu
to
3
S
S
O
CM
^^
^^
r-i
t
*«
CU
U
1
ra
1
O
u>
cn
cn
3
•P
02
r—
re
cu
c
o
+J
•^—
cu
a.
re
t__
)N—
-P
1— <
/~**
CM
CT>
r-^
9
*«
cu
cu
u
s.
1
O
CO
1?
^
S
CO
o.
re
cu
•r-
«J
're
cu
c
o
-p
'C
cu
a.
re
P
i— i
CM
at
H
^
*— *
re
CO
cu
1
ro
_|
O
US
5*
*!"
s
_i_%
•r—
J3
^
r~-
re
cu
o
si
a;
1
•p
re
•r-
E
j_
^^
•a
re
re
>>•->
re vo
^ 10
^»
v>
^
(H
"
c!
3 >,
*5 «
O *r~
S- i-
O. 3
tO O>
^ "*t
3 TJ
^2
o
cn
E
•P
^
tA
3
C
1
U
.a
to
CO
•p-
*T
r—
U
6-3
-------
O)
O
c
(U
O)
ID
>Ur-
,1 * iJ
• pM **
«— CO !•>
lO T— "O
4-> S- 0)
S- 3 U
o a>3
E'r- T3
i— OJ
&§ o s-
t^
^o
!>
•17-
•a
CO
0)
3
1— •
•\
13 in
<1) Q)
o en
3 C
CU JZ
s- a
•o
0
0
*!5
CO
•r- •»
S-. in
3 at
c -a
'55'sl
4-> 0
O r-
S- -C
0.0
" ^?
•r- fO
s- c
3 «r-
O
a.
atinine
and urine ere
i ncreased
u
ardmatii
CU
4-
T3
O)
in
to
CU
o
c,
1— 1
1
U)
•r"
"O >>
CU -P
in •<—
amines, decre
taminase acti
-a
CU
decreas
lood pressure
i
CQ
C
O
1
concent
in
0)
£
£_
C
-a
CU
decreas
lood pressure
00
c
o
•r-
1
concent
i— •
to
•p
CU
e
e
•r-
C
IE
CL)
10
O
in
O)
•r—
O
(U
CL
CO
O)
CD
O)
o
CO
o
LT>
1-1
OJ
^
>v
O)
O)
05
e
o
m
CM
o
o
If)
cn
co oo
-------
TD
CD
3
C
•r™
•P
C
O
u
rH
1
1 1 1
CQ
h-
CO
(J
C
CU
CU
CD
P
U
0)
f-
LU
CO
(A
O
O
(A
0)
•r—
U
CD
a.
to
0)
o
0-
rH
CT>
rH
0)
re
cu
j
T3
C
re
t-
cu
•o
3
a>
s_
LL.
"re
.c
•p
cu
•P
o
o>
^
*^x
1
m
o
-P
re
cu
r—
O)
re •!-
E (A
cu
o
c
0
•I—.
•p
re
o
a.
re"
rH
CD
i— 1
CU
re
|
•o
re
s-
*>x
^^
OJ
E,
O
00
rH
i-T
4*)
re
Q£
cu
"o)
re T-
E (/>
cu
0
e
o
• [—
re j
o
a.
a.
re
/»^
rH
CT> /*™S
rH f>.
CD
CU H
re
cu
1 ^—
«
C -P
re cu
t- rH
CU CO
C CT>
OJ v»»^'
^
•o
re
cu
-a
c§
o
rH
1
1
cu
-4
T3
C
re
i.
cu
•a
3
CU
u_
"re
i ^
cu
-p
o
jC
rH
S-
o
"^
C3>
V^
C3>
E
O
^-
*
D)
ol
re
cu
c
3
CD
"re
0)
a
CM <"-.
Co co
rH CT>
x- 1
"re •— '
re
CU -P
cu
•P r—.
E CO
co SB
c
o
•r-
"re
-p '
<^
^
•P^
•a
•r-
D)
E
00
H
rH
jP
U3
re
1
cu
a
/••^
r*-
CD
rH
v"'
.
r—
re
•p
cu
re
j—
re
"E
0
cT c
o re
•r- SZ
•P -P
re
3 (A
•a co
r—
aT >,
.^."r?
(A Ul
C
C CO
0 P
•c- C
10 •!-
CO
f
CM
i.
o
**~
D)
E
V£>
CO
CM
M
CO
rH
rH
-P
^
f->
re
're
CO
o
/*••%
r>-
CD
rH
v"~^
,
r—
re
-P
cu
CO
-o
re
i—~
re
Q-
W
•r-
W3
^f
U
CO
n
re
E
CO
S
,_
CM
s_
t^.
O]
•
o
CD
in
^j
«^_
f^
l">
re
1
CO
a
-------
CU
o
CU
CU
CD
02
(continued)
Effect
*
t-H
CD
LU
CO
H- CU
U>
o
U)
CU
•r-
O
CU
a.
CO
o
•P
o
on
CM
S
*
15
•p
CU
Lawrence
o
in
3
jz
^5
CD
l^^
tn
in
-P
•r-
.D.
CC
CU
o
CvT-->
o
s
r-T
-P
•r"
.a
1
15
CU
o
CM CM >> g g £ £
s § 1 a § § . s g
O O Oi ID v~' v--' ^ v~'
r— CT>
• • tO ' ' * * * ^_
15 15 15 15 m iHCUUOOU
c C CUCD CCCC
£ 1 IB f 1 1 1 1
5 5 |^dJ3J355
c
C 0
•o o cu «i- <—
-P •«
m CU r— »r- CU •!- CU
jS ^2 eu s- w s- a.
S- S- 5- -t-* *" ^ rrt
>'r"SS^
ES SS § « - «l §
00 00 0 +> rt ^^'43
+» •» *» ^ , £^ § " Id
C», C-> S CUCU ^+JC
oc oc w "a ^^S'C .2
TJ.? 5-? S. i 15 5-2 t 5
ta^-o (O+a w °- '£•£•'" 5
4J«OCU 4J<« 0 f. ^mo^W -^
T^ld T."S a * u»i c«ti S E
ifc-S £fc -oL ^ 5-S -^t 5 ^
.r- S_ «r- CU 18 *r— CU C •!— J3
S S 1 c^^cSil
in -a
. O> r— *>
81 ^ 1 *«V '5"S- "S>> r- - - -
9^2 §S^"S ^"§ ^= =0 c'5 co o
CO CM -P O 0 LO CU r— 0 '-"'£? ''""•,-, '*" -rl
OS«T- o^es«-p tO'>- to • c c g c
25 aaino »>cur- «>cu -o "0 ~o ~ o
•iSSr- l§Pi0 §-£u §•£ IS E§ -iS -i-S
CMOin'5 CMOO.^ -OUW "OU rHU r-jO rH .0 r-IU
d d rH t-iddoo
•«-» -^ ** +*.+*.•££:}:!
I i i I 1 1 2 n
(O ic f njnjnjioro
15
j-
CU r— r- r— i — <— >"~
_. -S (O tOfOfOcOfO
15 «
-------
The median lethal concentration for acute epichlorohydrin inhalation
exposure was determined by Laskin et al. (1980). Groups of 20 male Sprague-
Dawley rats were exposed to epichlorohydrin concentrations ranging from 283
to 445 ppm (1,075 to 1,691 mg/m3) for 6 hours. The animals were then
observed for mortality over 14 days. No deaths were observed at 283 ppm.
Only one death (5 percent) was observed at both 303 and 339 ppm (1,151 and
1,288 mg/m3). At higher concentrations, mortality increased sharply; at
369 ppm (1402 mg/m3), 15 of the 20 exposed animals died (75 percent) within
14 days after exposure. At 421 and 445 ppm (1,600 and 1,691 mg/m3), 16 and
17 animals died, respectively. From these data, the 14-day LCS€> for epi-
chlorohydrin was estimated to be approximately 360 ppm. Pathologic examin-
ation of four animals from each exposure group revealed acute respiratory
tract irritation, hemorrhage, and severe pulmonary edema. The lung-to-body
weight ratios were determined, and marked elevations were observed at the
higher exposure levels. For example, at 369 ppm epichlorohydrin, an 80
percent increase in lung-to-body weight ratio was detected when compared
with controls. At the lowest two exposure levels (283 and 303 ppm), there
were no increases over controls in lung-to-body weight ratio.
Freuder and Leake (1941) exposed white mice (unspecified strain, sex,
and age) to epichlorohydrin by inhalation at concentrations of 2,370 ppm
for 60 minutes and at 8,300 (31.54 g/m3) and 16,600 (63.1 g/m3) ppm for 30
minutes. In the group of 30 animals exposed at 2,370 ppm, no mortality was
observed within 24 hours after exposure. However, in the group of 20
animals exposed at 8,300 ppm and in the group of 30 animals exposed at
16,600 ppm, all the animals died. All animals exposed to epichlorohydrin
showed irritation of the nose and eyes. "Delirium" was observed 3 minutes
after exposure began at 16,600 ppm and within 14 minutes after exposure to
8,300 ppm. At the two highest exposure levels, the animals first became
quiescent and then developed cyanosis and muscular relaxation of the extrem-
ities. This was followed by tail stiffening and fine tremor of the body.
The respiration became increasingly depressed. Some animals experienced
clonic convulsions. The animals exposed to epichlorohydrin at 2,370 ppm
showed no symptoms of toxicity other than nose and eye irritation.
6-7
-------
Kremneva (1960) and Kremneva and Tolgskaya (1961) studied the acute
inhalation toxicity of epichlorohydrin in both white rats and mice (strain,
age, weight, and sex unspecified). The rats and mice were exposed for a
single 2-hour period in a 100-liter chamber. The epichlorohydrin was
placed in the chamber as a liquid and allowed to evaporate. Air samples
were withdrawn from the chamber between 15 and 30 minutes and then again at
90 minutes for analysis. After reaching a maximum value, the concentration
of epichlorohydrin fell to nearly half of the initial value during the
2-hour exposure period. The animals were observed for 14 days following
exposure. The results of these studies are shown in Table 6-2. All deaths,
except for that of one mouse, occurred within the first 3 days after exposure.
Table 6-2. Summary of Mortality Findings in Rats and Mice after
Acute Inhalation Exposure to Epichlorohydrin
Range of Concentration
rng/nr ppm
899-1,199 237-316
2,242-3,587 590-944
4,036-5,381 1,062-1,416
5,381-6,278 1,416-1,652
6,726-8,071 1,770-2,124
No. of Test
Rats
15
18
10
10
10
Animals
Mice
15
20
15
10
10
Mortal
Rats
0
55
80
100
100
ity (%)
Mice
0
40
93
100
100
Source: Kremneva (1960); Kremneva and Tolgskaya (1961).
The authors stated that rats and mice appeared to have essentially
identical sensitivities to inhaled epichlorohydrin vapor. The lethal
concentration for both rats and mice was 1,416 ppm (5,381 mg/m3) and the
LC56 ranged from 590 to 944 ppm (2,240 to 3,587 mg/m3). The maximum con-
centration that produced no observable signs of toxicity was 316 ppm
(1,199 mg/m3). Epichlorohydrin caused irritation of the mucous membranes
of the upper respiratory tract, initial stimulation followed by depressed
activity, increasingly depressed respiration, and dyspnea resulting in
asphyxia. Cutaneous hyperemia and areas of subcutaneous hemorrhage were
6-8
-------
observed. No "loss of righting reflex was observed during the 2-hour expo-
sure periods. Death from progressive respiratory dysfunction occurred
several hours following exposure. Microscopic examination of tissues and
organs from dead animals revealed inflammatory desquamative bronchitis,
necrosis of the bronchial mucosa, and pulmonary edema. The kidneys showed
degeneration and necrosis of the convoluted tubules and glomerular edema.
Hemorrhagic changes were observed in the n)ucosa of the stomach and the
small intestine. Sections of the myocardium showed fibers that were dis-
organized and fragmented.
Lawrence et al. (1972) determined an LT5o (lethal time 50 percent) for
male ICR mice exposed to air saturated with epichlorohydrin vapor. Groups
of mice were placed in an 8.75-liter glass chamber. The air in the chamber
was saturated with epichlorohydrin by bubbling air through liquid epichloro-
hydrin and then passing the air into the chamber. The concentration of
epichlorohydrin in the chamber was calculated by dividing the weight loss
of the liquid by the quantity of air passed through the liquid. Groups of
mice were exposed for specific time intervals, and then observed for 7
days. The LTso was determined to be 9.13 minutes, with 95 percent confi-
dence limits of 8.49-9.81. At 9.13 minutes, the exposure chamber should
have reached an 88 percent equilibrium with the saturated vapor entering
the chamber. At a room temperature of 23°C and a barometric pressure of
30.18 inches of mercury, the epichlorohydrin concentration was calculated
to be 71.89 mg/1 (18,907 ppm), with a maximum deviation over three separate
exposures of 1.26 mg/1.
6.1.1.2 Oral--Freuder and Leake (1941) examined the acute oral toxicity of
epichlorohydrin in mice. Epichlorohydrin was suspended in a 25 percent
aqueous gum arabic solution and mixed. Each animal received the same dose
volume based on body weight (0.1 ml/10 g). Groups of 15 white mice (un-
specified strain, sex, and age) were administered either 0.50 or 0.23 ml/kg
(588 or 270 mg/kg) epichlorohydrin by stomach tube. Immediately after
administration of 588 mg/kg epichlorohydrin, the mice showed intoxication
(erratic movements) for a few minutes, then the erratic movements ceased
and respiration slowed. A dose level of 588 mg/kg was lethal to all 15
test animals. At a dose level of 270 mg/kg, all 15 test animals survived
the 24-hour observation period.
6-9
-------
Kremneva (1960) and Kremneva and Tolgskaya (1961) examined the acute
oral toxicity of epichlorohydrin administered by gavage in 30 mice and 15
rats (strain, age, weight, and sex unspecified). Epichlorohydrin was
administered in aqueous solution at 250, 325, and 500 mg/kg. A dose of 250
rag/kg did not produce any observable signs of toxicity during 2 weeks of
observation. The two highest doses produced mortality in both rats and
mice usually within the first 48 hours after treatment. The signs of
toxicity observed were lethargy, slowed respiration, subcutaneous hemorrhage,
dyspnea, rales, ataxia, and tremors. Gross examination of the dead animals
revealed hyperemia and hemorrhage in the lungs and other organs, and a
yellow discoloration of the liver. Microscopic examination showed hemorrhages
and edema in the pulmonary tissues, degenerative changes with areas of
necrosis in the convoluted tubules of the kidneys, and fatty degeneration
of the liver. Foci of necrosis were also observed in the mucosa of the
stomach and intestine.
Lawrence et al. (1972) determined the acute oral LD5(j of epichloro-
hydrin in male ICR mice and in male Sprague-Dawley rats to be 0.20 and 0.22
ml/kg (235 and 260 mg/kg), respectively. The 95 percent confidence interval
was 0.16-0.25 ml/kg in mice and 0.12-0.39 ml/kg in rats. The epichlorohydrin
in this study was administered by gavage in cottonseed oil.
Smyth et al. (1962) and Weil et al. (1963) determined the oral LD56 of
undiluted epichlorohydrin in male Carworth-Wistar rats, which were 4-5
weeks old and weighed 90-120 g. Mortality observations were made for 14
days after compound administration. The LD50 was determined to be 0.21
ml/kg (260 mg/kg). No statistical information was provided.
6.1.1.3 Subcutaneous Injection—Several investigators have examined the
acute toxicity of epichlorohydrin by subcutaneous administration. Kremneva
and Tolgskaya (1961) administered epichlorohydrin subcutaneously to 50 mice
(strain, age, weight, and sex unspecified) at 125, 250, 375, and 500 mg/kg.
The 125 mg/kg dose was tolerated and the animals showed no observable
behavioral changes. A dose of 250 mg/kg was lethal to 7 of 10 animals.
Survival time was not described. The 375 and 500 mg/kg dose were lethal to
all the treated animals.
6-10
-------
Shumskaya and Karamzina (1966) reported the toxic effects of epichlor-
ohydrin in the kidneys from a single subcutaneous injection. It was the
intent of the authors in this study to investigate methods for detecting
kidney dysfunction and not to examine the role of epichlorohydrin in renal
toxicity. For this reason, there was little information provided concern-
ing the experimental details such as the number of animals treated, animal
observations, sample times, and the length Of time before the onset of
toxic effects. The authors also did not report the methods used for many
of the clinical determinations. Rats (strain, sex, and age unspecified)
were administered epichlorohydrin subcutaneously in single doses of 65,
125, or 250 mg/kg. Several parameters were examined in this study; how-
ever, not all parameters were reported for each dose level, so dose-
response relationships could not be identified. Generally, the authors
found that epichlorohydrin administered subcutaneously produced polyuria,
decreased urinary specific gravity, proteinuria, decreased urinary chlorides,
increased kidney-to-body weight ratios, and increased nitrogenous substances
in the blood.
Rotaru and Pallade (1966) and Pal lade et al. (1967) described gross
and microscopic findings in rats following subcutaneous injection of epi-
chlorohydrin. A total of 37 albino rats weighing 180-200 g (strain, sex,
and age unspecified) received a single subcutaneous injection of either 150
mg (23 rats) or 180 mg (14 rats) of epichlorohydrin. No control animals were
mentioned for this study. Animals were killed at 24 hours, 48 hours, 5
days, and 10 days after treatment. The number necropsied was not speci-
fied. The following tissues were examined: heart, lungs, kidneys, liver,
adrenals, spleen, stomach, intestine, and brain; most affected were the
kidneys. The changes observed in the kidneys at both dose levels were
qualitatively similar; however, more severe changes were observed at the
higher dose level (180 mg). At 24 hours after treatment, the rats examined
showed kidney toxicity consisting of ischemia of the cortex, and congestion
of the medulla with marked 'interstitial edema. Degenerative changes were
observed throughout the tubules with necrotic lesions observed also in the
proximar convoluted portions. Signs of regeneration were observed in the
kidneys 5 days after treatment. Ten days after treatment, only a few signs
6-11
-------
of ischemic necrosis were observed, and most of the tubular integrity was
restored. The changes in the other organs were not as severe as those
observed in the kidneys. The lungs showed areas of congestion of the
alveolar septa, desquamative bronchial inflammation, and some edema of the
bronchiovascular connective tissue. The heart tissue was normal except for
some limited myocardial congestion. The spleen showed stasis and some
limited hemorrhage, except in one animal, where hemorrhaging was extensive.
The stomach and intestines showed slight mucosal congestion and edema. The
liver and adrenals appeared normal except for some limited congestion in a
few animals (number unspecified).
6.1.1.4 Intraperitoneal Injection—Lawrence et al. (1972) reported LD5o
values for several animal species for the intraperitoneal injection of
epichlorohydrin. Male ICR mice, male Sprague-Dawley rats, male Hartley
albino guinea pigs, and male New Zealand albino rabbits (number, age, and
weight unspecified) were treated with epichlorohydrin dissolved in cotton-
seed oil. The LD56 values and confidence limits are shown in Table 6-3.
6.1.1.5 Intraveneous Injection—Freuder and Leake (1941) studied the
effects of epichlorohydrin injected intravenously in cats and dogs. Three
cats and two dogs (sex, age, and weight unspecified) were anesthetized with
sodium pentobarbital. The blood pressure was measured from the carotid
Table 6-3. Acute Intraperitoneal Toxicity of Epichlorohydrin
LD™ 95% Confidence Limit
Species (mg/Kg) (mg/kg)
Mouse
Rat
Guinea pig
Rabbit
170
113
118
160
153-188
94.6-134
29.5-472
83.6-306
Source: Lawrence et al. (1972).
6-12
-------
artery. The respiration was recorded directly from the trachea by measur-
ing pressure changes. Epichlorohydrin was suspended either in water or
acacia solution before injection. The doses administered to each animal
were not reported nor was the suspension vehicle nor the concentration of
epichlorohydrin in the vehicle. The authors reported that cats and dogs
showed similar blood pressure responses to epichlorohydrin administered
intravenously. Doses of epichlorohydrin below 9.3 mg/kg were essentially
inactive in affecting blood pressure or respiration; at 9.3 mg/kg, there
were only transitory decreases in blood pressure. The minimum lethal
concentration was approximately 93 mg/kg in both dogs and cats. Immedi-
ately after injection of a 93 mg/kg dose, there was a rapid decrease in
blood pressure followed by a moderate increase. Respiration increased and
deepened in the cats, whereas in the dogs there was a brief period of apnea
and then an increase in respiration rate. Death occurred in both the dogs
and cats within 2 hours.
6.1.1.6 Percutaneous Application—Epichloronvdrin has been shown to be
irritating to the skin and is readily absorbed to cause systemic toxicity.
Several investigators have examined the toxicity, irritation, and sensiti-
zation potential of epichlorohydrin in laboratory animals. Freuder and
Leake (1941) studied the acute percutaneous toxicity of epichlorohydrin in
white rats (strain, sex, age, and weight not specified). The abdomens of
the rats were shaved, and a 1-cm square piece of gauze wetted with a measured
amount of epichlorohydrin was applied to the shaved area. The gauze was
removed after 1 hour. Groups of 10 rats each were exposed to 0.5 ml/kg
(6.5 mmol/kg) and 1.0 ml/kg (13.0 mmol/kg) epichlorohydrin. Twenty rats
were exposed to 2.0 ml/kg (26.0 mmol/kg). The observation time was un-
specified; it is assumed to have been several days. At the lowest exposure
level (0.5 ml/kg), all 10 rats survived. At the intermediate dose level
(1.0 ml/kg), 8 of 10 rats survived; and at the highest dose level (2.0
ml/kg), only 2 of the 20 exposed rats survived. The authors noted dis-
coloration of the skin after exposure, with occasional superficial desqua-
mation within a few hours.
Smyth et al. (1962) and Weil et al. (1963) determined the acute dermal
LD56 in rabbits. Groups of four male albino New Zealand rabbits weighing
2.5-3.5 kg were immobilized for a 24-hour contact period. The fur was
6-13
-------
clipped from the skin, undiluted epichlorohydrin was applied, and then the
skin was covered with an impervious plastic film. After a 24-hour contact
period, the film was removed and the rabbits were observed for 14 days.
The approximate percutaneous dermal LD50 for rabbits was 1.3 ml/kg. From a
toxicologic viewpoint, great care should be exercised in the interpretation
of results from such experiments because restrained, conscious animals
usually have high levels of catecholamines in their circulation. It is
well known that catecholamines affect the cutaneous vascular tone and,
hence, the rate of absorption of substances applied to the skin.
Kremneva and Tolgskaya (1961) studied the skin absorption of epichlor-
ohydrin in mice. The tails of 20 mice (strain, sex, and age unspecified)
were immersed to three-quarters of their length in epichlorohydrin for
either 1 hour for a single exposure or for 20-30 minutes, 2-3 times, for a
repeat exposure. The single l^-hour exposure produced signs of toxicity and
death in 6 of 10 of the experimental mice. All 10 mice that received
multiple exposures to epichlorohydrin died. The signs of toxicity were
similar to those already described for inhalation or oral exposure. The
animals showed decreased activity, increasingly depressed respiration, and
loss of righting reflex. Examination of the dead animals showed congestion
and edema of the internal organs, hemorrhages in the brain, and necrosis of
the renal tubules.
Pallade et al. (1967) examined the percutaneous absorption of epi-
chlorohydrin in mice. The tails of 10 mice (strain, sex, age, and weight
unspecified) were immersed in epichlorohydrin for 15-20 minutes. Seven
mice died within 24 hours following exposure. The effects described were
similar to those observed by Kremneva and Tolgskaya (1961).
Smyth and Carpenter (1948) reported skin irritation following the
application of 0.01 ml of epichlorohydrin to the clipped abdominal skin of
five albino rabbits (strain, sex, age, and weight unspecified). The authors
described the irritation as a slight increase in local capillary perme-
ability.
In a study by Kremneva and Tolgskaya (1961), a small glass cap was
affixed to a rabbit's back (experimental details not provided). The glass
cap contained 0.5-1.0 mg epichlorohydrin. After 1 hour of exposure, the
glass cap was removed and the site was washed with soap and water. The
6-14
-------
skin appeared hyperemic; then ulcer!ike lesions developed followed by
scabbing. Complete recovery took 1-1.5 months.
Lawrence et al. (1972) examined the dermal irritation that occurred
when epichlorohydrin was applied to the shaved backs of male albino New
Zealand rabbits. A "Webril" patch (1.27 cm square) was wetted with 0.2 ml
of undiluted epichlorohydrin and placed on the shaved backs of rabbits and
covered with an occlusive bandage for 24 hours. An 8 percent aqueous
solution of sodium lauryl sulfate was used as a positive control and cot-
tonseed oil was used as the negative control. After the patch was removed,
the irritancy was evaluated on a 0 to 3+ scale. Epichlorohydrin showed
considerable irritant activity. Undiluted epichlorohydrin produced irri-
tation equal to, or greater than, the positive control (3+). The same
procedure was then used with epichlorohydrin diluted with cottonseed oil;
0.2 ml volumes of the various dilutions were tested. Table 6-4 shows the
results of the tests.
Lawrence et al. (1972), in determining the acute percutaneous LD50 for
epichlorohydrin in rabbits, used the same procedure as that for the irritancy
testing. Measured amounts of epichlorohydrin were placed on the "Webril"
patch, and it was covered with an occlusive bandage for 24 hours and then
Table 6-4. Dermal Irritation Scores for Solutions of
Epichlorohydrin in Cottonseed Oil
% Epichlorohydrin (v/v)
0.3125
0.625
1.25
2.5
5.0
Response
0
±
1+
2+
3+
Source: Lawrence et al. (1972).
removed. The mortality observations were recorded for 6 days. The LD50
value in male New Zealand albino rabbits was 0.64 ml/kg (755 mg/kg), with
95 percent confidence limits of 0.33-1.22 ml/kg (384-1,445 mg/kg).
6-15
-------
Lawrence et al. (1972, 1974) examined the sensitization potential of
epichlorohydrin using the guinea pig maximization test. Five male Hartley
albino guinea pigs weighing 300-500 g received intradermal injections of
0.01 percent epichlorohydrin in cottonseed oil and complete Freund's adjuvant.
Seven days after the first injection, epichlorohydrin was applied topically
over the injection site, and the site was then covered with an occlusive
bandage for 48 hours. Two weeks later, the hair was shaved from a different
site (hind flank) and the epichlorohydrin was applied topically and covered.
with an occlusive bandage for 24 hours. The bandage was then removed, and
the site was cleansed with alcohol. Twenty-four hours later, the site was
evaluated for sensitization reactions. No evidence of sensitization was
observed in any of the five treated guinea pigs. The positive control, 25
percent 2,4-dinitrochlorobenzene, produced a response of 3 grade (intense
redness and swelling).
Weil et al. (1963) also examined the sensitization potential of epi-
chlorohydrin in guinea pigs. Eighteen guinea pigs (strain, sex, age, and
weight unspecified) were injected intradermally with 0.1 ml of diluted
epichlorohydrin (concentration unspecified) three times a week on alternate
days for a total of eight injections. After a 3-week period with no expo-
sure, a challenge dose was injected and the animals were examined 24 and 48
hours thereafter for sensitization reactions. The concentration of epi-
chlorohydrin in the challenge dose was unspecified. Sensitization reac-
tions were not observed in any of the treated guinea pigs.
In contrast to the above test results, when Thorgeirsson and Fregert
(1977) examined the sensitization potential of epichlorohydrin using the
guinea pig maximization test, positive results were observed in more than
half of the animals. Fifteen female Hartley guinea pigs weighing 300-400 g
were injected intradermally with 0.1 ml of equal portions of 5 percent
epichlorohydrin (w/v) in ethanol and complete Freund's adjuvant. The same
procedure was followed as described in the study by Lawrence et al. (1974).
After 1 week, the occluded patch was wetted with 2 percent epichlorohydrin
in ethanol and applied to the skin over the injection site for 48 hours.
After 2 weeks, an occluded patch wetted with 1 percent epichlorohydrin
solution was applied to the shaved skin at another site on the body. This
sensitization test gave positive results in 9 of the 15 animals tested.
The authors classified epichlorohydrin as a grade 3 or moderate sensitizer.
6-16
-------
It is not clear from these studies whether epichlorohydrin causes skin
sensitization reactions. Further animal studies are necessary before the
sensitization potential of epichlorohydrin can be reliably determined.
However, there are reports (Jirasck and Kalinsky 1960; Lambert et al. 1978)
of skin sensitization reactions in humans occupationally exposed to epi-
chlorohydrin (see Section 5.4.2).
6.1.2 Subchronic and Chronic Toxicity
There are few subchronic and chronic epichlorohydrin toxicity studies
in the published literature. A recently completed 90-day study (Quast et
al. 1979a) has been designed and conducted in a much more thorough manner
than previous subchronic sjtudies. A summary of the subchronic toxicity
data appears in Table 6-5. The only chronic study found in the literature
was published by Laskin et al. (1980). This study was relatively complete
and well designed; however, all groups of animals, including controls, had
a high incidence of respiratory tract infections and pneumonia and, also,
poor survival rates.
6.1.2.1 Inhalation—Gage (1959) exposed five groups of eight albino Wistar
rats each (four males and four females) to epichlorohydrin vapor at concen-
trations of 9, 17, 27, 56, and 120 ppm (34, 65, 103, 213 and 456 mg/m3)
daily for 6-hour periods, 5 days/week, for a total of 11 to 19 exposures. The
test animals weighed between 160 and 200 g. No control animals were reported
in this exposure study. The epichlorohydrin vapor was prepared by atomizing
solutions of epichlorohydrin and propanol in a metered stream of air. The
concentration of epichlorohydrin was calculated based on the amount of
solution delivered by the feed syringe, the concentration of the epichlor-
ohydrin in the propanol solution, and the rate of airflow to the atomizer.
Daily checks were made of the chamber's atmospheric concentrations of
epichlorohydrin using a colorimetric method.
Rats exposed to epichlorohydrin at 120 ppm (456 mg/m3) showed labored
breathing after the first 3 hours, which continued throughout the remaining
exposures. Between exposures, the animals were lethargic and their condi-
tion progressively deteriorated during the study. Considerable loss of
weight, nasal discharge, and marked leukocytosis were observed. One rat
died after 11 exposures, at which time the study was terminated. At termi-
nation, the author reported that "the urinary protein was more than double
6-17
-------
the normal value", indicating possible kidney damage. At necropsy, the
kidney cortex was pale in color. Microscopic examination revealed areas of
leukocytic infiltration and atrophy of the peripheral cortical tubules in
four of the eight animals examined. The lungs showed congestion, edema,
consolidation, and inflamed areas with signs of abscess formation. Micro-
scopic examination of the liver revealed generalized congestion with one
animal's liver showing areas of necrosis.
Rats exposed to epichlorohydrin at 56 ppm (213 mg/m3) were lethargic
after the 10th exposure and later in the study exhibited respiratory distress,
loss of weight, and nasal discharge. Limited recovery was evident follow-
ing the weekends. Urinary protein, hemoglobin levels in blood, and dif-
ferential cell counts were normal. Eighteen exposures were made at 56 ppm
(213 mg/m3) before the study was terminated. No abnormalities were observed
at necropsy, and no abnormal microscopic findings were reported except for
an abscess formation in one lung, which the author did not attribute to
epichlorohydrin exposure.
Eighteen exposures to 27 ppm (103 mg/m3) caused mild nasal irritation
and 19 exposures to 17 ppm (65 mg/m3) epichlorohydrin caused no adverse
effects. Two rats exposed 18 times to 9 ppm (34 mg/m3) developed pulmonary
infections; however, the remaining animals in this group were healthy and
had normal weight gains. In this study, no results were presented for
vehicle controls (propanol); therefore, it is difficult to attribute the
changes observed solely to epichlorohydrin exposure.
Gage (1959) similarly exposed two New Zealand white male rabbits
weighing 1.8-2 kg to 35 ppm (133 mg/m3) epichlorohydrin by daily inhalation
for 20 days. These animals showed signs of nasal irritation, normal weight,
and no abnormal gross or microscopic findings. Two rabbits exposed to 16
ppm (61 mg/m3) epichlorohydrin showed nasal irritation after two exposures.
The concentration was then decreased to 9 ppm (34 mg/m3), and exposure con-
tinued for 20 days. No effects were observed and gross and microscopic
examination of the tissues from these animals revealed no tissue changes
that could be attributed to the epichlorohydrin exposure.
Kremneva and Tolgskaya (1961) exposed two groups of eight rats (strain,
sex, and age unspecified) to epichlorohydrin by inhalation. The first
group of animals was exposed 3 hours/day for 5 months at a concentration of
6-18
-------
rv
^^
OF EPICHLOROH
co
0
LU
u_
u_
LU
u
1— 1
0
OH
^"'
0
CD
CO
i
y
n
is
CO
"
s
s-
£
«
cc
-P
u
CO
IO (/)
4-> C
U 0
CO •!-
<*- -p
<*- o
CO CO
IO i-H
c
o
-p
CO
CO
x:
c
i— i
-P
CO
Q£
&
O
t. *©
O -C
•E-S
co a.
ii-
C i—
•i- co
- s-
C 0
•i- C
co .a
C33 CO
-P 0
£Z C.
.O)
co cT
S 0
•r~
I— -p
CO •!-
E ~a
i. C
0 0
Z 0
^
-o co
ID 3
X 0
E Q.
a. -a x
Q-^s CO
tn ^
c
o
'-p
(O
p^
.c
c
i— i
CO
OH
CO
E 1
t- •!-
>> O i—
T3 C 0
o X) in
.a to c
o
•POO
c c
ra ai
•p - c
to c o
C •!-
0 CO S-
U D> O
*»-
--P
c .e 4J
o a Q.
•i— »i— CO
•P CO U
CO 3: X
•P CO
•i- O O)
i- C >, C
S- 55 3
• f»» *, ^^ |__
•P r—
r- -C O T3
to rax: co
tO •!-+-> 4.)
CO CO CO CO
z s a. -a
^
-o eo
ID 3
V)
X O
E a.
Q."O X
Q-^v CO
X^
!*•«• 00
CM 10 t-H
C
0
•p
CO
"cO
C
1— 1
CO
"O
CO W
•a H
co
co
»
> c?
co -^
C U)
E D)
CO i—
1- O
-O T3 CO 1 S-
CCOWC O C
fO-PCOO -
1
C
rH
co
c:
c=
o
CO
x:
CO
OH
6-19
-------
fl>
u
O)
at
Qd
U3
•a rH
Cv^
n>
to
3 .*:
c u>
e o
l£>
£
* s,
,
'3 I-H j= co in
u s- • in i— 3
.^ _<» j-4J(i)i— a.
.^5*3^ rd'l g wife 3-0 i
^£§'SISH-iE^-§iy
ujzEioioa.O'— _-*->^g-
-a
g.
co
tn
t
«o
rl
mm
Islls*
!a -c .inoinl--ij> =
•Qu)u).^u)t--"-^ua)ca)E
T3-PO) 0)0)5-4-'-i-i— PC3
-pa) EO-U"—
S«??S&.51» > --xuag
a)
10
3 «
o
-------
/— -N
73
cu
3
C
•t—
•P
C
o
u
•
in
i
ID
LLJ
1
CO
l~~
cu
u
c
cu
cu
cu
^_>
o
M-
• UJ
cu
in
o
0
cu
•p
3
o
OS
cu
u
cu
CL
CO
re
CD
CD
r "i
•
jm.
re
-p
cu
-p
in
re
cy
in
-P
u
cu
<*-
tf_
cu
o
z:
X
CM
73 ID
^ 1
-C r-H 10
VD CU
10 S-
E 3: in
0-^v. 0
Q.73 Q.
X
in in cu
c
0
•p—
•p
re
f«
re
JZ
c
1 — 1
-p
re
Q£
5-
10 >> o
CU S- E
D) O 3
<— C -P -P
re i re u
o E -e re >>
o re u f- cu .
i— i— i— C C > • •!- 3
>,!_) .r- 73 _>»i m
4-> -P C O
•r- re re cu o.
> . s- ex
•r- C Ol >>O O)
-P 0 C S-
U T- CU O O
re -P OJ-P . -P
re cu re E
73 -P 73 S- 3 73
CU •<- -i- T- CU
in s~ •> Q-i — -t->
re s- >> in cu re
O) -I- S- CU -C .—
S- 0 S- +-> CU
U CU -P T- S-
o> >> re c Q. c
Q 3? E •- CU 3
X
73 CM
-C I
i-i in
ID t^ CU
S-
E ^: 3
O. S m
Q."^ O
73 a.
in x
CM in
re
-P
cu
-p
in
re
^
« in cu
1 C 3 OV
73 •!-• 1 O O C .
cu • i s- cu T- E re
-PO S-73-Pre -C-P
•arec-p T- re3 oo.
>, c CU CU >> •> S- CT73 CU
cures--poii- >>cucuincs-u
c c uc o s_^:c recux
73 >>•>- 73lOCUOt- oreCU- JC. Q)
i— cu s- >, •!-•!- ouore -p-PO)re>>-p
reinoins- Jsiin 4-> rUCCU-r- S- o «
ure-pcuoi— o cere s--r-aino cu
ocuroo)-pre73s-.cuor— -p- re-p s-
i— s-Ecuincuux D)-C- 3 Q.73 i— re .3
'uErereremcucuo-Pu mo • a. s- e tn
** cu re ^* i^— e re e i ^ tj re *r— cu o \ ^ s— *r— 3 o
>,O i— Or- CU J->,NU S-i)-J=CU O-'<- O-
-P <4- ocus-s-or— -r-m • cn o_ in •— x
•r- c cu x: u re u o>r— re ->73-i->>cucucu
> . 1— 1 > 73 -P C r— OM- . CUCUCU.CS-.C
•r-C -r-C t-H 3 r— S- 3 X OJIOS 4->E
•PO -pre1*- xjrereurecu s-re •><«- -r- o.
o-i-.re o 3c>— rec-P recu>>coo-CL.
re4->CS->> •(— CU3>OS- XZS-73O CU CU
rO'i-cus-Ein s-r— NO uoo-r-re os-
734->recO3CU r— 73 U WCUXJ-P-r- >,m CU<
CUT-O)CU-P'i--P.l)-CUCUC
ins- o>re< — remouin-r-i — 73 73Ereoincu
re£--pcus-cuc-p ore re c at e •— ; 4-> re m
— rerecucu
cj cu T™ 73 m *n" S— •!— cu o o re S— in «P cu M— • i * M— • E £••
cu>>cuccuo-3cuT3cuer— 73 reres-ecur— reo
ei cu 3s re S- cu -P 3c cu -c i — i o. re 21 -P u *r— E o m E
x in
X 1 >>
73 o re
73 CM ~-^ Q.73
-C 1 CU CM
rH in . ID i—l
tO UD CU
-------
CJ
CO
$-
CO
CO
cn
u
0)
•a t~
03 LU
3
c
4J
c
.§
in
LU
1
CO
in
o
o
CO
o
ce.
in
CO
*u
CO
a.
1/5
CO
ca
CO
_j
1
CO
t-t-H
0) .3-
TJ cn
3 r-l
Q) ^^
S-
LU
in
CO
s«
3
U)
0
CL
X •
cu cn
CO
ID S-
t-l 3
in
t- 0
M— CO
CO
"Si
•r- 01
in ca
ll
ca
r^"™
•arx:
i — in
S+j'ca
O TJ CO
P"^ **•*• "CJ
CO J=t— CO
•t r— CO
CM i-H COT3
C
O
'•Jj
CO
CO
c
h-l
CO
in
3
S
cn cn co
r- r-. .1^
cn cn co
rH t— 1 a)
_i
1-^ r-^ ,—N TJ
co ca cn c
m ca
•*•> +J cn <—>
CO CO r-t i- rH
v^/ co «=h
4J +J -Sen
in w co 3rH
co ca a> co>— '
3 3 CO S-
0" o- o u_
aj +J
N X:
• TJ W C «r- O3
OWCO COTJ COW-i- S~
•z. c. v> oio air— co
c: COCOT; c o. ococo 4Jj-
• •i-co 05 coci— ca t-+Jco o+J >>i— M- • t- 5- CO CO .C Jz > • >> >> TJ CO r— (O 4J CO W CO W
O)«(— WO O W CO U TJ OVr— W i — O . O E r— TJO TJO
j=i.coco ••- oca caSS-cat-cO4J .•— c r-co>, TJ TJ
O) J^ +J
•r-CT-+J .COT-J2S- S~>>CT-r—COC ECO +JTJ.r- WOO W«d-
cocojao CE 30 uTJcoja3^:o s- c r- s- t-
3 D) I- T-EWWC COOO>S-r— E OO 34JCO -r-S- -r- S-
C03C COCOCO ^3 TJ .O. CO =J r— TJ 3 CC W+J M-O) t-CO
CO >)— C CO «TJ OWCoi -- i— -r-o S-M- 1.1-
WTJr— W +JCCOUO- CCOTJr— OCO0.3 CW -r-U-E airfl COCO
caccaco ^:i-oo J=QO 4^5- cocOrH j=i — x:.—
U) TJ'r- C 0 ^-r- W T- O T- CO -r- C CO, 1- -r- O . S- i- 4J rO +J CO
+J 4J-r- TJcO+JCO+J S-TJ-tJ'i-TJ J3O S_C>> COTJ . CO4J CO+J
u >>co r— co-Pro.tJ-.-s- s- co ai-3 S-T-CJI+J .w > •*— o w w tn co TJ o TJ o
<»- j= ro CD s- co^cros+J r— ^cracncocor— o. r-xr'o E'tOwo •)-> -t->
CO CDr— CO S-TJCOXJ3O COCOi — CS--PCOO CO O)JC -r-EOTJ W>« WS«
•r-t-COC: U O O) t- ID W+JI-COUCOS- W-1--P E-r-TJ S-O S-O
o i — CX:JQ cooai3c:fz coccjztuoo+J rococo TIC t-< -r-o -r-o
Z I/5WOCO O1-TJ+-> O--r- Z'r-HHOQOU-rO ZSCX CCCOCMCM Ll-rH U-rH
X W
=< 1 >>
X TJ O CO E
CM TJCM TJ CM -x, Q.TJ Q.O
TJU5 -^l£) \U3 XIX O-CM
~»- i x: i x: i co CM
X:rHinr-)W t—IW IDrH OS- TOO!
^IOCOIDIDCO IDICICO gCn^ cr> ^ ^> ^ ^
BXwo-Sin a.Sw o-S1" icT-a CD i? 1°
O.^ O Q_^^ O Q.^ — , O *\ CO r-t — ^ W E
Q.TJ D-TJCL TJO- OTJS- JC >, OrH
xmxox 03 •. t& *$• cnP
iDLOcocMLnco LOIOCO r-HLnw cni^TJcn I-H CM
c c c c c
o o o o o
+3 \ ^ ^3 'r* *rr
CO CO CO CO CO
CO CO CO CO CO r— i — r—
x:x:.c x: x:caroro
ccc c cs-s-s-
HH . M 1— I 1— 1 HQ OO
CO CO CO CO •>-
w w in in .Q
33, 3 3 3+J+J+J
ooo o cocococa
SEE: s: Q^afc^of
6-22
-------
-a
aj
3
C
•r—
-P
C
O
u
*SM^
VD
LU
i
i
CO
ai
Ifl "l — tO
re »> to re cm
CU to S— -C O CU
s_ o cu -P T- 3
cj *f™* -p «c to to
C -P OJ Q.T3 OJ to
•r- re E O >
c -P re s- aj s- s-
•r- j= o-.a s- re cu
re Di o o c x:
OL> •!- r— HH E O
-P 3S O) 3 r-
-c o to 3
05 >>•— . Q. .
•i- TJ 0 E O) (/J
,i— i— a. O
Q^CCT3'i-OC
O1T3 S-
C i— 10
re E c >,
O i—
tn I-H -p re >,
O i— 1 U "O
O O i—
• C •!— O
C3 C3 -i- O CO
're
a>
c
o
•p
*jl
OJ
a.
re
&-
-p
c
i— i
•p
re
Of
i T3
OJ
1 OJ
OJ S-
•a u
QJ
""O
C
o *
•r- C
-P T-
OJ 1
5^ OJ
0 C
o
^ "O
-C OJ
^_> 40
2>» C
S-/ OJ
01 E
Q. re on
S OJT3 OJ
to -P
o o-
O •!-
OJ
o
**-• ^3
O
^j <^
cu
to T3
Decrea
crease
o>-o
r — tO
E C
^j
"st" 1 ^
f-- -P
^j- (_,
0
. c
0 •!-
^-«
re
OJ
c
o
-+J
"£
OJ
Q.
re
S-
-p
c
i— i
•p
re
C£
c to
re
o
JC U
^
*
"O ^^
s
CSI
r— I-l
•1 —
0 X
'
to i
OJ J—
•P CD
>5>
U-t-
O r—
^"
Q.-O
£ C
5*"> fO
f^
*^>
T3 OJ
OJ C
U) "O
re «i-
u «
OJ -P
T3 S-
re
"O QJ
C *""
re
to oj
i — t/>
Q.S-
S- C
•P 1-H
^~
re
i-
o
c
o
re
O
^y
.
CD
re
•p
01
•^
OJ
>>
J2
0
•P
.
^Q
^^
-JL>
s-
o
o.
cu
&
0
Q
^
-p
o.
01
«r™
-p
,
OJ
T3
« C
c re
o
•P OJ
Q. 3
3 "re
o c
U •!-
J3
•o o
O i —
O O)
"4- O
OJ ^
to
Decrea
crease
-o
\ s>
r— C
^z ^3
1 *
0 -P
S S
0
. c
"re
OJ
c
o
-p
"si
OJ
a.
re
S-
-p
c
h- 1
4J
re
o±
3
"O IO
cu to
re'-p
OJ
S- i—
oj E
Q S-
o
c
•P OJ
C O -P
Z5 Z. S-
O 0
u o.
cu to s_
-P OJ
U ^% O)
QUO
S- 0 r—
x: x: o
-p a~c
>>S-P
s- >> re
OJ r- 0.
^
25
*s^
^^
2«
r—
•r- .
O IO
PS CJ)
.^^ cn
^f* (p-
-a
QJ *r™
V) '|T J
re
OJ S-
5- OJ
OJ -P
Q 0
ra'Sx
\ w m
r— C 3C
^5 CD "^N^
. ^ "O
LO -P
O O
. c •«—
0 -i- O
re
OJ
o
-P
•r~
S-
OJ
Q.
re
s-
H->
C
1— 1
i %
re
CU
3:
CM
i-l
6-23
-------
0.17 to 0.25 mg/1 (170 to 250 mg/m3). During the first month of exposure,
these animals were not observably different from controls. In the follow-
ing month, there was some deterioration in the condition of the animals.
Hyperemia of the skin and subcutaneous hemorrhage were noted on some sites
of the body, respiratory difficulty and dyspnea were apparent, and two
animals died. There was some improvement after 2 months of exposure;
however, in the fourth month there was a marked deterioration in the condition
of the animals, and all remaining animals died at the start of the fifth
month. The rats showed elevated blood pressure after 1-2 months of exposure.
Microscopic examination of the tissues from these animals revealed changes
in the lungs, kidneys, liver, and heart. Respiratory tract changes included
bronchitis, with necrosis of the bronchial mucosa, thickened alveolar
septa, and pulmonary edema. The kidneys showed deterioration and necrosis
of the convoluted tubules. The hepatic cells showed some fatty degeneration
and vacuolization. The staining of myocardial tissue was irregular, and
there was fragmentation of the myocardial fibers.
The second group of rats was exposed to epichlorohydrin at a concen-
tration of 0.02-0.06 mg/1 (20-60 mg/m3) for 3 hours a day for 6.5 months.
No signs of toxicity were observed, and no deaths occurred during the
study. The weight gain was 5-10 percent below that of the control animals.
Two months after exposure began, blood pressures were slightly elevated
(95-100 mmHg compared with 90-95 mmHg for the control animals). The changes
observed in the tissues were similar in nature but not as severe as those
observed at the higher exposure level.
Fomin (1966) exposed three groups of 15 male white rats (strain and
weight unspecified) for 98 days (14 weeks) to epichlorohydrin at concen-
trations of 0.2, 2.0, and 20 mg/m3 (0.05, 1.06, and 5.28 ppm). A fourth
group not exposed to epichlorohydrin served as controls. The author examined
body weights, blood nucleic acids, erythrocytes, leukocytes, hemoglobin,
and urinary coproporphyrin. In addition, leukocytes were examined micro-
scopically for their ability to fix a fluorescent dye (dye unspecified).
Neurological/behavioral measurements were also made and are described in
Section 6.3.
6-24
-------
Nucleic acids in the blood decreased in concentration in the high
exposure group beginning the second month of treatment and in the middle
exposure group beginning the third month of treatment. The low exposure
group and the control group showed no significant changes in nucleic acid
levels during the study. After a recovery period of 4 weeks, the nucleic
acid levels in the two higher exposure groups returned to normal. None of
the rats exposed to epichlorohydrin had significant changes in erythrocyte,
leukocyte, or hemoglobin levels. There was a dose-related increase in the
number of leukocytes fixing a fluorescent dye, however, it was not made
clear by the authors whether this could be interpreted as a meaningful toxic
response. The necropsy microscopic examination of the tissues from the
animals at the highest exposure level (20 mg/m3) revealed emphysema, des-
quamative interstitial pneumonia, areas of edema, and deterioration of the
connective tissue surrounding the blood vessels in the lung. There were
intermuscular, microfocal hemorrhages and venous plethora (red florid
complexion) in the heart, and necrotic changes in the convoluted tubules of
the kidneys. Damage to the neurons in the medulla oblongata, cerebellum,
and the hippocampus was also reported; however, these changes were not
described in detail. Animals exposed to the lower levels had normal path-
ology.
Freuder and Leake (1941) exposed 10 white mice (strain, sex, and age
unspecified) to epichlorohydrin at a concentration of 0.1 mM/1 (2,370 ppm)
daily for 1 hour. Exposures were continued until all animals were dead.
Mortality was recorded daily. Table 6-6 shows the results of the study.
Table 6-6. Mortality in Mice Exposed to 2,500 ppm Epichlorohydrin
No. of Exposures
1-2
3-5
6
7
8
9-15
16
No. of Survivors
10/10
8/10
6/10
5/10
3/10
2/10
0/10
Source: Freuder and Leake (1941).
6-25
-------
During the first two exposures, the only abnormal signs were irritation of
the nose and eyes. The animals then showed decreased activity and muscular
relaxation of the extremities, and respiration was depressed and increas-
ingly difficult. Some animals experienced clonic convulsions.
The subchronic inhalation toxicity of epichlorohydrin was examined in
two strains of rats (Fischer 344 and Sprague-Dawley) and one strain of
mouse (B6C3F1) following repeated daily exposures (Quast et al. 1979a).
Inhalation exposures were at 0, 5, 25, and 50 ppm (0, 19, 95, and 190
mg/m3) of 99.8 percent epichlorohydrin for 6 hours/day, 5 days/week for a
total of 61 or 62 exposures in 87 or 88 days for male and female animals,
respectively. For each species and strain, 20 males and 20 females were
used in each group; rats were 9-11 weeks old and mice were 7-9 weeks old at
the start of inhalation exposure. An interim sacrifice of 10 animals of
each sex per exposure group was made for each species and strain after 30
days of exposure, and histopathologic examinations were conducted on five
animals of each sex of both the control and 50-ppm groups. After 90 days
all surviving animals were killed. Clinical studies conducted on animals
killed at 30 and 90 days included urinalysis (rats only), hematology, blood
urea nitrogen, serum glucose concentrations, and serum enzyme activities
glutamic-pyruvic transaminase (SGPT), glutamic-oxaloacetic transaminase (SCOT),
and alkaline phosphatase (AP). Control animals and the 50-ppm exposed group were
necropsied and the following organs were weighed and prepared for histo-
pathologic examination: brain, heart, liver, kidneys, testes, spleen, and
thymus. In addition, all possible target organs from the 5- and 25-ppm
exposure group animals were microscopically examined at the 90-day sacrifice.
Inhalation of 5 ppm of epichlorohydrin (Quast et al. 1979a) did not
result in toxicologically significant effects in rats or mice as determined
by clinical observations or changes in body weight, hematology, urinalysis,
clinical chemistry, organ weights, gross pathology, or histopathology.
During exposure, rats showed a dose-related conjunctival redness and
eyelid spasms without evidence of ocular involvement. These effects appeared
to be transient with recovery occurring overnight. Comparable observations
were not made in mice simultaneously exposed with these rats. During the
first 10 days of exposure, reduced activity was noted in the rats exposed
to epichlorohydrin at 25 and 50 ppm.
6-26
-------
There were no significant alterations in hematology, urinalysis or
clinical chemistry parameters in any of the test animals. There was a
slight decrease in body weight gain in male rats of both strains and in
male and female mice at 50 ppm, and in female Fischer 344 rats at 25 ppm.
The most severely affected tissues in both rats and mice were the
nasal cavities. There were inflammatory and degenerative changes in the
olfactory and respiratory epithelia of animals exposed to 25 or 50 ppm of
epichlorohydrin. The severity of lesions was dose related. Male rats
were more severely affected than females, and histologic changes were more
severe in Sprague-Dawley rats than in Fischer 344 rats. Mice were less
severely affected than either strain of rat, and there was no apparent
difference in severity of lesions between male and female mice.
Histopathologic changes were observed in the kidneys of rats of both
strains exposed to 50 ppm epichlorohydrin, consisting of increased inci-
dence of dilated tubules, focal tubular nephrosis and swelling of epithelial
cells of the renal cortex. The severity of lesions of the kidney did not
differ at 30-day or 90-day sacrifice, suggesting a lack of progression of
the effect on repeated exposure. At 25 ppm, there were no histopathologic
changes in rat kidneys; one female Sprague-Dawley rat had a unilateral
kidney tumor that was not considered exposure related. There were no
histopathologic changes in the kidneys of miqe exposed to epichlorohydrin.
The livers of rats exposed to 50 ppm epichlorohydrin showed decreased
glycogen deposits but no other histopathologic changes. A similar effect
was not seen in mice.
At final sacrifice the adrenal glands of some male rats exposed to 50
ppm epichlorohydrin showed slight microvacuolation of cells in the zona
fasciculata; this was possibly a stress response. In addition, in the
epididymis of several male Sprague-Dawley rats from the 50-ppm group, there
were increased numbers of nucleated cells and/or amorphous eosinophilic
staining material within the lumen, although there was a normal sperm
count.
In summary, the results of this study indicated that both rats and
mice exposed to 25 or 50 ppm epichlorohydrin consistently had substantial
changes in the epithelium of the nasal turbinates. Lesser effects in other
6-27
-------
tissues also occurred at these exposure levels, but with some variation in
response. Rats and mice exposed to 5 ppm epichlorohydrin had no adverse
effects in any of the parameters monitored in these studies.
To characterize target organ effects and to evaluate early changes in
the nasal turbinates, Quast et al. (1979b), in a subsequent experiment,
exposed Fischer 344 rats, Sprague-Dawley rats, and B6C3F1 mice to 100 ppm
(380 mg/m3) epichlorohydrin for 6 hours/day for 5 days during 1 week, and
for 4 days during a 2nd week for a total of nine exposures in 12 days.
Exposures were carried out in a 4.3-m3 stainless steel and glass Rochester-
type dynamic flow inhalation chamber. Five animals of either sex were used
as test animals and controls; rats were 9 to 12 weeks old and mice were 7
weeks old at the start of the exposure.
When groups of animals were placed in the exposure chamber they huddled
together and slept. No evidence of eye or nasal irritation was detected
during exposure; however, upon removing the rats from the chamber following
exposure, there was a slight amount of moist nasal discharge and discolora-
tion of the hair immediately around the nasal orifice, suggestive of exuda-
tive rhinitis. This was not noticeable in the mice due to the dark hair
and skin color of the species used. Immediately after the animals were
removed from the exposure chamber they sneezed and rubbed their noses.
Signs of respiratory distress, apparent decreased food intake (actual food
consumption not measured), and reduced fecal excretion were observed during
exposure periods with some recovery observed on the weekends.
A marked decrease in the body weight of rats and mice was observed
during the exposure to 100 ppm (380 mg/m3) epichlorohydrin and equally
apparent was a transient partial recovery of body weight following the
weekend.
After nine exposures (day 12), all animals were killed. Weights of
brain, heart, liver, kidneys, and spleen were recorded. Samples of blood
and urine were collected on day 11 for hematology and urinalysis evalua-
tions. On day 12, at time of necropsy, blood was obtained from rats to
determine the serum concentrations of urea nitrogen, glucose, and the glu-
tamic pyruvic transaminase (SGPT), glutamic-oxaloacetic transaminase (SCOT),
and alkaline phosphatase (AP) activities.
6-28
-------
Multiple effects were associated with repeated exposures to 100 ppm
epichlorohydrin. The most consistently recognized treatment-related effects
were changes in the mucosa of the nasal turbinates, decreased body weight
gain, leukocytosis secondary to nasal inflammation, decreased specific
gravity of urine (hematologic and urinary examination were conducted only
on rats), and increased kidney weights in rats, but not mice.
Upon histopathologic examination of tissues, the most consistent and
readily detectable changes were present in the nasal turbinates, with
degeneration, inflammation, hyperplasia, and squamous metaplasia present to
some degree in all exposed rats and mice. This condition extended through-
out the regions lined by respiratory and olfactory epithelium. These
changes were more severe than those noted in the rats and mice exposed to
25 and 50 ppm epichlorohydrin in the 90-day subchronic study by Quast et
al. (1979a). Changes in the respiratory tract of mice were much less
severe than those in either species of rat but were similar in nature.
Occasional sections of rat trachea had an increased number of inflammatory
cells migrating through the epithelial lining. Dose-related changes in the
liver were minimal for both animals. These changes were characterized by
decreased hepatocellular glycogen content, decreased hepatocyte size, and
increased variability in cytoplasmic staining.
Degenerative changes were noted in the kidneys of both strains of
rats, but. not the mice. Minor nondegenerative liver effects and thymic
atrophy, both secondary to stress, were noted in rats and mice. Male rats
of both strains had slight changes in the contents of the epididymides.
Male Sprague-Dawley rats had minor changes in the adrenal glands, possibly
secondary to stress. In general, there was a decreasing order of toxicity
observed as follows: Sprague-Dawley rats, Fischer 344 rats, and B6C3F1
mice.
Laskin et al. (1980) examined the chronic toxicity of epichlorohydrin
in rats. Two groups of 100 male Sprague-Dawley rats were exposed by inha-
lation to epichlorohydrin concentrations of 10 and 30 ppm (37.8 and 114
mg/m3). The animals were exposed 6 hours/day, 5 days/week over their
lifetimes. An additional group of 140 rats was exposed to 100 ppm epichlor-
ohydrin for thirty 6-hour exposures and then observed over their lifetimes.
A sham control group of 100 rats was exposed to air alone in an exposure
6-29
-------
chamber using the lifetime exposure schedule. Finally, a group of 50 rats
was maintained as untreated controls. The laboratory methods for inhalation
exposure, necropsy, preparation of tissues, and histopathologic observations
are reported in Section 7.1 (Carcinogenicity).
The weight gain for the,animals exposed to 10 ppm epichlorohydrin was
comparable to that in controls, but the group exposed to 30 ppm began to
show marked decreases in body weight after 40 weeks. For the first 16
weeks of the study, there was no significant mortality in either exposure
group. However, by 48 weeks, 45 percent of the group exposed to 10 ppm had
died and by 60 weeks a similar number of the group exposed to 30 ppm had
died. In all cases, pulmonary congestion and pneumonia were observed.
Mortality was not treatment related; in fact, a slightly greater mortality
rate was noted in control groups of rats than in treated groups.
At necropsy, renal damage was observed at a high incidence in epi-
chlorohydrin-treated animals. The severities of the lesions were similar
for the animals treated at 30 and 10 ppm. The incidence of kidney lesions
was 65, 37, 24, and 14 percent for the 30-ppm, 10-ppm, sham controls, and
untreated controls, respectively. Tubular degenerative changes were the
most common lesions. The tubules were atrophied, dilated, and some were
filled with hyaline casts. Occasionally, atrophy of the glomeruli was also
observed.
The authors observed a high incidence of rhinitis and pulmonary infec-
tion in both control groups. Approximately 90 percent of the control
animals showed severe inflammatory changes in the nasal cavity. For this
reason, the authors could not attribute effects observed in the nasal
cavity of exposed animals to epichlorohydrin exposure. However, none of
the control animals showed squamous metaplasia of the nasal mucosa.
6.1.2.2 Oral—Freuder and Leake (1941) examined the oral toxicity in mice
administered repeated doses of epichlorohydrin by gavage. Three groups of
15 white mice (strain, sex, and weight unspecified) received 10 mg/kg
epichlorohydrin suspended in a 25 percent aqueous gum arabic. Single daily
doses were given until all animals died. The number of doses required to
produce the first death and 100 percent mortality are shown in Table 6-7.
6-30
-------
Table 6-7. Mortality of Mice Administered Epichlorohydrin Orally
Dose
(mg/kg)
94
190
271
First Death
(day)
2
1
1
100% Mortality
(day)
21
8
4
Source: Freuder and Leake (1941).
Repeated oral administration produced many of the same signs described for
inhalation exposure: decreased activity, muscular relaxation of the extrem-
ities, stiffening of the tail, fine tremors, depressed respiration, and
clonic convulsions.
6.1.2.3 Intraperitoneal Injection—Lawrence et al. (1972, 1974) examined
the cumulative toxicity of epichlorohydrin by repeated intraperitoneal
injection in rats. Two groups of 12 male Sprague-Dawley rats, weighing
100-150 g received 30 daily injections of 0.00955 and 0.01910 ml/kg epi-
chlorohydrin dissolved in cottonseed oil. A control group received injec-
tions of cottonseed oil alone for the 30-day period. The rats were weighed
at 5-day intervals throughout the study. Clinical blood chemistry, rate of
clearance of sodium sulfobromophthalein from plasma, organ-to-body weight
ratios, and organ pathology were examined at the end of the 30-day exposure
period. No deaths occurred during the treatment period. Weight gain was
significantly less (p ^0.05) in the low dose group than in the control
group beginning on day 20 and in the high dose group on day 15. Impaired
hepatic function was not detected by the rate of clearance of sodium sulfo-
bromophthalein. There were no marked differences in the hematologic data
between dosed and control animals. However, kidney-to-body weight ratios
were significantly increased (p ^0.05) in both epichlorohydrin-dosed groups.
Microscopic examination of the tissues showed a greater incidence and
severity of lesions in the lungs of the dosed animals than in control
animals; however, the control group as well as the test groups showed
6-31
-------
pulmonary changes such as bronchitis, peribronchitis, interstitial pneumonia,
bronchopneumonia, and emphysema. Microscopic examination of the other
organs did not reveal any significant changes.
Lawrence et al. (1972, 1974) also examined the cumulative toxicity of
epichlorohydrin by repeated intraperitoneal injection in rats, but used a
different dosing schedule. In this study, three groups of 12 immature,
male, Sprague-Dawley rats weighing 60 to 100 g received repeated injections
of 0.0095, 0.0190, or 0.04774 ml/kg epichlorohydrin in cottonseed oil 3
days per week for 12 weeks. A fourth control group received injections of
cottonseed oil alone according to the same schedule. Food consumption
(weeks 1, 7, and 12) and weekly body weights were monitored throughout the
study and clinical blood chemistry, organ-to-body weight ratios, and organ
pathology were examined at the end of the 12th week of the study. Food
consumption was generally lower for the two high dose groups than for the
controls. Body weight gain was significantly lower (p ^0.05) for the high
dose group each week of the study, except weeks 2, 3, and 12. The hemato-
logical studies showed a dose-related decrease in hemoglobin, hematocrit
values, and erythrocyte counts in the epichlorohydrin-treated animals.
Hemoglobin concentration was significantly (p ^0.05) decreased at all three
dose levels and the hematocrit value was significantly (p = 0.05) decreased
only at the middle dose level. An increase in segmented neutrophils was
observed at the high dose level, and reductions in the percentage of lympho-
cytes were observed in the two higher dose groups. Organ-to-body weight
ratios were not significantly different in the low and middle dose groups;
however, in the high dose group, significant (p ^0.05) increases were
observed for the heart, kidneys, liver, and brain. No abnormal organ
pathology was reported.
6.1.2.4 Dermal—Freuder and Leake (1941) examined toxicity associated with
the repeated dermal exposure of rats to epichlorohydrin. Undiluted epi-
chlorohydrin was placed on a square centimeter piece of gauze and applied
to the shaved skin on the abdomen of 10 rats (strain, sex, and age unspeci-
fied). The gauze was removed after 1 hour; the application was repeated
daily. Two dose levels were examined, 6.5 mmol/kg and 13.0 mmol/kg (0.5
ml/kg and 1.0 ml/kg). Mortality data are shown in Table 6-8. Repeated
6-32
-------
applications caused superficial necroses of the skin that exceeded the size
of the gauze. The skin was parched and brown. In the rats that survived,
the skin gradually healed with no other symptoms of toxicity noted.
Table 6-8. Lethality Following Repeated Dermal Application
of Epichlorohydrin in Rats
No. of
Dose (ml /kg) Applications
0.5 1
2
3
4
1.0 1
2
3
No. of
Survivors
10/10
6/10
4/10
0/10
8/10
6/10
6/10
Source: Freuder and Leake (1941).
6.2 EFFECTS ON THE LIVER, KIDNEYS, AND LUNGS
Toxic effects in the kidneys and possibly in .the liver and lungs have
been reported following exposure to epichlorohydrin, and have been discussed
in Section 6.1; the following is a summary of these effects.
6.2.1 Liver
Inhalation exposure of rats (170-250 mg/m3, 3 hours/day for 5 months)
caused a moderate degree of fatty degeneration and vacuolization of hepatic
cells (Kremneva and Tolgskaya 1961). Exposure of rats to 190 mg/m3 epi-
chlorohydrin, 6 hours/day, 5 days/week for 61 exposures caused a decreased
hepatocellular glycogen content (Quast et al. 1979a).
Gage (1959) also found that nineteen 6-hour exposures to 120 ppm (456
mg/m3) epichlorohydrin caused congestion of the liver of rats. Lawrence
and Autian (1972) found that exposure of ICR mice to epichlorohydrin caused
a dose-related increase in phenobarbital-induced sleep time. This assay
indicates inhibition of liver microsomal enzymes. When microsomal enzymes
6-33
-------
are inhibited, sleep time is prolonged. Four groups of 10 mice each were
exposed to 98.20 mg/1 epichlorohydrin for 0.1, 0.2. and 0.5 of the LT5Q
(0.92, 1.83, and 4.58 minutes, respectively). A control group was placed
in the exposure chamber for 4.58 minutes and exposed to air. Increased
sleep times were observed for mice exposed to increasing concentrations of
epichlorohydrin. The sleep time for the control mice was 69.20 ±3.88
minutes. For those exposed for 0.92 minutes, the sleep time was 73.14 ±
6.46 minutes; for those exposed for 1.83 minutes, sleep time was 85.31 ±
6.40 minutes; and for those exposed for 4.58 minutes, sleep time was 108.93
± 9.34 minutes. This dose-related increase indicates inhibition of the
liver microsomal enzyme system.
6.2.2 Kidneys
Epichlorohydrin exposure has been shown by several investigations to
cause severe renal toxicity by different routes of exposure (Table 6-5).
The most extensive changes observed were in the convoluted tubules (Kremneva
and Tolgskaya 1961, Rotaru and Pallade 1966, Laskin et al. 1980). The
earliest changes were swollen, dilated, and ischemic convoluted tubules.
This was followed by epithelial degeneration. Later, the epithelia became
completely necrotic, and cells were desquamated into the lumen of the
tubules where they underwent calcification. After exposure ceased, regener-
ation of the tubular epithelium occurred. Rotaru and Pallade (1966) and
Pallade et al. (1967) found signs of regeneration in the kidneys in rats 5
days after subcutaneous injection of epichlorohydrin. Ten days after
exposure the investigators found marked regeneration of the tissue in the
kidneys with most of the tubular integrity restored. In addition to changes
in tubules, several authors (Kremneva and Tol(gskaya 1961; Rotaru and Pallade
1966; Gage 1959; Laskin et al. 1980) have reported minor glomerular changes
in rodent kidneys following epichlorohydrin exposure.
6.2.3 Lungs
Carpenter et al. (1949) reported acute respiratory irritation, hemorrhage,
and severe pulmonary edema in rats exposed to epichlorohydrin by inhalation
at concentrations ranging from 283 to 445 ppm (1,075 to 1,691 mg/m3) for 4
hours. There were also marked increases in the lung-to-body weight ratios
for animals exposed at higher concentrations. At 369 ppm (1,042 mg/ms)
6-34
-------
epichlorohydrin, an 80 percent increase in lung-to-body weight ratio was
observed when compared with controls. The two lowest exposure levels (283
and 303 ppm) showed no lung-to-body weight ratio increases over controls.
Quast et al. (1979a) found that inhalation of 50 ppm epichlorohydrin
(6 hours day, 5 days/week for 88 days) caused focal subacute pneumonitis in
mice, but not in rats.
Changes in the lungs have also been observed following epichlorohydrin
exposure by routes other than inhalation. Rotaru and Pallade (1966) and
Pallade et al. (1967) described the pulmonary changes in rats following a
single subcutaneous injection of either 150 or 180 mg/kg of epichlorohydrin.
There was inflammatory desquamative bronchitis, edema of the bronchio-vascular
connective tissue, and congestion of. the alveolar septa.
6.3 BEHAVIORAL TOXICITY AND CENTRAL NERVOUS SYSTEM EFFECTS
Depression of the central nervous system (CNS) has been linked with
acute exposure to high levels of epichlorohydrin. A range of 1,416 to 2,124
ppm epichlorohydrin was found to be lethal in rats during a 2-hour exposure
(Kremneva 1960). These animals first became quiescent and then developed
cyanosis and muscular relaxation of the extremities. This was followed by
tail stiffening and fine tremor of the body; the respiration became increas-
ingly depressed and some animals experienced clonic convulsions. Death occurred
from depression of the respiratory center. This follows the common clinical
development of toxicity from high acute exposures to epichlorohydrin.
Freuder and Leake (1941) exposed white mice (sex, age, and strain
unspecified) by inhalation to epichlorohydrin at concentrations of 2,370
ppm (9,000 mg/m3) for 60 minutes and at 8,300 and 16,600 ppm (31,540 and
63,080 mg/m3) for 30 minutes. Within 24 hours after exposure, all the
animals that were exposed at 8,300 and 16,600 ppm died. Delirium was
observed 3 minutes after exposure started at 16,600 ppm and within 14
minutes after exposure started at 8,300 ppm. This was followed by the
progressive depression of the CNS as previously described.
Fomin (1966) measured the latency time for defensive unconditioned
reflex reactions in rats exposed to" epichlorohydrin. Three groups, each
containing 15 white male rats (strain and weight unspecified), were exposed
6-35
-------
by inhalation to epichlorohydrin at concentrations of 0.2, 2.0, and 20.0
mg/m3 (0.05, 0.52, and 5.3 ppm). The animals were exposed continuously for
98 days. A fourth group, not exposed to epichlorohydrin, served as a
control.The latency time was measured weekly using a technique described by
Gusev and Minayev (1973). In this technique, the rat was placed in a
chamber with a floor containing parallel metal bars or plates on which its
extremities rested. The lid of the chamber contained a switch.that rested
on the back of the animal. Sufficient current was used in the parallel
metal plates to startle the rat, causing the switch on the back of the
animal to break the circuit. The time that elapsed between the application
of the electrical stimulus and the rat breaking the circuit was the latency
time for the defensive unconditioned reflex reaction.
The rats in the highest exposure group were hyperactive and restless
on the 1st day of exposure. This was replaced by depression and decreased
activity as exposure continued. After 1.5 months of exposure at 20.0
mg/m3, the latency time for defensive unconditioned reflex reaction increased
significantly in the exposed animals. The groups exposed at 0.2 and 2.0
mg/m3 epichlorohydrin had latency times similar to the control group.
Microscopic examination of the tissues and organs of the rats exposed at 20
mg/m3 showed abnormal changes in the lungs and kidneys similar to those
already described. The authors also observed damage to the neurons in the
medulla oblongata, cerebellum, and hippocampus. These changes were not
described in detail. No differences were observed between the animals
exposed at 0.2 and 2.0 mg/m3 and the control animals.
Kremneva and Tolgskaya (1961) examined the effects of prolonged
epichlorohydrin exposure on the CNS of two groups of rats. The first group
of eight rats (strain, sex, and age unspecified) were exposed daily (exposure
period unspecified; probably 2 hours/day) for 5 months to epichlorohydrin
vapor at 0.17-0.25 mg/1 (50-60 ppm). The threshold of irritation (stimula-
tion threshold) was measured in these experimental animals at various
intervals throughout exposure. The stimulation threshold was measured by
determining the amount of current required to elicit a withdrawal response
in the animals. No further experimental details were provided. The animals
6-36
-------
exposed to epichlorohydrin at 0.17-0.25 rag/1 showed an increase in stimula-
tion threshold (8 mA to approximately 9 mA) during the first 2 months of
exposure. At approximately 3 months, the threshold in the treated animals
decreased to 8.5 mA and then by the 4th month it had increased again to
approximately 9 mA. The control animals remained at a threshold of approx-
imately 8 mA throughout the study. In this study, two rats died after 1.5
months of exposure. The remaining animals died at the start of the 5th
month of exposure. The second group of 10 rats (strain, sex, and age
unspecified) were exposed to epichlorohydrin vapors at a concentration of
0.02 to 0.06 mg/1 (5-16 ppm), 3 hours/day for 6.5 months. The stimulation
threshold was also measured in these animals. The threshold was higher in
the animals exposed to epichlorohydrin from months 2 through 5. This
threshold was 7.5-8.0 mA compared with 6.5-7.0 mA for the control animals.
Six months into the study the stimulation threshold in the epichlorohydrin-
treated animals decreased to approximately normal levels. No animals died
at this exposure level during 6.5 months of exposure.
6.4 OTHER TISSUES OR ORGANS
6.4.1 Nasal Cavity
Irritation of the mucous membranes of the upper respiratory tract was
a common finding in laboratory animals exposed to epichlorohydrin vapor.
Irritation was normally followed by rhinitis (inflammation of the nasal
cavities) and degeneration and necrosis of the nasal mucosa (Quast et al.
1979a). These effects have been described in detail in Section 6.1.2.1.
Laskin et al. (1980) reported the development of neoplastic lesions of the
nasal cavity in rats during chronic inhalation studies.
6.4.2 Eyes
Lawrence et al. (1972) instilled 0.1 ml of different concentrations of
epichlorohydrin in cottonseed oil into the superior temporal quadrant of a
rabbit's right eye; the left eye served as the untreated control. The eyes
were then examined every 30 minutes for 3 hours and scored for the degree
of irritation. No irritation was observed at 5 percent, and doubtful
irritation was observed at 10 percent epichlorohydrin. However, an epi-
chlorohydrin concentration of 20. percent produced conjunctival and palpebral
irritation with edema; a 40 percent concentration produced iritis and
palpebral irritation with edema; and an 80 percent epichlorohydrin solution
produced cornea! injury.
6-37
-------
Smyth and Carpenter (1948) described cornea! injury of grade 4 to the
rabbit's eye (experimental methods not described) after epichlorohydrin
instillation. The amount of test compound instilled was not stated. Grade
4 was described as moderately severe cornea! injury.
Kremnera and Tolgskaya (1961) instilled a single drop of epichlorohydrin
into the conjunctiva! sac of a rabbit's eye. The epichlorohydrin produced
blepharospasm, hypermia of the mucosa, excessive lacrimation, papillary
constriction, and corneal clouding. The cornea! clouding had cleared after
2-4 days and improvement in the condition of the eye was noted. Complete
recovery was reported within 7-10 days following exposure.
6.4.3 Circulatory System
Kremneva and Tolgskaya (1961) found elevated blood pressure and some
pathologic changes in the myocardium (heart muscle) in rats exposed to
epichlorohydrin at 170-250 mg/m3 for 3 hours/day for 5 months. These changes
were described as moderate. No other reports of cardiotoxicity have been
found in the literature.
6.5 SUMMARY
Acute exposure to high levels of epichlorohydrin was shown to cause
CMS depression and death resulting from respiratory paralysis. The LC5o in
rats was 360 ppm (1,368 mg/m3) for 6 hours, and the no-observed-effect-
level (NOEL) was 283 ppm (1,075 mg/m3) epichlorohydrin for 6 hours. With
shorter periods of exposure, there were higher LC5o values. These LC5o
values were similar when rats and mice were analogously exposed. Intraper-
itoneal LD5e>s were found at 187 mg/kg for rats and 165 mg/kg for mice; and
oral LD56S for rats and mice were 248 and 236 mg/kg, respectively. In
addition, epichlorohydrin was acutely toxic by the dermal route; a single
immersion of a mouse's tail for 1 hour caused 100 percent mortality, and
0.64-1.3 ml/kg was the dermal LD5e» in rabbits. A single nonlethal dose
of epichlorohydrin can cause kidney and lung damage in rats.
Repeated exposures to epichlorohydrin were found to be highly irrita-
ting to the nasal cavity and to produce damage of the nasal mucosa in
rodents. Subchronic exposure to epichlorohydrin has been shown to cause
severe renal toxicity in rats via different routes of administration.
By both the inhalation and subcutaneous routes, epichlorohydrin has
been shown to cause changes in the lungs and bronchi. Subchronic exposure
6-38
-------
of rats to toxic levels of epichlorohydrin caused mild effects on the liver
and moderate changes in the myocardium. Epichlorohydrin has been found
irritating to the skin and eyes of rabbits.
6-39
-------
-------
7. CARCINOGENICITY, MUTAGENICITY, AND REPRODUCTIVE
AND TERAT06ENIC EFFECTS
7.1 CARCINOGENICITY
7.1.1 Introduction
The purpose of this section is to provide an evaluation of the likelihood
that epichlorohydrin is a human carcinogen, and, on the assumption that it is a
human carcinogen, to provide a basis for estimating its public health impact,
including a potency evaluation in relation to other carcinogens. The evaluation
of carcinogenicity depends in part on available animal bioassays and epidemiologic
evidence. Additionally, information on mutagenicity and metabolism (reviewed in
other sections of this document) particularly in relation to chemical interaction
with DMA and pharmacokinetic behavior, has an important bearing on qualitative
and quantitative assessments of carcinogenicity. This section presents
evaluations of the animal bioassays, each of which are followed by a qualitative
statement as to the evidence of carcinogenicity and its relevance to quantitative
human risk assessment. This section also evaluates the epidemiologic evidence.
The epidemiologic and animal studies are used, where appropriate, for human
quantitative risk assessment. Lastly, a summary and conclusions are presented
dealing with relevant aspects of the carcinogenicity of epichlorohydrin.
7.1.2 Animal Studies
In separate studies, epichlorohydrin exposure produced carcinogenic responses
proximal to the site of exposure. The results of these studies were as follows:
anterior nasal cavity squamous cell carcinomas in an inhalation experiment,
sarcomas at the site of subcutaneous injection, and papillomas and carcinomas
in the forestomach in a drinking water study and in a gavage (in water) study.
These carcinogenic responses were related to the amounts of epichlorohydrin to
which the test animals were exposed.
7-1
-------
7.1.2.1 Inhalation Exposure: Rat—Rats were exposed to epichlorohydrin vapor
(Aldrich Chemical Company, >_ 99 percent pure by gas chromatography, Laskin et
al., 1980). Exposures were perfomed in 128-liter or 1.3-m3 inhalation chambers.
Ambient epichlorohydrin levels were monitored spectrophotometrically during
exposures. Test animals were non-inbred male Sprague-Dawley rats initially 8
weeks old. Body weights were recorded monthly. Rats were allowed to live until
natural death or were killed in extremis. Necropsies were performed, and tumors,
lesions, and major organs were examined histopathologically. Heads were
fixed, decalcified, and sectioned for examination of the entire nasal cavity.
Based on the results of preliminary 1X50 studies, 100 ppm epichlorohydrin
was selected as the exposure level for a 30-day exposure period. Initially, 40
rats were exposed to 30 daily exposures of 6 hours each; subsequently, after
the exposure period for the 40 rats was completed, another group of 100 rats
was also given 30 daily 6-hour exposures. Both groups were followed by lifetime
observation.
Early mortality, attributed to respiratory disease, was higher in sham
(exposed to air only) and untreated control groups as compared to treated animals
(Figure 7-1). Body weight gain was comparable among groups. A maximum weight
gain of 200-220 percent achieved by 48 weeks following the first exposure was
sustained or slightly declined during the rest of the study.
Severe inflammatory changes in the respiratory tract were found in almost
all treated animals as well as in 90 percent of the control animals. Edema,
congestion, and pneumonia were observed in the lungs of exposed rats. Control
rats had congestion, edema, bronchiectasis, and pneumonia in the lungs. Control
rats had some expected kidney changes commonly found in aging rats. Renal
damage, including dilatation of cortical and medullary tubules, was found in
63 percent of the treated rats.
7-2
-------
01
II I I I I I I
I I I I I I I i I
00
CM
CN
LU
CD OC
cn D
O
Q.
X
O
00
CO
(A
cc
tr.
LU
(/}
oo 'LU
CD
o
o
o
CD
CM
3AI±ViniAinO lN3OH3d
£= CL)
S- -U
-o e
o*—r-
Q. to 13
,
o .c
«t- i- o
o •*-» s-
s§3
CL o JC
Q. O
•O 'r-
CD O) CL
O -P
,|_i -[ ^
i -a
O S- O)
S- .'r- tn
3 X
CL C CD
X CU
at (r> tn
eu r—
en s- IB
c CL g
••f-' to «
•••-• >,
i — to ~O
co "a c
+•> ns
s- o
O OO o
I
i — .
>>
ns to
o
oo
CTl
cu
c
to
ns
OJ •— O
S- S- S_
a -e -t->
en c
•r- 10 O O
U_ s—^ cj c/}
_
a:
7-3
-------
Squamous cell carcinomas were found in the nasal tracts of treated animals,
as enumerated in Tables 7-1 and 7-2. Many of these nasal carcinomas infiltrated
the proximal bones of the skull, but metastasis to distal organs was not found.
The nasal carcinomas were described in terms of their histologic appearance
as well-differentiated tumors with keratin pearls. All of the nasal carcinomas
appeared in the anterior portion of the nasal cavity. Some of these solid
tumors almost filled the nasal cavity, thereby causing dyspnea and wheezing.
Three other rats were diagnosed with either nasal or bronchial and larynx
papillomas. The incidence of other tumor types in nonrespiratory organs was
similar between treated and control groups, thereby indicating the lack of
epichlorohydrin-induced carcinogenicity in other organs, and the lack of
metastasis from the proximal site, the nasal cavity.
TABLE 7-1. SQUAMOUS CELL CARCINOMAS OF THE NASAL CAVITY OF MALE WISTAR
RATS FOLLOWING THIRTY 6-HOUR EXPOSURES TO 100 PPM EPICHLOROHYDRIN VAPOR
Experiment
1
2
Dose
(ppm)
100
100
No. of
animals
40
100
No. of
tumor-bearing
animals
(percent)
4 (10)
11 (ID
Tumor observation time
Mean
540
623
(days)
Range
462-610
330-933
SOURCE: Laskin et al., 1980.
A second study was done in which 100 male rats per treatment group were ex-
posed to 10 or 30 ppm epichlorohydrin 6 hours/day, 5 days/week, for their lifetimes,
Treated animals were compared with concurrent sham and untreated control groups.
7-4
-------
TABLE 7-2. DOSE-RESPONSE FOR INDUCTION OF SQUAMOUS CELL CARCINOMAS IN
THE NASAL CAVITY OF MALE WISTAR RATS EXPOSED TO EPICHLOROHYDRIN VAPOR
Concentration
ppm
100 (combined studies)
30
10
Air (sham) controls for life
Untreated controls
Total
Number of exposure
exposures (ppm x days)
30 3,000
290a 8,700
250* 2,500
— ---
— —
Cancer incidence
(No. with cancer)
(No. exposed)
15/14QDC
1/100
0/100
0/100
0/50
aLifetime exposures were based on median survival time.
bP < 0.00001 versus combined controls.
cThere were also two nasal papillomas in this group.
SOURCE: Adapted from Laskin et al., 1980.
Early mortality was high in all groups, with 50 percent mortality evident
by 64 weeks (Figure 7-2). Lung congestion and pneumonia were common in decedents,
Body weights were lower in the 30-ppm group, as shown in Figure 7-3.
Two respiratory tract tumors were found: a larynx squamous papilloma (at
90 weeks) and a squamous cell carcinoma (at 107 weeks) in two rats exposed to
30 ppm. The Laskin et al. (1980) paper incorrectly described the larynx papil-
loma as a nasal papilloma. This was corrected in a telephone conversation
with one of the authors. Severe inflammation of the nasal cavity was noted in
90 percent of the control animals. Exposure to 10 and 30 ppm epichlorohydrin
produced 2 and 4 percent incidences, respectively, of squamous cell metaplasia
in the nasal cavity, with none in the controls. Squamous cell metaplasia is
the transformation of the nasal mucosal epithelium into stratified squamous
epithelium.
7-5
-------
I I I I i I I I I
I I I I i I i I I I
AinviaoiAi 3Aii\nni/\ino iiviaoaad
c/> «-*•
>> •
«-—' £
-O O.
n Q.
if, i—>
*~»O
" rrl
>>
-^. S-
S_ 4-> O)
jr c s_
O 3
>—' O
-D Q.
OJ O) X
S- 4J Q)
3 tO
10
O>
r— >
i— S- O)
O 3 S-
«»- O 3
to O) O
-(-> -C D-
(0 I— X
S-
o c
•r- O
• Q. O
CM O)
i -o
r-» o o>
-P 4J
0) (O
S-^-> 0)
3 -^ S-
CD O) 4->
•r- OJ C
U- 3 3
O
CO
(O
o
o;
o
7-6
-------
ro
I i i I I I I I I I i
I I I I I I
5
00
CN
Ol
8
<*
to
$
S
LLI
oc
(O
o
0.
X
LLJ
cc
LU
0}
*:
IU
UJ
S
CO
.c
O IO
o o
i— S-
o s=
••- o
Q. O
CU
O CU
tO E
O 3
o.
X •>
CU »-»•
^D
(O
CD
•i- M-
s- o
o
>, in
£= 0)
O £
t. T-
O •!->
i— CU
J= <4-
O T-
O.
Ol CU
O
S_
o
S_
Q. O)
Q. O)
3
O --^
i— i 1/1
>>
S- (O
o -&
o.
o
O -i->
4- to
O)
to s_
-i-> +J
ns i
S- i-
O vo
i.
o o
•»-> CU
o o
i- S-
CD CL
CO to i—
10 o
o -a
a.
x cu
-------
Renal damage occurred in 65, 37, 24, and 17 percent of the 30 ppm, 10 ppm,
sham, and untreated groups, respectively. Severity of renal damage, diagnosed
as mainly tubular degenerative changes, was related to epichlorohydrin dose.
The results of the Laskin et al. (1980) study have provided qualitative
evidence for the carcinogenicity of epichlorohydrin. Epichlorohydrin produced
a much greater increase in nasal cavity carcinomas with a high dose given early
in the study (15/140 plus two nasal papillomas), but produced a much lesser
response (1/100) when given as a lifetime treatment with one-third of the
higher concentration. The authors stated that nasal carcinomas had not been
observed in 1,920 control rats over 14 years in their laboratory; however, no
data were made available on the rate of nasal inflammation. Laskin et al.
(1980) hypothesized that the latency period for cancer development could have
been shorter with the more intense (though shorter) exposure, whereas in the
lifetime study at 10 and 30 ppm the latency period could have been longer,
perhaps approximating or even exceeding the rats' lifetime. Furthermore, the
rather high mortality in the lifetime exposure groups reduced the number of
animals available for development of late tumors.
Since epichlorohydrin was not found to be a complete carcinogen in a
dermal study (Van Duuren et al., 1972a, b), there is a question as to whether
nasal tumors would have been observed in the absence of nasal inflammation.
One of the authors of the Laskin et al. (1980) study, Roy E. Albert, stated
that inflammation in control rats used in their laboratory for lifetime inha-
lation carcinogenicity studies is not apparent before 1 year. Since exposure
to the carcinogenic 100-ppm exposure level of epichlorohydrin occurred dur-
ing the initial 30 exposure days (at age 8 weeks) of the Laskin et al. (1980)
studies, the probability that nasal inflammation as observed in matched controls
could have been an initiating event in the induction of nasal carcinomas is
7-8
-------
considered to be low because inflammation was not observed during the early or
initiating-event period. It has generally been observed that most promoters
are irritants, but that not all irritants are promoters. There is presently no
evidence available to suggest a promoting action of epichlorohydrin. Further,
it has not been determined that the observed nasal inflammation necessarily
caused the induction of the observed nasal carcinomas (personal communication
with R.E. Albert).
The Laskin et al. study indicates qualitatively that epichlorohydrin is an
animal carcinogen. The SOx-exposure study showed a significant increase in a
rare tumor type, nasal carcinomas, 15/40 vs. controls (P < 0.0001), whereas
the 30-ppm response in the lifetime exposure response (1/100) was not signifi-
cant compared to internal controls (P > 0.05) but was significant when compared
to external control rat nasal carcinomas (0/1920) observed in the authors'
laboratory over 14 years (P = 0.05). On these bases, the Laskin et al. study
can be used to extrapolate a 95% upper-bound risk estimate for humans.
7.1.2.2 Oral Administration in Drinking Water: Rat—Konishi et al. (1980)
and Kawabata (1981), both reporting on the same study, described a carcino-
genicity bioassay on epichlorohydrin given orally in drinking water to rats.
The epichlorohydrin (Hani Kagaku, Kyoto) was 99.96 percent pure; impurities, if
known, were not reported. Seventy-two male outbred Wistar rats, 6 weeks old
and each weighing 160 g, were divided into four groups of 18 rats each. Six
animals were housed in each cage. The animals were given fresh solutions of
epichlorohydrin in drinking water each day. The epichlorohydrin solutions were
protected from light. One group served as untreated controls, and the other
three groups were treated with 375, 750, or 1,500 ppm epichlorohydrin. These
concentrations of epichlorohydrin in water (ppm) were initially 0, 62, 95.5, and
187 mg/kg body weight/day (approx.). The amount of epichlorohydrin ingested per
7-9
-------
rat was constant throughout the experiment (because daily water consumption was
constant), and thus, the dosage rate (mg epichlorohydrin/kg/day) varied inversely
with body weight. The total epichlorohydrin dosage for the whole experiment
was 0, 5.0, 8.9 and 15.1 grams for each of the groups. Although survival among
all groups was similar, treatment with epichlorohydrin was discontinued for
short periods after 60 weeks due to debilitation of the rats (Figure 7-4). It
was concluded that pulmonary infection was the cause of death in animals that
died during the course of the experiment. Epichlorohydrin intake patterns
during the 81-week experimental period are shown in Figure 7-5.
All of the animals were necropsied at 81 weeks terminal sacrifice, and
tissues and organs, as well as tumors, were examined histopathologically.
However, the authors did not report pathologic results for animals that died
during the study. Major organs from survivors were weighed, and blood was
collected from survivors for biochemical and hematologic analysis.
Dose-related decreases in body weight gain occurred as shown in Figure 7-6.
Statistically significant (P < 0.05) increases in organ/body weight ratios were
common in the groups given 750 or 1,500 ppm epichlorohydrin, largely due to com-
parable organ weights among control and treatment groups but decreased body
weights in treated animals. A significant (P < 0.05) increase in pancreas/body
weight ratios was also evident in the 375-ppm group. The results of the pathologic
examination of the kidneys were not reported; however, significant (P < 0.05)
increases in both kidney weights and kidney/body weight ratios in each treatment
group (Table 7-3) may be indicative of injury at this organ site from treatment
with epichlorohydrin, since treatment-related kidney damage was also seen in the
Laskin et al. (1980) study, in which exposure was by inhalation.
The results of blood analyses were normal except for significant (P < 0.05)
increases in cholesterol and neutral lipid levels in each treatment group as
7-10
-------
GROUP 0 20
NO. Ly/_l_
EXPERIMENTAL PERIOD, weeks
40
60
_JL
81
WATER WITHOUT EPICHLOROHYDRIN IN DRINKING WATER
375 ppm EPICHLOROHYDRIN IN DRINKING WATER
750 ppm EPICHLOROHYDRIN IN DRINKING WATER
1500 ppm EPICHLOROHYDRIN IN DRINKING WATER
Figure 7-4. Patterns of epichlorohydrin administration in male Wistar rats,
SOURCE: Konishi et al., 1980.
7-11
-------
CM CO
o
-CD
O
r-
_o
CO
o
— LO
_ O
0
CO
c/>
•!£
0)
CD
£
*"""
•0
•—
CD
Q.
WH
CD
C
1
v_
o>
Q.
X
LU
C
S- -P
•o i o
to >>••- c
CU JC Q.
C O Li-1 10 C
•r- S- CO CO
i — O 2
r- ' • >>
CU iC C C jO
.C O O O
1— ~'r- 'r- -r- "O
Q..P -P CU
UJ Q. CO -P
• E £- ^
to 3 CO E
•P • (/) O *^
CO C CU -P
s_ o. o s- to
3 O O CU
s- o s-
CO S- S- C CU
•p en cu o >
to -p -I- O
^
A c/i CU i —
>>^^ C C td
f^t +^^ Q >
, — . CO CM iC
C S 0
•>t-4 QJ S-
•^ j__ fC~ Q
C *— *' CO -P r—
S- E 4-> CD'O
•O Q-'i- C •!-
>, CL C 'r- Q.
J= «'r- S- CU
O O 3
s- in E tJ <*-
O r-^ 3 O
r— O tO
.E « C •!- (0
_ o
\
O
I
O
CO
I
o
cs
1
o
QJ • O
»— ' 5-
q_ t)_
O E
0.-0
QJ O. QJ
^: •(->
co LO re
• -p
O
0) r—
CO
O
i — CO
O 4->
J- C
CU -r-
S- S- -P
CU O C
-P *4— CO
CO CU 3
3 S- CT
cu
c j= -a
•r** >P CU
I/)
(Aep/iBJ/Buu) U!
ai
cu to cu
i- cu ^
3 S- CO
CD CL-P
•i- CU C
U_ S- «r-
O)
>,-o
J= CU
O 5-
S- 3 -P
o to cu
i— CO T3
^: cu c
O E 3
CO
cr>
CO
4->
CO
ja
CO
i
UiJ
O
C£.
O
CO
7-12
-------
t/1
-P E
CO Q.
•i- a.
•3.
in
•r- »i—
CD
•M
J= CO
en ex
•r- 3
o» o
S 5-
CT
-O Q)
O W
.Q O
so
^: o ' —
o o _.
^ •
O T3 ^
r— O> ---
O (O E
•r- OJ Q.
Q. i. Q.
o; -P
c o
M- =5 O
O LO
4-> .— I
4-> C
O OJ T3
O) tfl C
«t-
I O)
r-. >
00
(B
*
JO
re
o:
o
oo
'1HOI3M AOO8
7-13
-------
TABLE 7-3. KIDNEY WEIGHTS AND KIDNEY/BODY WEIGHT RATIOS IN MALE WISTAR RATS
GIVEN EPICHLOROHYDRIN IN DRINKING WATER FOR 81 WEEKS
Dose
(PPm)
0
375
750
1500
Number
Initial
18
18
18
18
of rats
Effectiveb
10
9
10
12
Body wei
(g + s.
Initial
157+J.O
159+_ 6
157+_ 8
160+_ 7
ghta
D.)
Final
595+75
494+45
415+46C
295+46C
Organ weights
(% of body
Left
1.7+0.2
(0.31+0.07)
2.2+0.6C
(0.44+0. ll)c
2.1+0.2C
(0.52+_0.07)c
1.9+0.2C
(0.66+_0.09)C
(g + S.D.ja
wei gift)
Right
1.6+0.2
(0.31+0.07)
2.1+0.4C
(0.31+0.08)
2.2+0.2C
(0.53+0.05)c
1.9+0.2C
(0.65+_0.09)c
aS.D. = standard deviation.
bBased on rats sacrificed at 81 weeks.
cp < 0.05.
SOURCE: Adapted from Kawabata, 1981.
compared to controls. The relationship between these blood analysis results
and epichlorohydrin treatment is presently not understood, and a stronger
examination of this response to epichlorohydrin treatment could have been made
if blood analyses had been done prior to treatment and periodically throughout
the study.
Dose-related pathologic changes in the forestomach were diagnosed as shown
in Table 7-4. Macroscopic examination of forestomachs revealed uneven
protuberant large tumors and "countless" small nodular tumors. The lesions
were histologicaly diagnosed as hyperplasia, papilloma, and squamous cell
carcinoma. Both localized and diffuse hyperplasia were observed. Proliferation
of squamous epithelium and multistage stratification of basal cells were observed
7-14
-------
CO
H-
^c
fy
Qi
1 •--
CO OO
1— 1 NX
3 LU
LU
LU 3
1
^C »—*
sr oo
sr Q;
1—1 o
u.
LU
O Oi
•Z. LU
LU (•—
t 1 ^c
i— i 3
^_^
Z. f n
(—4 Z!
i— t
O ^
E: 2=
Q:
CO Q
o;
o z:
3E! " — '
^3
1 — - «ST
I — i
O O
"^ ^~
0 0
OO O
LU _J
as z
P°
U- *-<
CL.
U— LU
*-*
C£. LU
LU >
m t— i
y C5
ID
Z.
•
1
1 —
LU
CO
H-
C <-~
O S«
*— ^
in
o> E
C 3
•r— •!—
"O r~~
C CU
•t- .C
M — 4-9
•r—
i— Q.
re cu
0
•c- in
O) 3
0 0
'o re
4-> 3
in o"
•i— in
1C
re
g
°
Brj
0
t_
re
o
re
f—
r*^
tr*~
CL
Q.
re
•^-1
in
re
CL
CO
Q.
T-
40
re
c_
t_
CO
Q.
.C
o
1
40
tn
cu
£_
o
^~
c:
•1—
in
£_
O
5
,,_
O
f
cu
_Q
E
3
"Z.
CM
V
LO
— - CM
E
g
.w^-
CO
N
in
O
3 LO
1— A
r—
re
4->
1^
in
re
ll-
CS
£_
CO
E
3
"Z.
re
cu
•i —
4_>
O
<«-
LU
"re
•«—
4-9
•r-
C
1 — 1
cu ^^
0 CL
O Q.
s"--x
*—* x-^
0 0
o o
^-^ ^^--.
0 0
*»—* *^— -*
o o
^ — ^
00
--^ r-.
0 1
— — •
<"s f«^
CM
OO
O -f- |
p«^
CO
^_^
o
o
t— 1
> ^
f— 1
1^^>
0
•
o
t— 1
^_^
.1—1
X— ^
o
•
0
Ol
—
en
o
o
t—
4->
o
re
LU
in
i.
CU
s:
in
•r-
U-
cu
• T—
re
4-J
i
CO
e~
o
-Q
r-t
O
0
O
V
a.
f,
-&
cu
• fH»
>> o
O res
t/) O
r~~ £T
1?'
c re
re o
T3 T3
CU c:
u re
M! in
•r- re
t- E
O O
re i —
in i—
•— re
i — CL
re
c: t-
•i— O
cu
in
CU i— •
"=£^>
t— +J C
0 c: 0
c: o
•i- O CU
o
i/) in -r-
CL 3 it-
c! 52 r
t_ CO O
CD > re
in
CO LO
> 0 r-
•i- o re
4-J • c:
O O •!-
CJ || £
4— CU
LU Q- 4-J
re f>
4
o
co
i— t
n
1v
-p
OJ
•1—
(/I
•1 —
c
o
• •
LU
o
— ^
O
OO
7-15
-------
in hyperplastic regions. Papillomas consisted of squamous epithelium projecting
into the lumen. Marked keratinization with little nuclear division was apparent
1n the papillomas. Carcinomas were characterized as highly differentiated,
keratinized squamous epithelium that proliferated and invaded the basal
membrane. Irregularly sized nuclei and characteristics of nuclear division
were common in the carcinomas. Metastases of the carcinomas were not found.
Additional tumor findings included squamous cell carcinomas of the oral cavity
1n two rats given 1,500 ppm epichlorohydrin, and interstitial cell tumors of the
testes in two to four rats in each group, with no relation to dose.
The results of the study by Konishi et al. (1980) and Kawabata (1981) pro-
vide contributing evidence for carcinogenic activity of epichlorohydrin by the
oral route of administration in the forestomach of male Wistar rats. However,
there were several problems with the study. First, a stronger indication of
carcinogenicity could possibly have been obtained if the study protocol had
included larger numbers of animals and lifetime treatment and observation.
Second, it is not known whether pathologic changes suggestive of carcinogenic
activity of epichlorohydrin were also evident in animals that died during the
study, since pathologic data for these animals were considered unreliable by
the authors. Third, the dose levels in water of epichlorohydrin used were
toxic to the rats, as indicated by the reduction in body weight and the need to
periodically stop treatment after 60 weeks; this indicates that the MTD was
exceeded. Fourth, the epichlorohydrin had a half-life in water of 1,000
minutes (0.69 day) in this experiment, and yet solutions were made daily;
epichlorohydrin dosage was thus likely to have been overestimated.
Nonetheless, induction of forestomach neoplasia as seen in this Japanese
study by a direct action of epichlorohydrin is supported by: 1) a Dutch oral
gavage study (forestomach tumors), 2) other studies discussed herein that
7-16
-------
indicate a direct tumorigem'c action of epichlorohydrin at other exposure
sites, and 3) the chemical nature of epichlorohydrin as an alkylating agent.
The upper-bound unit risk can be estimated on the basis of the Konishi
et al. study, but such an estimate will be of limited reliability because of
the above-described problems with the study.
7.1.2.3 Oral Administration-Gavage: Rat—G.J. van Esch (Rijsinstitut voor
de Volksgezonheid Bilthoven, 1982) has written a laboratory report on the induc-
tion of forestomach lesions in SPF-derived Wistar RIV:Tox rats as a result of
epichlorohydrin exposure by gavage at rates of 2 and 10 mg/kg body weight/day
(administered freshly made in water). Controls and both treated groups all had
50 males and 50 females. As of October 1984, this preliminary report had not
been completed, peer reviewed, or published.
Intergroup comparisons showed no compound-related effects on weight during
the course of the 104-week experiment. Females showed increased early mortality
(up to 40 weeks) (Table 7-5) as compared to males. Increased mortality in both
males and females in the 10-40 week period was attributed by van Esch to
"hairballs" or trichobezoars in the stomach (as previously seen in other experi-
ments). This situation was corrected by diet change with different fiber
content at 56 weeks, after which the mortality rate was normal. The survival
to 104 weeks (2 years) is judged to be normal for this strain or rat.
The hematology findings were unremarkable except in the case of 10-mg/kg
dosed females, which showed decreased leukocyte counts (mainly eosinophilic cells).
The male leukocyte counts seemed to be somewhat decreased, but the variation in
the white cell counts was too great in males for this determination to be made
with certainty.
After 20 months of exposure, all but one rat in the high-dose group
(10 mg/kg) showed macroscopically neoplastic lesions in the forestomach mucosa,
which varied from small protrusions to rough surfaces to verrucous masses
7-17
-------
c o»
•i- C Cfl
ui i— .ai
C *r- O ^J
•a; .^ **- oo
UJ
1
CO
o
Cu
o
i
$
CO CM CO
i-H 0
CO CO
O
CO
co
CM
O
1^.
1
O -T-
to c
I <0
0)
-g
f-H CO CO
T-H tO IO
o
CO
I
T-H CO tO
f-H CO ID
CM -a- ID
CM to en
CM CO
O CO CO
O T-H T-H
O T-H O
000
•: a a 42 a a
o o
t_ en en ui e- en cr>
+J E E CJ 4J E E
OJ
(O
O
o
CVJ
S2
> O
i— (/> (O
r— CO E
re ui o
.,_ (jj 4J
a i- c
g c «
!=§15
7-18
-------
occupying extensive luminal space to masses with ulcerations and necrosis. In
several cases, the stomach wall was observed to be thickened. The low-dose
group (2 mg/kg) showed macroscopically far fewer neoplastic lesions of the
forestomach, and the lesions were less extensive in pattern, with the first
lesion seen after 22 months. Only one control rat showed a macroscopic neo-
plastic lesion. The observed incidences of forestomach macroscopic changes at
autopsy are given in Table 7-6.
TABLE 7-6'. INCIDENCE OF RAT FORESTOMACH MACROSCOPIC NEOPLASTIC LESIONS
Sex
Male
Female
0 mg/kg
0/50
1/50
2 mg/kg
6/50
1/50
10 mg/kg
39/50
23/49
SOURCE: van Esch, 1982.
Histopathology showed forestomach proliferative changes mainly in the 10-
mg/kg group (HOT). Table 7-7 shows the chronology of discovery of forestomach
hyperplasia, papillomas, and carcinomas. The summed histopathologic occurrence
of carcinogenic activity (papillomas and carcinomas) by epichlorohydrin in the
rat forestomach in the van Esch study is shown in Table 7-8.
Distal tumors were not described by van Esch, but the lung was described
as being tumor-free (no aspiration-induced tumors or metastases to the lung).
One in situ tumor of the esophagus was mentioned in the high-dose group.
This study presents better oral exposure incidence data in the rat fore-
tomach than the Konishi et al. study because more rats per group were used, the
MTD was apparently not exceeded as in the Konishi et al. study, and health
7-19
-------
TABLE 7-7. HISTOPATHOLOGY OF PROLIFERATIVE LESIONS IN THE FORESTOMACH
OF RATS GIVEN EPICHLOROHYDRIN (ECH) BY GAVAGE
Period (months) 0-12 12-14 14-16
Group I control
females
no. of animals iu(i)
died or kllleda
Hyperplasla
Papllloma
Carcinoma
Group I control
males
no. of animals 1
died or kllleda
Hyperplasla
Papllloma
Carcinoma
Group II, 2 mg ECH/
kq BW. females
no. of animals I9(s>) z
died or kllleda
Hyperplasla 1
Papllloma 1
Carcinoma
Group III, Z mg ECH/
kq BW, males
no. of animals 3 1
died or killeda
Hyperplasla 2 1
Papllloma
Carcinoma
Group II, 10 mg ECH/
kq BW, females
ho. of animals 19(6) 2(1) z('<2)
died or killeda
Hyperplasla 5 1
Papllloma
Carcinoma
Group III, 10 mg EHC/
kq BW, males
no. of animals s>(i) . i
died of kllleda
Hyperplasla 1 1
Papllloma
Card nctna
16-18 18-20 20-22 22-24 24-26 26-28 28-30 cumulative
2346 10(1) 15(1) 50(3)
2 1 3
11 2
0
1 4 4 4 15 21 50
113 5
I I
0
1115 6(1) 3 12 50(6)
1325 12
1 2
1 1 2
2 2 9(1) 2 7 12 12 50(1)
1 3 1 2 7 7 24
1 236
312 6
2(1) 2 8 3 11 49(10)
I
2 8 3 11 24
1 2 6(1) 6(1) 8 13 8 50(3)
12 1 6
1 1 2
3 4 7 13 8 35
due to advanced autolysls.'
8K1lled 1n moribund state or at termination of the experiment.
SOURCE: van Esch, 1982.
7-20
-------
TABLE 7-8. HISTOPATHOL06Y OF PAPILLOMAS AND CARCINOMAS IN THE FORESTOMACH
OF RATS GIVEN EPICHLOROHYDRIN BY GAVAGE3
Sex
Male
Female
0 mg/kg
1/49 (2.0%)
2/38 (5.3%)
2 mg/kg
12/45 (26.7%)b
4/28 (14.3%)
10 mg/kg
37/42 (88.1%)c
24/26 (92.3%)c
aAll animals that died prior to appearance of first tumor and those for which no
histologic examination was performed have not been included.
bP < 0.01
cp < 0.00001
SOURCE: van Esch, 1982.
status and survival were better. Even though the early high mortality from tri-
chobezoar obstruction could indicate a diet-treatment interaction effect, the
Carcinogen Assessment Group (CAG) concludes that the van Esch epichlorohydrin-
induced response in the rat forestomach is strong qualitative evidence (test for
linear trend, P < 0.0001 for both sexes) that epichlorohydrin induces carcinog-
enic activity at 2 and 10 mg/kg body weight by the oral route of administration.
7.1.2.4 Dermal Exposure: Mouse—Van Duuren et al. (1974) reported the results
of a topical application study of epichlorohydrin on female ICR/Ha Swiss mice
(6-8 weeks of age). The epichlorohydrin sample (Eastman Organic Chemicals) was
purified by distillation and checked for purity by infrared spectroscopy, nuclear
magnetic resonance spectroscopy, and gas chromatography; the epichlorohydrin
sample was 99.8 percent pure (personal communication with B.L. Van Duuren).
Dose selection was based on the results of preliminary 4-week tests, and the
highest possible doses producing minimal cytotoxicity were used.
Fifty mice received 2 mg epichlorohydrin in 0.1 ml acetone thrice weekly
on the clipped dorsal skin. Thus, epichlorohydrin was tested as a complete
7-21
-------
carcinogen. The study lasted for 580 days, and the median survival time was
506 days. No skin tumors were observed, and it is concluded that epichlorohydrin
is not a complete carcinogen.
Weil et al. (1963) painted one "brushful" of undiluted epichlorohydrin
(purity not reported) onto the clipped dorsal skin of 40 C3H strain mice,
initially 90 days old, thrice weekly for life. Thirty were alive at 17 months,
and one survived for 25 months. No local or distant tumors.due to the effect of
epichlorohydrin were found in this study, which corresponds to the results of
the repeated topical application study of epichlorohydrin in female ICR/Ha mice
by Van Duuren et al. (1972a, b; 1974) described previously. The mice used by
Weil et al. (1963) were initially 90 days old, which did not allow an evaluation
of carcinogenicity during early growth of the animals. A "brushful" does not
give any indication of the actual dose applied. This is supportive evidence
that epichlorohydrin is not a complete carcinogen.
7.1.2.5 Initiation-Promotion: Mouse—In an initiation-promotion study on
mouse skin, Van Duuren and coworkers (1974) applied single doses of 2 mg
epichlorohydrin (99.8 percent pure) in 0.1 ml acetone to the dorsal skin of
30 female ICR/Ha mice, followed 2 weeks later by thrice-weekly skin applica-
tions of 2.5 ug phorbol myristate acetate in 0.1 ml acetone for the duration
of the experiment (median survial > 385 days). Nine of thirty (9/30) mice
developed skin papillomas (the first observed at 92 days), and one mouse
developed a skin carcinoma. Of 30 control mice treated with phorbol myristate
acetate alone, three developed papillomas (the first at 224 days), whereas no
tumors occurred in 30 sol vent-treated controls. Thus, epichlorohydrin was
shown to be a tumor initiator, requiring complementation by a promoter in
this system. Qualitatively, it may be surmised that epichlorohydrin is an
initiator and not a good promoter or not a promoter at all. In relation to
7-22
-------
the quantitative risk assessment, these data suggest that epichlorohydrin acts
early in the carcinogenic sequence.
7'1>2'6 Subcutaneous or Intraperitoneal Administration: Mouse—For an assay by
subcutaneous injection into the flank, 50 mice were given 1 mg epichlorohydrin
in 0.5 ml tricaprylin (highest possible dose producing minimal cytotoxicity in
a 4-week preliminary test) once each week for 580 days. Median survival time
was 486 days. Van Duuren and coworkers (1974) reported that six mice developed
local sarcomas and one had a local adenocarcinoma (P <_ 0.05), whereas only one
local sarcoma occurred in 50 tricaprylin-treated controls.
In an intraperitoneal assay by Van Duuren et al. (1974), 30 mice received
weekly injections into the lower abdomen of 1 mg epichlorohydrin in 0.05 ml of
tricaprylin for 450 days. None of the mice developed local sarcomas, but 11
had papillary lung tumors. Of 30 tricaprylin-treated control mice, 10 had
papillary lung tumors and one had a local sarcoma. Thus, epichlorohydrin pro-
duced local sarcomas at the site of subcutaneous injection but did not produce
distant tumors after intraperitoneal injections.
Kotin and Falk (1963) administered single subcutaneous injections of 5 uM
(462 ug) of epichlorohydrin in 0.1 ml ethyl laurate or tricaprylin to 30 C3H-
strain mice, which were observed along with sol vent-treated control mice for 2
years. Of the experimental mice, four showed malignant lymphomas within 6
months, one showed a skin papilloma after 11.5 months, one showed a hepatoma
after 13 months, and one showed two lung adenomas after 24 months. However,
survival was poor (12 mice died during the first year) and, except for the
papilloma, the tumors were of similar types and not significantly higher in
frequency than those in the control group. Animals were given only one
treatment with a rather low dose of epichlorohydrin at the beginning of the
study; this dosing procedure appears weak for carcinogenicity testing compared
7-23
-------
to a stronger challenge of repeated treatment over a lifetime at doses as
high as those maximally tolerated.
These studies indicated that epichlorohydrin is an administration-site or
proximal-site animal carcinogen, but that it is not transported to form distal
cumrrs. The tumors that are formed by epichlorohydrin apparently do not metasta-
size.
7.1.3 Epidemiologic Studies
A retrospective cohort mortality study of epichlorohydrin workers was
conducted for Shell Oil Company by Dr. Phillip Enterline of the University of
Pittsburgh (Enterline, 1978, 1981). The cohort of 864 comprised workers from
Shell plants at Norco, Louisiana, and Deer Park, Texas. Deaths were compared by
cause with those expected in Louisiana and Texas, respectively. Results were
analyzed by vital status as of December 31, 1977 (reported by Enterline in 1978)
and in the most recent update by vital status as of December 31, 1979 (reported
by Enterline in 1981) for the cohort exposed to epichlorohydrin for at least one
quarter before January 1, 1966. The Carcinogen Assessment Group (CAG) previ-
ously reported on the 1977 update (CAG, 1980). Those data (see Table 7-9 plus
footnote) showed less observed mortality than expected (54 versus 97.3, respec-
tively) but also showed an increase (which was not statistically significant)
in both respiratory cancer and leukemias with overall standardized mortality
ratios (SMRs) of 146.2 and 224.7, respectively. Furthermore, the data published
in 1978 showed an apparent increase with increasing latent period since, of the
12 respiratory cancer or leukemia deaths, 11 occurred in workers 15 years or more
after first exposure (Table 7-9). Even though these increases were not sta-
tistically significant, the trend provided reason for concern that increasing
observation time would produce more of these cancers, leading to positive con-
clusions about the carcinogenicity of epichlorohydrin in humans.
cs
7-24
-------
TABLE 7-9. COMPARISON OF MORTALITY IN ENTERLINE'S EPICHLOROHYDRIN STUDY
UPDATES BY CAUSE AND BY LATENCY
(1978 versus 1981)*
fi
Enter! ine
(1978)
(1981)
Time since
rst exposure
Overall
< 15 years
>^ 15 years
Overal 1
< 15 years
_>. 15 years
Al 1 cases
Observed/
Expected
54/97.3
19/45.8
35/51.6
65/115.7
19/46.0
46/69.8
Respiratory
cancer
SMR
55.5
41.5
67.9
56.2
41.3
65.9
Observed/
Expected
10/6.8
1/2.2
9/4.7
10/8.7
1/2.2
9/6.5
SMR
146.2
45.9
193.1
114.1
45.0
137.8
Leukemia
Observed/
Expected
2/0.9
0/0.5
2/0.4
2/1.0
0/0.5
2/0.5
SMR
224.7
0
500
194.2
0
377.4
^Included in the 1981 report are two additional deaths missed in the 1978
reports—one due to lung cancer and one due to heart disease. The numbers
above referring to the 1978 report include this correction.
7-25
-------
However, the most recent data (Enterline, 1981) have produced a reversal
of the trend of respiratory cancers and leukemia deaths. This is shown in
Tables 7-9 and 7-10. In this latest 2-year followup period, 1978-1979, there
were 11 additional deaths, only one of which was due to cancer, and this was
not respiratory cancer or leukemia. As can be seen with respect to the SMRs
for both respiratory cancers and leukemia, this most recent update has shown
decreases in both of the overall SMRs, and especially in those for the group
with greater than 15 years since first exposure. However, neither of these
SMRs is statistically significant.
In the most recent update, Enterline (1981) also presented a smoking
history of 12 of the cancer deaths. He found that for the 10 deaths diagnosed
as lung cancer deaths on death certificates, 7 individuals were known smokers,
1 was a nonsmoker, and 2 had unknown smoking histories. This confounding
factor makes it even more difficult to assert a positive causal relationship
between epichlorohydrin and human lung cancer.
Additionally, Enterline considered the severity of epichlorohydrin expo-
sure. Regarding the group with at least 15 years since first exposure, he
stratified "heavy to moderate" versus "light to nil" groups. This analysis
failed to show a dose-response trend, as the death ratios for cancer in both
groups were similar.
Finally, there is a problem of exposure to multiple chemicals. This is
examined in two separate studies by Enterline that share some of the same
cohort (Table 7-11). In the 1981 study, Enterline provided a further analysis
contrasting the mortality experience of 124 men from Deer Park who had prior
exposure in the isopropyl alcohol (IPA) unit.* The results show that the
*The exposures to IPA and epichlorohydrin were considered by the CA6 in its
most recent report (1980). It was concluded that the confounding effects
between the exposures detracted from the significance of the findings.
7-26
-------
LU
LU O£
O CO
ST 0
O.
QL X
O LU
"- (_
S3
CO »-«
O Lu
O_
X LU
LU O
Z
CO I-H
LU co re
rf uj r-~
2t- 21. CD
I-H i — 1
OO 1 — 1
tO CO
CO >- «d-
CD CD
Z •>*""'
O Z CO
sr HH - |—
^o «
t/} O** ^X
o a:
a —i Z Z
Q£ I-H
LU
co co
/y\ T"
T3
M-
O
0)
to
3
re
0
oo
3E ^H
CO -5t-
r>-
O}
0- •
X LO
I t 1 <^-
co 1^
t-1
oo
t-H
CM
LO
CD
CD CD
OO OO CD
CD tO O
CO CD
CD
r— (
CD
to c
CO CO
3
re C -0
•i- CX
C 3
"~ ^ CO
CT> O)
C
re co c
^— C_ ••»
.^_ im. *r~*
O Qu ^
t ^ _ ^
i: c.
O> co
•r— C e
11 r— 3
CO U
•P CO O
CCO
| t 1 tf.
c -o
>> re
.U ^).C
^^
t/J C. C
,
•r- f" re
•P +-> 3
co c o
o co re
>
-p re
re A.C
CO ^) 4-^
•O r—
•— -o
^D (O GJ
-cos
•P T- O
c 'o
re co c
0- 0
•P CO 'I-
&_ -p
o re
o co -^
o -P E
re re
CO T3 X
.C CL CO
•P 3 CO
t
•p 3 re
0 0
"° '> H-T
o t_ o
a. o.
c: co
O ft) t_
reexaminatii
mbers of thi
e previous i
78-1979.
3 .£= CD
reCC4J^
t — i
CO
CD
i-H
CO
c:
11
CO
c
LU
* •
UJ
0
3
o
CO
-------
TABLE 7-11. COMPARISON OF MORTALITY IN EPICHLOROHYDRIN (ECH) ALONE,
AND COMBINED WITH ISOPROPYL ALCOHOL (IPA) EXPOSURE GROUPS
IN DEER PARK, TEXAS
Exposure group Number
Enter! ine
(1981)
ECH (alone) 308
IPA and ECHa 124
(1980)
IPA (alone) 350
IPA and ECHa 125
All cases
Observed/
Expected
38/60.89
16/23.24
24/44.30
16/22.64
All cancer
Observed/
SMR Expected SMR
62.4 11/11.79 93.3
68.9 5/4.68 107.0
54.2 5/8.21 60.9
70.7 5/4.37 114.4
Res pi ratory
cancer
Observed/
Expected
3/4.61
4/1.86
2/3.16
4/1.71
SMR
65.1
214.8
63.3
233.9
aThese represent the same cohort (except for one unidentified man) with 1 year's
additional values.
SOURCE: Enterline, 1980 and 1981.
7-28
-------
respiratory cancer SMR is much higher in the group exposed in the IPA unit (to
chemicals other than epichlorohydrin) than in the group exposed to epichloro-
hydrin alone (SMRs = 214.8 versus 63.3, respectively).
While the above data suggest that the IPA process is responsible for the
respiratory cancer increase, an additional study by Enterline of the IPA cohort
suggests a different interpretation. This is the 1980 report on the mortality
experience of a cohort of 433 men who worked in the IPA unit at Deer Park, Texas,
from its startup in 1941 to 1965 (Enterline, 1980). Table 7-11 shows the mor-
tality patterns for all causes, all cancers, and all respiratory cancers. As
can be seen, the IPA plus epichlorohydrin combined group had higher SMRs than
the IPA group alone in all three categories, with the major increase being in
respiratory cancer deaths. Also, the epichlorohydrin group alone had approxi-
mately the same respiratory cancer mortality as the IPA group alone.
The conclusion made by the CAG is that the most recent update of the
Enterline data has provided less clear evidence for the human carcinbgenicity of
epichlorohydrin. The evidence for the carcinogenicity of epichlorohydrin
includes increased respiratory cancer with increasing latent period and the
higher respiratory SMR in cancer in the combined IPA plus epichlorohydrin
group versus the IPA group alone. Further, there was elevated respiratory
cancer in both epichlorohydrin production plants (CAG, 1980). Contrary to
this evidence, there are no statistically significant increases and actually
a decreased SMR in the latest 2-year update as compared to the earlier update.
Also, the increase in respiratory cancer SMR in the combined IPA plus epichlo-
rohydrin exposure group, as compared with either the epichlorohydrin group
alone or the IPA group alone, suggests that the interaction between IPA and
epichlorohydrin exposure leads to increased respiratory cancer. Significantly,
7-29
-------
in the Deer Park, Texas epichlorohydrin alone subgroup of 350, there was no
increase in respiratory cancer versus controls (see also Enterline, 1982).
Considering, in addition, the confounding factor of smoking, the CAG is of the
opinion that these studies provide only limited evidence for the human carcino-
genicity of epichlorohydrin.
Shellenberger et al. (1979) conducted a restrospective cohort mortality
study of 533 white male full-time Dow Chemical Company employees who had poten-
tial epichlorohydrin exposure in a production area for at least 1 month between
October 1957 (the date on which commercial production of epichlorohydrin began
in the Dow Chemical Company, Texas Division) and November 1976. In all, there
were 12 deaths during this period: one cancer death from adenocarcinoma of the
stomach, one death due to metastatic malignant melanoma, five deaths due to
cardiovascular diseases, and five deaths from accidents. The two observed
cancer deaths were less than the number expected (3.50) for the entire group.
In a further breakdown, Shellenberger et al. subdivided the cohort, enumerating
the 202 persons with at least 1 year of epichlorohydrin exposure and holding at
least one job in which epichlorohydrin exposure was estimated to be >1 ppm.
Neither of the two cancers was from this group.
Although this study was negative with respect to cancer mortality, it has
drawbacks relative to carcinogenicity assessment. First, only 2 percent (12/553)
of the cohort died during the 11-year follow-up period. This would have been
only 1.3 percent had it not been for the five accidental deaths. The expected
death rate for accidents was larger than that for cancer, indicating a very
young cohort. The actual average age at the end of the follow-up period was
only 39 years, and 61.8 percent of the cohort was less than 40 years of age as
of the cutoff date. The average duration of exposure for the study cohort was
only 3 years, with 43.9 percent exposed less than 1 year and 58.7 percent
7-30
-------
exposed less than 2 years. The average interval since first exposure was only
7.7 years, with 47.7 percent having less than 7 years from first exposure to
the end of the study. This epidemiologic study on epichlorohydrin, while
negative, is inadequate for the evaluation of carcinogenicity because of its
low exposures, short exposure durations, short latent period, and very young
age of the cohort.
Tassignon et al. (1983) presented a report to the EPA detailing an historic
prospective study of workers exposed to epichlorohydrin at four European plants
that manufactured epichlorohydrin and epichlorohydrin-derived chemicals. A total
of 606 males, who had had at least 1 year of exposure to epichlorohydrin at least
10 years prior to the final date of the study on December 1978, were analyzed
for their mortality experience. The combined results showed 10 observed deaths
versus 18.4 expected (SMR = 54.3), arid four cancer deaths versus five expected
(SMR = 80). While these four cancers were all from different sites and provide
no evidence of an association between epichlorohydrin and cancer, the study by
itself lacks the power to establish the safety of epichlorohydrin. Several
limiting factors are: 1) the relatively small study size, especially the size
of the group (274) with at least 10 years of exposure; 2) the low average age
of the cohort (42 years); 3) the lack of any analysis by time since first
exposure (the authors state that they would have preferred a longer minimal
observation period but that the cohort would then have become very small); and
4) personal exposure measurements taken in 1977 and 1978 showing that exposure
levels were low (at or below 1 ppm, 8-hour time weighted average, even though
earlier exposures are stated by the authors to have been considerably higher.
Finally, there is the peculiar result that the cohort yielded zero dead
and 90 live retirees. Considering that retirees generally have about 5 percent
mortality per year, this is a highly improbable result. However, the paper
7-31
-------
does not report either the person-years of these retirees, their length of
exposure, or their average age, with the result that further analysis is
impossible. Furthermore, if the average retiree's age is 65, then the average
age of.the other 516 men in the cohort would have to be low—around 38—in
order for the average age of the total cohort to be 42 years. A more complete
report"would perhaps answer these questions. In short, this study appears to
suffer from many of the same deficiencies as the Shellenberger et al. study.
7.1.4 Quantitative Estimation
This section deals with the incremental unit risk for epichlorohydrin in
air and water, and the potency of epichlorohydrin relative to other carcinogens
that the CA6 has evaluated. The incremental unit risk estimate for an air or
water pollutant is defined as the incremental lifetime cancer risk to an indivi-
dual due to continuous exposure from birth throughout life to a concentration
of 1 ug/m3 of the agent in the air breathed, or to 1 ug/L in drinking water.
This calculation estimates in quantitative terms the impact of the agent as a
carcinogen. Unit risk estimates are used for two purposes: 1) to compare the
carcinogenic potency of several agents with each other, and 2) to give a crude
indication of the population risk that might be associated with air or water
exposure to these agents if the actual exposures were known.
7.1.4.1 Procedures for Determination of Unit Risk—The data used for the
quantitative estimate are taken from one or both of the following: 1) life-
time animal studies, and 2) human studies where excess cancer risk has been
associated with exposure to the agent. In animal studies it is assumed, unless
evidence exists to the contrary, that if a carcinogenic response occurs at the
dose levels used in the study, then responses will also occur at all lower doses,
with an incidence determined by an extrapolation model.
There is no solid scientific basis for any mathematical extrapolation model
7-32
-------
that relates carcinogen exposure to cancer risks at the extremely low concentra-
tions that must be dealt with in evaluating environmental hazards. For practical
reasons, such low levels of risk cannot be measured directly either by animal
experiments or by epidemiologic studies. We must, therefore, depend on our
current understanding of the mechanisms of carcinogenesis for guidance as to
which risk model to use. At the present time the dominant view of the carci-
nogenic process involves the concept that most agents that cause cancer also
cause irreversible damage to DNA. This position is reflected by the fact that
a very large proportion of agents that cause cancer are also mutagenic. There
is reason to expect that the quantal type of biological response, which is
characteristic of mutagenesis, is associated with a linear non-threshold dose-
response relationship. Indeed, there is substantial evidence from mutagenicity
studies with both ionizing radiation and a wide variety of chemicals that this
type of dose-response model is the appropriate one to use. This is particularly
true at the lower end of the dose-response curve; at higher doses, there can be
an upward curvature, probably reflecting the effects of multistage processes on
the mutagenic response. The linear nonthreshold dose-response relationship is
also consistent with the relatively few epidemiologic studies of cancer responses
to specific agents that contain enough information to make the evaluation possi-
ble (e.g., radiation-induced leukemia, breast and thyroid cancer, skin cancer
induced by arsenic in drinking water, liver cancer induced by aflatoxin in the
diet). There is also some evidence fronranimal experiments that is consistent
with the linear nonthreshold model (e.g., liver tumors induced in mice by
2-acetylaminofluorene in the large-scale EOgi study at the National Center for
Toxicological Research, and the initiation stage of the two-stage carcinoge-
nesis model in rat liver and mouse skin).
Because its scientific basis, although limited, is the best of any of the
7-33
-------
current mathematical extrapolation models, the linear nonthreshold model has
been adopted as the primary basis for risk extrapolation to low levels of the
dose-response relationship. The risk estimates made with this model should be
regarded as conservative, representing the most plausible upper limit for the
risk, i.e., the true risk is not likely to be higher than the estimate, but it
could be lower.
The mathematical formulation chosen to describe the linear nonthreshold
dose-response relationship at low doses is the linearized multistage model.
This model employs enough arbitrary constants to fit almost any monotonically
increasing dose-response data, and it incorporates a procedure for estimating
the largest possible linear slope (in the 95 percent confidence limit sense)
at low extrapolated doses that is consistent with the data at all dose levels
of the experiment.
7.1.4.1.1 Description of the Low-Dose Extrapolation Model. Let P(d) represent
the lifetime risk (probability) of cancer at dose d. The multistagejnodel has
the form
P(d) = 1 - exp [-(q0
qkdk)]
where
Equivalently,
q-i _> 0, i = 0, 1, 2 k
Pt(d) = 1 - exp C-(q1d + q2d2 + ... + qkdk)3
7-34
-------
where
P+(d) = P(d) - P(0)
* 1 - P(0)
is the extra risk over background rate at dose d.
The point estimate of the coefficients q-j, i =0, 1, 2, ..., k, and
consequently, the extra risk function", Pt(d), at any given dose d, is
calculated by maximizing the likelihood function of the data.
The point estimate and the 95 percent upper confidence limit of the extra
risk, Pt(d), are calculated by using the computer program GLOBAL79, developed
by Crump and Watson (1979). At low doses, upper 95 percent confidence limits
on the extra risk and lower 95 percent confidence limits on the dose producing
a given risk are determined from a 95 percent upper confidence limit, qj, on
parameter qj. Whenever m > 0, at low doses the extra risk, Pt(d), has approx-
imately the form Pt(d) = qx x d. Therefore, q£ x d is a 95 percent upper
confidence limit on the extra risk, and R/q^ is a 95 percent lower confidence
limit on the dose, producing an extra risk of R. Let L0 be the maximum value
of the log-likelihood function. The upper-limit, qj, is calculated by increas-
sing qx to a value q£ such that when the log-likelihood is remaximized subject
to this fixed value q^ for the linear coefficient, the resulting maximum
value of the log-likelihood LI satisfies the equation
2 (L0 - LI) = 2.70554
where 2.70554 is the cumulative 90 percent point of the chi-square distribution
with one degree of freedom, which corresponds to a 95 percent upper limit (one-
7-35
-------
sided). This approach of computing the upper confidence limit for the extra
risk Pt(d) is an improvement on the Crump et al. (1977) model. The upper
confidence limit for the extra risk calculated at low doses is always linear.
This is conceptually consistent with the linear nonthreshold concept discussed
earlier. The slope, qj, is taken as an upper bound of the potency of the
chemical in inducing cancer at low doses.
In fitting the dose-response model, the number of terms in the polynomial
is chosen equal to (h-1), where h is the number of dose groups in the experiment,
including the control group.
Whenever the multistage model does not fit the data sufficiently well, data
at the highest dose is deleted and the model is refit to the rest of the data.
This is continued until an acceptable fit to the data is obtained. To
determine whether or not a fit is acceptable, the chi-square statistic
X2= E
is calculated where N1 is the number of animals in the ith dose group, Xi is
the number of animals in the ith dose group with a tumor response, P1 is the
probability of a response in the itn dose group estimated by fitting the
multistage model to the data, and h is the number of remaining groups. The
fit is determined to be unacceptable whenever X2 is larger than the cumulative
99 percent point of the chi-square distribution with f degrees of freedom,
where f equals the number of dose groups minus the number of non-zero multistage
coeffi ci ents.
For cases of partial lifetime exposure where time-to-tumor or time-to-
tumor death is known, Crump and Howe (1984) have developed the multistage
7-36
-------
model to Include a time-dependent dose pattern. The form of this model is
one which is linear in dose and in which time has a power and form determined
by both the number of assumed stages and the stage affected by the carcinogen.
This model is used for the epichlorohydrin inhalation study of Laskin et al.
(1980). A best fit will be determined by the method of maximum likelihood in
the ADOLL1-83 computer (Crump and Howe, 1983).
7.1.4.1.2 Selection of Data. For some chemicals, several studies in different
animal species, strains, and sexes, each run at several doses and different
routes of exposure, are available. A choice must be made as to which of the
data sets from several studies to use in the model. It may also be appropriate
to correct for metabolism differences between species and for absorption factors
via different routes of administration. The procedures used in evaluating
these data are consistent with the approach of making a maximum-likely risk
estimate. They are listed below:
1. The tumor incidence data are separated according to organ sites or
tumor types. The set of data (i.e., dose and tumor incidence) used in the
model is the set in which the incidence is statistically significantly higher
than the control for at least one test dose level, or in which the tumor
incidence rate shows a statistically significant trend with respect to dose
level. The data set that gives the highest estimate of the lifetime carci-
nogenic risk, q-^, is selected in most cases. However, efforts are made to
exclude data sets that produce spuriously high risk estimates because of a
small number of animals. That is, if two sets of data show a similar dose-
response relationship, and one has a very small sample size, the set of data
having the larger sample size is selected for calculating the carcinogenic
potency.
2. If there are two or more data sets of comparable size that are
7-37
-------
identical with respect to species, strain, sex, and tumor sites, the geometric
mean of q^, estimated from each of these data sets, is used for risk assessment.
3. If two or more significant tumor sites are observed in the same study,
and if the data are available, the number of animals with at least one of the
specific tumor sites under consideration is used as incidence data in the model.
7.1.4.1.3 Calculation of Human Equivalent Dosages from Animal Data. Following
the suggestion of Mantel and Schneiderman (1975), it is assumed that nig/surface
area/day is an equivalent dose between species. Since, to a close approximation,
surface area is proportional to the two-thirds power of weight, as would be the
case for a perfect sphere, the exposure in nig/day per two-thirds power of the
weight is also considered to be equivalent exposure. In an animal experiment,
this equivalent dose is computed in the following manner:
Let
Le = duration of experiment
le = duration of exposure
m = average dose per day in mg during administration of the agent (i.e.,
during le)> and
W = average weight of the experimental animal
The lifetime exposure is then
IP x m
d = -e-
Le x W
2/3
7.1.4.1.3.1 Oral. Often exposures are not given in units of mg/day, and
it becomes necessary to convert the given exposures into mg/day. For example,
in most feeding studies exposure is given in terms of ppm in the diet. In these
cases, the exposure in mg/day is
7-38
-------
m = ppm x F x r
where ppm is parts per million of the carcinogenic agent in the diet or water,
F is the weight of the food or water consumed per day in kg, and r is the
absorption fraction. In the absence of any data to the contrary, r is assumed
to be equal to one. For a uniform diet, the weight of the food consumed is
proportional to the calories required, which in turn is proportional to the
surface area, or two-thirds power of the weight, so that
m <* ppm x W2/3 x r
or
m
rW2/3
ppm.
As a result, ppm in the diet is often assumed to be an equivalent exposure
between species. However, this is not justified for the present study, since
the ratio of calories to food weight is very different in the diet of man as
compared to laboratory animals, primarily due to differences in the moisture
content of the foods eaten. It is therefore necessary to use an empirically-
derived factor, f = F/W, which is the fraction of an organism's body weight
that is consumed per day as food, expressed as follows:
Species
Man
Rats
Mice
W
70
0.35
0.03
Fraction of body
weight consumed as
ffood
0.028
0.05
0.13
'water
0.029
0.078
0.17
7-39
-------
Thus, when exposure is given as a certain dietary or water concentration in
ppm, the exposure in mg/W2/3 is
m = ppm x F = ppm >cf x'W = ppm x f x W1/3
W2/3
7.1.4.1.3.2 Inhalation. When exposure is given in terms of mg/kg/day
m/Wr = s, the conversion is simply
m = s x W1/3
When exposure is via inhalation, the calculation of dose can be considered
for two cases where 1) the carcinogenic agent is either a completely water-soluble
gas or an aerosol and is absorbed proportionally to the amount of air breathed
in, and 2) where the carcinogen is a poorly water-soluble gas which reaches an
equilibrium between the air breathed and the bo.dy compartments. After equilibrium
is reached, the rate of absorption of these agents is expected to be proportional
to the metabolic rate, which is proportional to the rate of oxygen consumption,
which in turn is a function of surface area.
Case 1— Agents that are in the form of particulate matter or virtually
completely absorbed gases, such as sulfur dioxide, can reasonably be expected
to be absorbed proportionally to the breathing rate. In this case the exposure
in mg/day may be expressed as:
K
m = I x v x r
where I = inhalation rate per day in m3, v = mg/m3 of the agent in air, and
7-40
-------
r = the absorption fraction.
The inhalation rates, I, for various species can be calculated from the
observations of the Federation of American Societies for Experimental Biology
(FASEB, 1974) that 25-g mice breathe 34.5 liters/day and 113-g rats breathe 105
liters/day. For mice and rats of other weights, W (in kilograms), the surface
area proportionality can be used to find breathing rates in m3/day as follows:
For mice, I = 0.0345 (W/0.025)2/3 m3/day
For rats, I = 0.105 (W/0.113)2/3 m3/day
For humans, the value of 20 m3/day is adopted as a standard breathing rate
by the International Commission on Radiological Protection (ICRP, 1977)*.
The equivalent exposure in mg/W2/3 for these agents can be derived from
the air intake data in a way analogous to the food intake data. The empirical
factors for the air intake per kg per day, i = I/W, based upon the previously
stated relationships, are tabulated as follows:
Species
Man
Rats
Mice
W
70
0.35
0.03
i = I/W
0.29
0.64
1.3
Therefore, for particulates or completely absorbed gases, the equivalent
exposure in mg/W2/3 is
*From "Recommendation of the International Commission on Radiological Pro-
tection," page 9. The average breathing rate is 107 cm3 per 8-hour workday
and 2 x 10' cm3 in 24 hours.
7-41
-------
m
Ivr = mffir . iwl/3vr
w2/3
In the absence of experimental information or a sound theoretical argument
to the contrary, the fraction absorbed, r, is assumed to be the same for all
spacies.
Case 2—The dose in mg/day of partially soluble vapors is proportional to
the 02 consumption, which in turn is proportional to W2/3 and is also proportional
to the solubility of the gas in body fluids, which can be expressed as an absorp-
tion coefficient, r, for the gas. Therefore, expressing the 02 consumption as
02 » k W2/3, where k is a constant independent of species, it follows that:
m
= k W2/3 x v x r
or
d =
= kvr
As with Case 1, in the absence of experimental information or a sound theoretical
argument to the contrary, the absorption fraction, r, is assumed to be the same
for all species. Therefore, for these substances a certain concentration in
ppm or ug/m3 in experimental animals is equivalent to the same concentration
in humans. This is supported by .the observation that the minimum alveolar
concentration necessary to produce a given "stage" of anesthesia is similar in
man and animals (Dripps et al., 1977). When the animals are exposed via the
oral route and human exposure is via inhalation or vice versa, the assumption is
made, unless there is pharmacokinetic evidence to the contrary, that absorption
7-42
-------
is equal by either exposure route.
7.1.4.1.4 Calculation of the Unit Risk from Animal Studies. The risk
associated with d mg/kg2/3/day is obtained from GLOBAL79 and, for most cases of
interest to risk assessment, can be adequately approximated by P(d) = 1 - exp
(-qjd). A "unit risk" in units X is simply the risk corresponding to an
exposure of X = 1. To estimate this value, the number of mg/kg2/3/day corre-
sponding to one unit of X is determined and substituted into the above relation-
ship. Thus, for example, if X is in units of ug/m3 in the air, then for case
1, d = 0.29 x 701/3 x lO'3 mg/kg2/3/day, and for case 2, d = 1, when ug/m3 is the
unit used to compute parameters in animal experiments.
If exposures are given in terms of ppm in air, then the conversion factor
to mg/m3 is
1 ppm = 1.2 x molecular weight (gas) mg/m3
molecular weight (air)
Note that an equivalent method of calculating unit risk would be to use mg/kg
for the animal exposures, and then to increase the jtn polynomial coefficient
by an amount:
(Wh/Wa)J/3 j = 1, 2, .... K
and use mg/kg equivalents for the unit risk values.
7.1.4.1.5 Adjustments for Less Than Lifespan Duration of Experiment. If the
duration of experiment, Le, is less than the natural lifespan of the test
animal L, the slope qlf or more generally the exponent g(d), is increased
by multiplying by a factor (L/l_e)3. We assume that if the average dose d is
continued, the age-specific rate of cancer will continue to increase as a con-
7-43
-------
stant function of the background rate. The age-specific rates for humans
increase at least by the second power of the age and often by a considerably
higher power, as demonstrated by Doll (1971). Thus, it is expected that the
cumulative tumor rate would increase by at least the third power of age. Using
this fact, it is assumed that the slope q£, or more generally the exponent
g(d), would also increase by at least .the third power of age. As a result, if
the slope qj [or g(d)] is calculated at age Le, it is expected that if the
experiment had been continued for the full lifespan, L, at the given average
A
exposure, the slope q^ [or g(d)] would have been increased by at least (L/Le) .
This adjustment is conceptually consistent with the proportional hazard
model proposed by Cox (1972) and the time-to-tumor model considered .by Crump
(1979), where the probabiity of cancer by age t and at dose d is given by
P(d,t) = 1 - expC-f(t) x g(d)].
It is also consistent with the partial lifetime exposure extension of the
multistage model developed by Crump and Howe (1984).
7.1.4.1.6 Interpretation of Quantitative Estimates. For several reasons, the
unit risk estimate based on animal bioassays is only an approximate indica-
tion of the absolute risk in populations exposed to known carcinogen concen-
trations. First, there are important species differences in uptake, metabo-
lism, and organ distribution of carcinogens, as well as species differences
in target site susceptibility, immunological responses, hormone function,
dietary factors, and disease. Second, the concept of equivalent doses for
humans compared to animals on a mg/surface area basis is virtually without
experimental verification regarding carcinogenic response. Finally, human
populations are variable with respect to genetic constitution and diet, living
7-44
-------
environment, activity patterns, and other cultural factors.
The unit risk estimate can give a rough indication of the relative potency
of a given agent as compared with other carcinogens. The comparative potency
of different agents is more reliable when the comparison is based on studies
in the same test species, strain, and sex, and by the same route of exposure,
preferably inhalation.
The quantitative aspect of carcinogen risk assessment is included here
because it may be of use in the regulatory decision-making process, e.g.,
setting regulatory priorities, evaluating the adequacy of technology-based
controls, etc. However, it should be recognized that the estimation of cancer
risks to humans at low levels of exposure is uncertain. Because of the limited
data available from animal bioassays, especially at the high dose levels re-
quired for testing, almost nothing is known about the true shape of the dose-
response curve at low environmental levels. At best, the linear extrapolation
model used here provides a rough.but plausible estimate of the upper limit of
risk; i.e., it is not likely that the true risk would be much more than the
estimated risk, but it could be considerably lower. The risk estimates pre-
sented in subsequent sections should not be regarded as accurate representations
of the true cancer risk even when the exposures are accurately defined. The
estimates presented may, however, be factored into regulatory decisions to the
extent that the concept of upper risk limits is found to be useful.
7.1.4.1.7 Alternative Methodological Approaches. The methods used by the CA6
for quantitative assessment are consistently conservative, i.e., tending toward
high estimates of risk. The most important part of the methodology contribu-
ting to this conservatism is the CAG's use of the linear nonthreshold extrapo-
lation model. There are a variety of other extrapolation models that could be
used, all of which would give lower risk estimates. These alternative models,
7-45
-------
the one-hit, log-probit, and Weibull models, have not been used in the
following analysis, but are included for comparison in Appendix E. Another
alternative method involves basing extrapolations on animal bioassay data, using
either the most sensitive responses or averages of the responses of all ade-
quately tested bioassay animals.
Extrapolations from animals to humans can also be done on the basis of
relative weights rather than surface areas. The latter approach, used here,
has more basis in human pharmacological responses; however, since it is not yet
clear which of the two approaches is more appropriate for carcinogens, it seems
appropriate to use the method in most general use, which is also the more con-
servative method. In the case of epichlorohydrin drinking water studies, the
use of extrapolations based on surface area rather than weight increases the
unit risk estimates by a factor of 5.8.
7.1.4.1.8. Model for Estimation of Unit Risk Based on Human Data. If human epi-
demiologic studies and sufficiently valid exposure information are available for
a compound, the CA6 always makes use of these data in its analyses. If these
studies show a carcinogenic effect, the data are analyzed to give an estimate
of the linear dependence of cancer rates on lifetime average dose, which is equi-
valent to the factor 64. If the epidemiologic data show no carcinogenic effects
when positive animal evidence is available, then it is assumed that a risk does
exist, but that it is smaller than could have been observed in the epidemiologic
study. An upper limit to the cancer incidence is then calculated, assuming
hypothetically that the true incidence is just below the level of detection in
the cohort studied. Whenever possible, human data are used in preference to
animal bioassay data.
s.
Very little information exists that can be utilized to extrapolate from
high-exposure occupational studies to low environmental levels. However, if a
7-46
-------
number of simplifying assumptions are made, it is possible to construct a crude
dose-response model whose parameters can be estimated using vital statistics,
epidemiologic studies, and estimates of worker exposures.
In human studies, the response is measured in terms of the relative risk
of the exposed cohort of individuals as compared to the control group. The
mathematical model employed assumes that for low exposures the lifetime proba-
bility of death from lung cancer (or any cancer), P0, may be represented by
the linear equation
PQ = A + BHx
where A is the lifetime probability in the absence of the agent, and x is the
average lifetime exposure to environmental levels in units such as ppm. The
factor BH is the increased probability of cancer associated with each unit in-
crease of the agent in air.
If we make the assumption that R, the relative risk of lung cancer for
exposed workers in comparison to the general population, is independent of the
length or age of exposure but depends only upon the average lifetime exposure,
it follows that
A + B
+ x2)
or
RPo = A + BH (xi + x2)
7-47
-------
where KI s lifetfme average daily exposure to the agent for the general
population, X£ = lifetime average daily exposure to the agent in the occu-
pational setting, and P0 = lifetime probability of dying of cancer with no
or negligible epichlorohydrin exposure.
Substituting PQ = A + BH xi and rearranging gives
BH = PO (R - l)/*2
To use this model, estimates of R and X2 must be obtained from the epidemio-
logic studies. The value PQ is derived by means of life table methodology
from the age-cause-specific combined death rates for males found in the 1976
U.S. Vital Statistics tables. For lung cancer, the estimate of PQ is 0.036.
This methodology is used in the following section, which deals with unit risk
estimates based on human studies.
7.1.4.2 Calculation of Quantitative Estimates—
7.1.4.2.1 Unit Risk Estimates Based on Human Studies. In making a risk estimate
on the basis of data from the study of Shell Oil epichlorohydrin workers, the
confounding effect of epichlorohydrin and IPA exposures cannot be ignored.
Neither can the fact that the SMR increase was not statistically significant.
Because this study was inconclusive and not positive, it is possible only to
calculate an upper bound to the risk based on the sufficient evidence of carci-
nogenicity in animals.
The basis for calculating a risk estimate is the one-sided upper 95 per-
cent confidence limit of the SMR for respiratory cancer (International Classi-
fication of Diseases 160-164). Enterline's corrected cause of death classifi-
cation (Enterline, 1981, Table 7), which attributes two of the 10 lung cancer
deaths to other cancers, was also used. In addition, all eight of the remain-
7-48
-------
ing lung cancer deaths occurred after a 15-year latency period. For lung
cancer, the corrected SMR increases from 0 for < 15 years to 122.5 for _> 15
years (8 observed vs. 6.53 expected deaths). The associated 95 percent confi-
dence limit for these eight observed deaths is 14.4.* The corresponding SMR is
100 x 14.4 + 6.53 = 221. Thus, 221 was chosen as the upper 95 percent limit.
The average age of the cohort at the time of this follow-up was 50 years.
The years of exposure were not given for the whole cohort, but Enterline's
report gave the duration of exposure prior to January 1, 1966, and the dates of
death for the eight who died from respiratory cancer. For these eight, the
average length of time from beginning of exposure to death was about 19.7 years,
If the known time of non-exposure between the beginning of exposure and
January 1966 is subtracted, the average duration of exposure is 13.4 years.
Since six of the eight had retired or left the employment of Shell Oil prior
to the time of death, the actual years of exposure are fewer than 13.4.
No exposure data are given for the Shell Oil study other than to separate
the workers into two exposure groups of (1) light to nil, and (2) moderate to
heavy exposure. Lung cancer deaths occurred in both groups, with the heavier
exposure group having a higher SMR. An exposure level of 5 ppm was chosen as
an average exposure based on the following considerations: Exposure must have
been less than 20 ppm, since workers reported extreme discomfort from only one
hour of exposure to 20 ppm (NIOSH, 1976a, 1978). Exposure must have been more
than 1 ppm, since recent plant improvements in epichlorohydrin manufacturing
facilities had reduced exposures to 1 ppm or less (NIOSH, 1978). Since half of
the cohort of workers had "moderate to heavy" exposure, according to Shell Oil,
it was reasoned that these workers were probably exposed to more than the
*If the observed eight deaths were from a Poisson distribution with 14.4
expected deaths, the probability of observing eight or fewer deaths is equal
to 0.05. ^
7-49
-------
current Threshold Limit Value (TLV), on the basis that a company would not
label as "heavy exposure" values that were at or below the TLV. The TLV
current for many years was 5 ppm, so it was chosen as a reasonable average for
a cohort of workers divided approximately half and half into light and heavy
exposure.
The exposure of these workers to epichlorohydrin averaged over a lifetime
is given by
Exposure = 5 ppm x _§. hrs x 240 days x 1-Ld yrs
24 365 50
Exposure =0.29 ppm
The probability of dying from respiratory cancer from a lifetime exposure
to 1 ppm epichlorohydrin/m3 air is given by
- p°
where PQ, the background lifetime probability of dying from respiratory cancer
in the United States, is 0.036, R is the respiratory cancer relative risk of
the workers, Xi is exposure at 1 ppm, and K£ is the exposure experienced by
the workers. Substituting the appropriate numbers, we get
B - 0.036 x (2.21 - 1) x 1 ppm = Q.15
H 0.29 ppm
Thus, the upper 95% limit of the SMR for lung cancer based on the observed 8
7-50
-------
deaths and the expected 6.53 deaths yields a unit risk of 0.15. To convert ppm
to ug/m3, the formula is
ug/m3 =
10"3
1.2 (m.w. chemical)/(m.w. air)
ug/m"
10
-3
1.2 (92.
1 ug/m3 epichlorohydrin = 2.59 x 10~4
ppm
Thus, the upper limit of risk of death from lung cancer as a result of breathing
1 ug/m3 epichlorohydrin is
P = 2.59 x 10-4 x o.l5 = 3.9 x 10-5 (Ug/m3)-l
These are considered to be upper-bound risk estimates, since they are based
on a linear extrapolation to low doses. The lower bound of risk approaches zero
in view of the uncertainties in both the qualitative evaluation and the quanti-
tative extrapolation process. The plausibility of the upper bound is enhanced
when there is clear evidence of mutagenicity, which is the case for epichloro-
hydrin.
7.1.4.2.2 Unit Risk Estimates Based on Animal Studies. Two positive animal
oral studies are available for a quantitative risk assessment, one gavage study
and one drinking water study. The van Esch (1982) gavage study showed increased
papillomas and carcinomas of the forestomach, as did the drinking water study.
However, the gavage study was not used in the present quantitative estimation
for several reasons: (1) the report has not yet been published; (2) the diet
7-51
-------
used for the first 56 weeks of this study caused the death of animals due to
"hairballs" in the stomach. The interaction of diet and epichlorohydrin with
respect to stomach lesions could not be dismissed; and (3) the gavage route is
not as appropriate for animal-to-man extrapolation as the drinking water route,
especially for a contact carcinogen such as epichlorohydrin.
Because of the limitations discussed previously, the bioassay of epichlor-
ohydrin in the drinking water of male Wistar rats (Konishi et al., 1980; Kawa-
bata, 1981) may be considered a pilot study. The results of this study never-
theless are chosen as presenting adequate evidence of carcinogenicity for
calculating a unit risk by the drinking water route, mainly because it is the
only drinking water study that has been done to date. The pertinent cancer
data, shown in Table 7-4, present a doseresponse trend for both papillomas and
carcinomas of the forestomach. Tumor response of papillomas and carcinomas
combined on terminal sacrifice at 81 weeks was 0/10, 0/9, 2/10, and 9/12 for
the control, 375-ppm, 750-ppm, and 1,500-ppm dose groups, respectively. This
increase was statistically significant at the P < 0.001 level for high-dose
animals vs. controls, and is considered biologically significant, especially in
view of the early terminal sacrifice. ;
Since epichlorohydrin is a direct-acting alkylating agent, this response
to the forestomach can be considered a local reaction. As such, the effect is
dependent not on the dose per body weight, but on the dose per square inch of
forestomach area. Unfortunately, the relative surface areas of rats and humans
are not comparable. Furthermore, dose to the target organ is also dependent
on comparative residence time, on which no information is as yet available.
The estimates of unit risk based on this study must be viewed in the light of
these uncertainties.
The method chosen by the CAG to estimate human exposure to epichlorohydrin
7-52
-------
is that of determining the equivalent concentration a human must ingest in
drinking water, allowing for the difference in ratio of water ingested to body
weight. In formula terms this is
fwater, rat x water cone., rat = ^water, human x water cone., human
The use of this formula results in doses that are equal on a dose/body weight
basis, as opposed to dose/(body weight)2/3 as discussed in the methodology
section. The water-to-body-weight ratios were given earlier as fwater = 0.078
for the rat and fwater = 0.029 for the human. Thus, the rat-to-human dose
ratio is 0.078/0.029 = 2.7. For the rat experiment, the drinking water concen-
trations for the three dose groups were 375, 750, and 1,500 ppm. (For amount
ingested, the authors state that the total doses were 5.0 g, 8.9 g, and 15.1 g,
respectively; this would lead to slightly different estimates.) However, since
all of the dose groups had their treatments interrupted for varying times, this
figure must be adjusted to give equivalent concentrations on a continuous basis.
As estimated in Figure 7-4, the numbers of weeks of treatment for the low- to
high-dose groups were 76, 79, and 75.5, respectively. Multiplying these by
the low- to high-dose water concentrations and dividing by 81 weeks yields
352, 731, and 1,398 ppm, respectively. Adjusting the treatment levels in the
bioassay, 375 ppm in the rat would be equivalent to 352 x 2.7 = 950 ppm in
the human. The other equivalent human doses are 1,974 ppm and 3,775 ppm for the
middle- and high-dose groups, respectively. Assuming that a 70-kg human drinks
2 liters of water/day, the equivalent human dosages are 1.9, 3.9, and 7.6 g.
Dividing by 70 kg gives doses of 27.1, 55.7, and 108.6 mg/kg/day.
The above responses on papillomas and carcinomas of the forestomach were
fitted using the linearized multistage model with the equivalent human dosages.
7-53
-------
The upper-limit maximum likelihood estimate of the linear component is
= 4.7 x ID'3 (mg/kg/day)-]
Because the experiment was conducted for only 81 weeks, the adjustment
factor for the less-than-natural-lifetime experiment is (104/81)3 = 2.1, as
discussed in a preceding section. Thus, the final value of the linear
component is
4.7 x
x 2.1 = 9.9 x
v-3
(mg/kg/day)"
In order to estimate a unit risk for 1 ug/L of water, it was assumed
that the average 70-kg human drinks 2 liters of water per day. Since 2 liters
weigh approximately 2 kg, it was estimated that 1 ug/L water corresponds to
2 ug/day. Dividing by 70 kg gives 2.9 x 10-2 Ug/kg/day or 2.9 x 10'5 mg/kg/
body weight/day. The upper-limit unit risk corresponding to 1 ug/L epichloro-
hydrin concentration in water is then
p = 1 _ exp (-9.9 x ID'3 x 2.86 x 10'5) = 2.8 x 10'7
For comparison purposes only, the following paragraph relating the animal
drinking water study and inhalation study risk estimates is included.
The dose rate d(mg/kg/day) resulting from breathing 20 m3/day of air
containing a concentration of 1 ug/m3 can be determined if it is assumed that
100 percent of the inhaled epichlorohydrin is absorbed into the body. With
this assumption the dose rate is
7-54
-------
1 ug/m3 = 1 ug/m3 x 20 m3/day x lO"3 mg/ug x 1/70 kg
= 2.86 x 10-4 mg/kg/day
The upper-limit estimate of the unit risk, P, of 1 ug/m3 can be found
using this value of d and the value of q-^ estimated above as follows:
P » 1 - exp (-9.9 x 10-3 x 2.86 x 10"4) = 2.8 x lO'6
This is about twice that of the animal inhalation study (see below) and
about, 7 percent as large as the upper limit for the human data.
In the Laskin et al. (1980) inhalation study, 15 of 140 rats exposed to
short-term relatively intense exposure (100 ppm for 30 exposures) developed
squamous cell carcinomas of the nasal cavity. Two others developed nasal
papillomas. The same study, however, had lifetime exposure groups at the lower
concentrations of 10 ppm and 30 ppm with squamous cell carcinoma incidences of
0/100 and 1/100, respectively, and no papillomas. These results are summarized
in Table 7-12. The authors attempt to explain the result that dose rate rather
than total dose is related to cancer incidence. This explanation is that the
relationship
dtn = constant
holds, where d is the dose rate with chronic lifetime exposure, t is the time
required to reach a given level of tumor incidence, and n is a power of t,
usually between 2 and 3. Thus, with increasing dose rate, not only does the
incidence increase, but the time-to-tumor for a given incidence decreases. For
dose rates that yield similar incidences, the time-to-tumor is greater in the
7-55
-------
TABLE 7-12. ESTIMATION OF 95 PERCENT UPPER-LIMIT UNIT RISK
ESTIMATES FOR NASAL PAPILLOMAS AND SQUAMOUS CELL CARCINOMAS
USING THE MULTISTAGE MODEL WITH AND WITHOUT TIME DEPENDENCE
Model
95 percent
upper-limit unit
risk estimate for
lifetime exposure to
1 ppm 1 ug/n)3
Comments
Without time dependence
Eliminating highest
dose-response group
Including highest
dose-response group
as if 100 ppm given
5 days/week the
entire study
With time dependence
Three-stage,
first stage active
4.8 x 10-3 1.2 x 10-6
4.7 x lO-3 1.2 x 10-6
4.6 x lO-2 1.2 x 10-5
Considered to be the
best estimate of
effects of chronic
low exposures.
Best estimate of
linear term is zero;
terms of 2nd and 3rd
degree are positive.
The higher estimates
show the effects of
high exposure early
in life.
SOURCE: Laskin et al., 1980.
7-56
-------
lower dose rate group. In such cases, even though the incidence may not
actually decrease, it may appear smaller in the lower dose rate groups because
death may occur before tumorigenesis.
While the above explanation does have some experimental basis, the effects
of long-term low-dose exposure are still of concern. Therefore, the low doses
of 10 ppm and 30 ppm were chosen as being more representative of environmental
exposures. The control groups of 100 (sham) for life and 50 untreated were conn
bined; no squamous cell carcinomas were seen. Since epichlorohydrin was admini-
stered as a partially soluble vapor, the concentrations in ppm in experimental
animals are considered equivalent to the same concentrations in humans. Thus,
no corrections have been made for.weight differences between species. The 95
percent upper-limit estimate for slope based on the two long-term exposures to
10 ppm and 30 ppm (1.79 ppm and 5.36 ppm continuous dose) is
q* = 4.8 x 10-3 (ppm)-1
h
For unit risk in terms of ug/m3, we make the transformation
3 _ 10"3 ppm 10"3 ppm
1 u9/m ~ 1.2 (m.w. chemicals)/(m.w. air) = 1.2 (92.5)7(28.8)
= 2.59 x 10-4
ppm
thus, in terms of ug/m3
q* = 4.8 x 10-3 (ppm)-l x 2.59 x 1Q-4 ppm = 1.2 x 10-6 (Ug/m3)-l
1 ug/m
3
7-57
-------
For comparison purposes, the above data were extrapolated to low-dose
estimates using the nasal papilloma and squamous cell carcinoma responses to
the 30-exposure 100-ppm dose as though the dose had been given for the animals'
* ^
lifetimes. The results produced an upper-limit estimate of q^ = 4.7 x 10"°,
but with higher-order terms of the second and third degree. These are shown in
Table 7-12. Also shown in Table 7-12 are the results based on the partial life-
time exposure model, computed with the ADOLL1-83 computer program (Crump and
Howe, 1983, 1984). The partial lifetime exposure model is one which is linear in
dose but one in which the time-to-tumor (or time-to-tumor death) variable has
an exponent related to both the number of stages and the stage affected by the
carcinogen (single active stage). For the epichlorohydrin data, all animals
with nasal papillomas (2) or carcinomas (15) were considered to have died as a
result of these tumors. This approach is conceptually consistent with that of
the above analysis of the quantal data which counted the number of animals with
tumors. It is also consistent with the study design, which did not include
scheduled sacrifice, so that all animals sacrificed were killed in moribund
condition. Finally, the data often note times at which the tumors were first
observed; these times were close to the actual death times. The result of
ascertaining that all animals with nasal carcinomas died from them is that the
time-to-tumor death variable can present a good time-response function. Based
on the method of maximum likelihood, the best-fitting model is one with three
stages, with the first stage active. Satisfactory fits were also obtained with
models of stages 4, 5, and 6, all with stage 1 active. The results show that
accounting for partial lifetime exposure leads to a 95 percent upper-limit
estimate of the lifetime unit risk of 4.6.x 10-2 (ppm)-1, which is almost 10
times that derived when partial exposure is not accounted for.
The results of the above analysis differ from those of most other EPA
7-58
-------
analyses in that they show how the response is related to dose rate in a non-
linear way. While the design of the study could not detect whether epichloro-
hydrin given at high doses later in life would be a late-stage carcinogen, the
model was very definitely able to detect the danger of high doses given early.
The implication for humans is that high exposures to epichlorohydrin early in
life for a short time can be more hazardous than larger cumulative doses given
over a longer period. Whether or not that same high dose given later in life
would be as hazardous, the study cannot answer. The fact that only an early-
stage active carcinogen model fits the data, however, is consistent with the
observed response.
7.1.4.2.3 Summary of Unit Risk Estimates. Three upper-limit risk estimates
were calculated for epichlorohydrin. All three have more uncertainty than
those of other suspect carcinogens that the CA6 has evaluated. Epichlorohydrin
is, however, among the weakest of these in terms of unit risks.
Quantitative unit risks were calculated for epichlorohydrin via both the
drinking water and the inhalation routes. The study of male Wistar rats (Konishi
et al., 1980) was used to estimate a unit risk, 2.8 x 10-7, for a lifetime
exposure to drinking water containing 1 ug/L of epichlorohydrin. This esti-
mate has the uncertainty of estimated exposure to the target organ, the small
number of animals used in the study, and the high mortality that was observed.
Quantitative risk assessments were also calculated using both the rat
inhalation study (Laskin et al., 1980) and the study of Shell Oil workers
(Enterline, 1981). The two 95 percent upper-limit unit risk estimates are not
very close: the upper limit for animal data is 1.2 x 10"^ (ug/m-*)-!; the upper-
limit for human data is 3.9 x 10~5 (ug/m^)-!. in units of risks per ppm, the
upper-limit estimates range from 4.8 x 10"3 to 4.6 x 10~2 (ppm)-l for animals and
0.153 (ppm)-l for humans. In view of the weakness of both inhalation data bases,
7-59
-------
these inhalation unit risk estimates must be taken with caution. Animal expo-
sures in the Laskin et al. (1980) inhalation study, on a continuous daily
equivalent basis, were 0.86 ppm, 2.98 ppm, and 1.03 ppm for actual daily
exposures of 10 ppm, 30 ppm, and 100 ppm, respectively. The high concentration,
100 ppm, was given for only 30 exposures, whereas the lower two concentrations
were given for the animals' lifetimes; however, only this high-concentration
group developed a significant increase in cancers. Because of the dose-rate
effect and/or the early active stage effect, each estimate based on the animal
inhalation study should extrapolate to the specific human exposure condition.
1 ' 'ff Q 1
The estimate q^ = 4.8 x 10""3 (ppm)"1 should be used for long-term low exposures;
if O
the estimate q^ = 4.6 x 10"^ should be used under conditions of short-term high
exposure. For human-to-human extrapolations, the 95 percent upper-limit unit
risk estimate based on human studies is also an upper bound on non-statistically-
significant increases in cancer mortality.
7.1.4.3 Relative Potency—One of the uses of unit risk is to compare the
potency of carcinogens. To estimate the relative potency, the unit risk slope
factor is multiplied by the molecular weights, and the resulting number is ex-
pressed in terms of (mmol/kg/day)"1. This is called the relative potency index.
Figure 7-7 is a histogram representing the frequency distribution of
potency indices of 53 suspect carcinogens evaluated by the CA6. The actual
data summarized by the histogram are presented in Table 7-13. Where positive
human data are available for a compound, they have been used to calculate the
index. Where no human data are available, animal oral studies and animal
inhalation studies have been used, in that order. In the present case, the
human data are only suggestive; therefore, animal oral studies were used.
The potency index for epichlorohydrin based on the drinking water study
(Konishi et al., 1980; Kawabata, 1981) is 0.92 (mmol/kg/day)"1. This is derived
7-60
-------
20-
18-
16-
14-
O
Z
LU
=)
o
UJ
12-
10-
6-
4-
2-
0-
4th
QUARTILE
1
1x
3rd
QUARTILE
2nd
QUARTILE
1st
QUARTILE
I I
+1 +2 +3
I0 4x10 2x10
-I
0
.2 34 5
LOG OF POTENCY INDEX
8
Figure 7-7. Histogram representing frequency distribution of the potency indices of 53 suspect
carcinogens evaluated by the Carcinogen Assessment Group.
7-61
-------
Q.
O
on
CD
\—
UJ
UJ
CO
CO
zs
UJ
g
•z,
I-H
o
s
o
UJ
a:
H-
co
.s
l_. .
5
— i
UJ UJ
C3
CO CD
CJ 0
»-4 C£.
jj^l f^
"1
CO SE
i^ i •
o o
**— "O
O 3 — •
4J OX
C- M- i-H<0
OJ c en *o
•o en o c
c_ re r-~ »i—
Q £ « *
C O)
CU *T3
O •!-
Q.
t_
re
r— +J
3 -C
o en
Q) T-
i— 0)
0 5
SE
^ [
1
re
QJ "O
o en
CO -^.
£
en c re
CO •!—
••" <-2 5^
d.*o O- 0)
3
O -r-
0) >
_r oj > re tl
C- i— T3
«< «i «=c
o
i-H
1
X
CD
in
CD
P-.
CM
CD
T-l
X
CD
i— 1
i— 1
O)
TD
E
0
r—
JT
>
^
5
CO
CO
CD
X
CM
OO
CD
»-H
c5
r-H
i-H
h- *
CO
CVJ
OO
CO
1
0
^«
p—
0
c
, E
c t- -o
w oj re
CO CO O
i-H
"o
X
CVJ
co
CO
LO
i-H
I-H
1
0
^
CO
rH
CO
CM
• co
i — i
LO
CO
CM
1
ID
LO
CU
2
fO
cr
-------
cu
it- T3
O 3 '--
+•> OX
t- •!— I-H CO
co c cn-o I-H o t— i i— i
•o en o c + + +
£- ro i— -r-
O E
OX • i-H
C CO 0 O + 0
cy -a o o o o
O T- X X X X
Q. r-. co co oo
J_
ro
r— 4-i CTl r-- CTl •*
o en oo ID r-~ co
CO -I— CTl CO ID OO
1 — CO CVI i-H i-H
o 5
2^
i-H
1 CVI CVI
•— » CVJ 1 1
>5 100
ro O i— I t-l
CO "O r-H X X
Q.-^ X CVJ O CO
o en CTI
— 1 ro i —
CO CO rO
<: o E to _i _j _i
1— c T-
r— CO C
CO -0 f.
> -r-
co >
— 1 CU «/)
c
O Er
3
3T
t- CVJ r-H LO LO
CO 1 1 1 1
.Q ID CVI •* O
E 0 P-. CO 0
3 1 1 1 1
"so. r^- r*- CTI CTI
CD ^D r** r*^
tO r-H
•=f CO
o c:
ro
4_> CU
CU c
O 0}
CU t_ jr
«n c: o 4J
CO (O O £_ '
-C co jr O 4_> t_ i —
CO I- CO 4-> -C
O O CO O
to -a r— t- 4-> •!—
•o cu .E: o i t-
C +•> O i— CVI 4->
3 ro T- JT " 1
o c -a o cvi cvi
Q. •!— 1 « « •>
§£- CVI X •— 1 r-H
o * cu ** **
O I— i-H JZ i-H r-H
c~
o
r-l
O
O
0?
«*
CTl
I— 1
I— 1
CVI
1
o
I-H
X
CO
CVJ
to
CO
1
ID
ID
p*.^
ID
£
o
E
•P~
§,
t- I—
J= O
O Q
7-63
CO
CVI
&
X
r-l
CO
in
cvj
CTl
(D
•
r-H
CO
CVJ
to
i-H
1
CTl
r_]
CTl
OJ
<=
•5
IM
C
CO
o
o
_C^
0
•r—
O
r-H
r-H
fe
X
r-l
r--
CTI
I-H
1
O
i-H
X
^»
•
i-H
CO
1
^>
1
CO
in
i —
CO
c
cu
>^
c—
j \
CO
o
o
o
*r~
•o
1
I-H
t,
i-H
^
•ti-
cs
X
r-H
CTl
0
oo
CO
^~
•
o
CO
CO
CVI
to
-
i-H
1
1^ —
LO
1
o
ID
c
•r—
!_
•o
"oi
•r~
o
cvi
t-H
O
T"H
X
ID
CVI
00
r-H
I-H
00
•
O
CO
CVJ
to
CVI
1
i-H
1
i-H
cvi
I-H
cu
c
CO
3
"o
4->
O
£_
•r—
C
•r-
•o
^~
A
CVI
CVI
CVJ
0
X
1— 1
0
I-H
iji^^
f*^
«
o
CO
CVI
to
(^
1
ID
ID
cvj
CVJ
I-H
CO
C
N
ro
T3
r—
C
CO
JC
Q.
Q
O
r-H
1
O
T— <
X
CTl
IO
CVI
CTl
•
CO
o
r-H
X
CTl
CTl
CO
CVI
to
oo
1
CTl
CO
I
OJ
o
£_
0
0
1
cvi
to
CO
ID
ID
0
1 — 1
X
1— 1
I-H
.1—1
^_^
1 — 1
o
o
CO
CTl
^
to
to
i-H
1
oo
00
CVI
LO
t
CO
JC
CO
,_
^_^
f^
OJ
0
0
-£T
o
t/)
CO
^— ^
CO
en
fO
o
01
c:
•r—
0
1 —
O
CO
c:
O
.J-J
CO
3
4->
C
O
o
9
cu
o
CO
-o
^~
CO
CO
4-J
ro
cr
0)
"S
c
II
(— 1
• »1
I
CO
"O
^
CO
"S
,
E
•r—
1
II
\
• •*
CO
o
c
CO
»!—
CO
t ^
c
CO
o
•r—
•4-
3
to
II
to
to
-------
^w**
•g
3
C
•r-
•£;
O
O
c
CO
T"H
I
UJ
_J
CO
*""
CO
tj-ir, *^3
03 *—
•«•> OX
C_ •!- t-ICO
co c cn-a
-a cn o c:
t_ (O 1 •!-
o E — •
o" x
C CO
COTS
I"
C_
re
r- 4->
3 SZ.
o cn
i— CO
£~
T— 1
1
(O
co -o
O cn
r— ^
CO — .
cn
• ^w^
cn c re
CO -r-
•r- O £-
O.-O Di CO
3 CO
—1 CO in
c
*0 E
3:
C-
CO
p:
3
CO
•f.
O
in
T3
C
3
O
a.
e
o
o
+
CO
0
^
f—
£j
CM
,_,
O
X
to
T-l
5
^^
rH
to
CM
I-H
CM
_J
CO
I
r-l
CM
1
LO
f-
CO
•a
•r—
X
O
CO
c
CO
>,
s
UJ
CO CO
CO CM
+ +
^J IJ
X X
rH LO
CO •ci'
CO *^1~
r~» co
CO CM
f- 1 —
CO VO
m c
CO r-l
CO CO
CM CM
CO CO
CO r-l
1 1
«=r i~-
1 1
IO CO
,_{
CO
cr
CO
N
c
CO
I- 0
O £-
r- O
O J=
re o
+j re
O. X
CO CO
rH
rH
o
3
rH
10
CM
CM
1
o
rH
X
LO
f^
•
p^
CO
—1
CO
CO
1
CO
CO
c:
CO
• r-
•a
re
•y
JO
0
o
IE
o
re
X
CO
31
CO CO CO CO
CO CO CM CM
O CD O O
X X X X
rH co LO -st-
cn cn cn cn
o o o o
CM CM CM CM
10 CM
• « CO CO
•*«-!••
I— 1 rH rH
CO CO
CM CO CM
CO _1 _J
i — cn
1 1 1
<=)• LO cn
CO CO CO
1 1 1
cn cn co
rH rH LO
CO CO
CO
re
X
CO CO
r** «rj
o re
i— t_ t- t_
o cn co £- co
O i — O E O
o re in o m
c_ o -r- in -t—
O •!- •!-
r— c re re
J= jc J= re E
O O Q.4J E
re co i — co re
x +J re ja cn
CO
to
to
o
X
CM
rH
CO
CO
O
rH
X
CM
•
to
CQ
CM
CO
CO
to
*^(-
I
LO
to
^~
co
c
X
0
•r—
-o
0
NJ
C
CO
•I—
TJ
O
£_
0
r-
O
re
X
CO
rH
CM
1
0
X
LO
cn
CO
•=*•
o
rH
X
CO
•
to
CO
_l
CM
1
cn
o
i
LO
f^
0)
-a
•i —
r
0
0
CO
• c
CO
1 —
>^
c*
t s
£
CM
rH
0
X
t-.
l~~
CO
LO
, t
3
LO
t— 1
•
rH
CM
CO
|
0
CM
0
|
O
«^f-
^~
^-
f—
CO
vx
O
CO t- C_
in T- +J 4->
CO C •!- T-
£= i— C C
•r- >,! — i —
E -^ >> >.
re +J J= +->
in co +J 3
0 E CO J2
£- -I— •!— «r-
+j a a o
•T—
^— •*
CO
cn
re
a.
cn
c
'£
o
|H
o
CO
o
•a
CO
3
•r"
4-S
0
U
•
s
c
CO
-o
'>
CO
CO
re
3
cr
CO
•o
ra
c
t — i
ii
(•--I
• f.
CO
o
c
CO
•r—
'CO
-a
ai
^
E
•r-
_J
II
_J
.«
CO
O
CO
-a
>
CO
J ^
£Z
CO
o
•r—
4—
(f—
3
CO
II
CO
re
7-64
-------
73
CO
3
•r—
"c
o
0
*
CO
T— 1
1
f-
UJ
_J
CO
£5
1
co
«t- 73
. 4J OX
£- •!— t— leu
CO C 0)73
73 0) O C
L_ (Q f«> .(— .
0 E —
C CO
CU 73
"O T-
CL.
^_
re
r— 4->
^y c—
O O)
CU •!-
i r— CU
o 3
r-H
1
1?
CU 73
O Ol
* i— v^
co -~^.
O)
^^,
01 c re
CO -r-
•i— O t-
0-73 C£. CO
3 CO cac 4->
O OO t-H .,-
t_ re s-
CB jQ O
c/l
re i —
cu ro
0 E
C *r—
i — CO C
CO 73 cC
> -r-
cu >
1 CO OO
**— ro
O E
-f-
t_
CO
CO
•a:
CJ
Compounds
CM^-^-O
CM CO •*
+ -f + 0
O 0 O O
X X X X
CM «* CO i-H
CM r-l t-H
O 1--. OO CO
O r-H O CD
t— 1 i — 1 t-H t — 1
CO
o
T-H
„ 10 X
CO CD "CM
r-H • CM CD
• CM CD •
CM OO OO «dr
CO CO CO CO
CM CM CM CM
CO CO CO CO
t— t i— i (—1 i i
CM CD in CO
in co co CD
in r-^ CD co
iiii
O CD «=J- IO
co in oo co
CD r~ 10
CO
N-nitrosopyrrolidine
N-nitroso-N-ethyl urea
N-nitroso-N-methylurec
N-nitroso-diphenylamir
CO
CO
0
^H
X
t— 1
•=3-
CM
CO
CO
•*
CO
CM
CO
CO
1
10
CO
CO
CO
T— t
\
OO
CO
CJ
Q.
,
i-H
0
O
i-H
X
<3-
•
I —
CD
i-H
CM
1
O
T— 1
X
CD
CD
t— 1
CO
CM
CO
1 '
CM
10
°
CO
CO
enols
2,4,6-trichlorophenol
-C
a.
OO i-H
-1- +
+ 0
O 0
r-H r-H
X X
in 10
CO
CM in
CM 10
CO I-H
tn
+ CM
O 1
r-H O
X I-H
VO X
in in
I-H CO
CQ
CM CO
CO _J
>-H r-H
.
CO
o
c
CO
73
•r-
y*
CU
cu
ro
3
o-
cu
73
re
c:
i — i
It
i — i
• «N
CU
o
c:
cu
73
•i—
>
CU
73
CO
'i
-Hi
ii
.^
= Sufficient evidence
CO
re
c nn
o
73
73 C
cu ro
re «* •
-Q'— r—
cu cu
73 I- 73
CO 3 O
4-3 OO E
re o
i — Q.73
3 X r—
O CU O
r— .£!
re r— oo
o re cu
C S-
CO 0 -C
C- -1- 40
ro 4-3 c
ro o
• >> Q. c
CU 3
-C O t_
1— o ro
o cu
c
• C i-
r— re i—
cu E
73 3 CO
O -C .C
E -r-5
CU 3 C
Ol O
ro .«
OO C CO
•r- O t/)
4-> •>- re
i— •<-> -Q
3 ro
E p— * oo
ro cu
73 -C 4-3
cu c ro
IM -r- E
• r— *P—
C. r— 4-3
re re oo
CO E CO
c: -r-
•r- e 4-3
i— re c
cu o
J= >-H Q.
4-3
^} GJ
c: xi t-
o re
73
73 CO OO
CO 4-3 CO
oo ro Q.
re o o
l~t «r™ r—
73 OO
00 C
CU -r- C
Q. re
O CO E
I— 00 3
in o 3r
<—
4-3 4-3
•1— •
•r- O CO
T 3
£- 4-3 00
CO Q- O
Q- cu a.
O- O X
3 X CO
CO
^S £-
in * co
CD 00 4-3
co re
CU -r- 3
£- 73
re 3 o>
OO OO -I—
CO v^
Q-i— C.
o re ••-
r— t- t_
00 O 73
t— i — c:
re re ro
•• E E E
00 •!— •>— 3
-s£ c: c -c
t_ <: re ^-^
ro
cD •
CC I-H
c:
•r—
OO
CO
Q.
O
in
CO
Ol
c:
•^
a.
•r—
4-3
3
E
>^
XI
"S
4-3
re
3
O
re
o
V)
-a
cr
re
i—i
^b.
re
73 «
"^-~ 73
O) C
— -. O
r— O.
O E
E S
«•— ^
CO
C JC
•r- 4-3
CO <(-
0. 0
o
r— — -(— J
(rt .c:
01
If— .(—
U— (I)
rounded-o
olecular w
ro E
in CO
•r- -C
4-3
X
CO >>
73 -Q
C
•i— i— 1
1
c ro
CO 73
4-3 ^~
O Ol
CU Ol
^
CM
CO
ro
r—
t^
£
ro
tf
ft\
o
o
CO
cu
£_
Ol
cu
73
CU
E
re
00
cu
4-3
• c:
CO
si
CU
,-_
J3
to
4-3
(/)
g—
4-5 •
CO
C i —
-r~ ro
73 i —
CU T-
4-3 re
c >
CU CO
in
CO l/>
£_ CU
' Q- E
O
00
•r- CU
CO 5
Ol CO
o c:
•i— 00
o re
c_
re co
o 01
co ro
4-3 0
1- O
0 4->
"its co
4-3 -0
O 3
'Z. in
00
7-65
-------
as follows: the upper-limit slope estimate from the drinking water study is
9.9 x ID"3 (mg/kg/day)-1. Multiplying by the molecular weight of 92.5 gives a
potency index of 9.2 x 10-1. Rounding off to the nearest order of magnitude
gives a value of 10°, which is the scale presented on the horizontal axis of
Figure 7-7. The index of 0.92 lies in the fourth quartile of the 53 suspect
carcinogens that the CAG has evaluated, placing epichlorohydrin among the
weakest of these carcinogens.
7.1.5 Summary
7.1.5.1 Qualitative Assessment—The carcinogenicity of epichlorohydrin has
been demonstrated in rats and mice. Epichlorohydrin vapor produced papillomas
"'and sq'uamous cell carcinomas in the nasal tracts of male Sprague-Dawley rats
initially given 30 daily exposures, 6 hours/day, followed by lifetime observa-
tion. Consumption of epichlorohydrin in drinking water elicited neoplastic
lesions in the forestomachs of male Wistar rats, including a statistically
significant increase in the combined incidence of papillomas and carcinomas in
high-dose animals. The drinking water study is compromised in that pathologic
evaluations of decedents were not reported due to postmortem changes, the 81-
week duration of the study was less than the lifetimes of the animals, and the
number of animals in each dosage group was small. In a draft report of a
study in which epichlorohydrin was administered in water via gavage, a strong
dose-response with forestomach papillomas and carcinomas was present in both
male and female Wistar rats. This stucjy presented strong qualitative evidence
on the proximal-site carcinogenicity of epichlorohydrin.
Two studies involving dermal application of epichlorohydrin on the skin
of mice for their lifetimes elicited no tumor response. In one, thrice weekly
applications at a maximally tolerated dose were given to ICR/Ha mice, and in
the other, an uncertain dose (i.e., one brushful) was applied three times
7-66
-------
weekly to the skin of C3H mice. On the basis of these results, epichlorohydrin
is apparently ineffective as a complete carcinogen when applied to the skin.
Skin tumor initiating activity was found in a lifetime initiation-promotion
study with female ICR/Ha mice.
Weekly subcutaneous injection of epichlorohydrin at a maximally tolerated
dose in a lifetime study in female ICR/Ha mice produced a statistically signi-
ficant increase in local sarcomas. However, intraperitoneal injections once
weekly in females of this strain proved ineffective. None of the above studies
showed evidence of metastases of the proximal-site tumors.
A single subcutaneous injection of a low dose of epichlorohydrin did not
'produce a carcinogenic effect in a lifetime observation study with C3H mice;
however, survival in this study was poor, and the single low dose used would
appear to be a relatively weak challenge compared to lifetime treatment with
doses as high as those maximally tolerated.
Three epidemic!ogic studies of mortality in epichlorohydrin workers have
been conducted. One study of epichlorohydrin workers at Dow Chemical Company
in Texas failed to show an increase in cancer. This study is considered
inadequate for the evaluation of carcinogenicity, however, due to low exposure,
short exposure duration, short latency period, and young age of the cohort.
A second study of European epichlorohydrin workers at plants in four different
countries also showed no association between epichlorohydrin and cancer, but
was also judged to be inadequate for many of the same reasons as the Dow study.
A third study, a 1979 update of an ongoing study of workers at Shell Oil
Company, showed increased death's from respiratory cancer. Leukemia was also
present in an otherwise healthy cohort. This increase, however, was not statis-
tically significant, and the trend in the most recent 2-year follow-up period
actually weakened the evidence that epichlorohydrin is a human carcinogen.
7-67
-------
While the previous update had showed increasing lung cancer trends with time
since first exposure, this most recent update produced only one additional cancer
death and no additional lung cancer deaths.
7.1.5.2 Quantitative Assessment—Unit risk estimates for exposure to epichloro-
hydrin have been calculated by the CAG for both the animal and human studies.
For the animal studies, unit risk estimates were calculated from both drinking
water and inhalation studies. The study of male Wistar rats exposed to epi-
chlorohydrin in drinking water indicated that epichlorohydrin caused tumors of
the forestomach. Based on this study, the CAG estimated 2.81 x 10"7 as the
upper-limit lifetime risk from exposure to drinking water containing 1 ug/L of
•
epichlorohydrin. For the animal inhalation studies, a unit risk was calculated
using the nasal carcinoma response in male Sprague-Dawley rats exposed to
epichlorohydrin vapor in the Laskin et al. (1980) study. In this study, the
10 percent nasal carcinoma response at the 100 ppm exposure level was not used
for low-level environmental exposure extrapolation because it is a high-dose
short-term exposure and is not consistent with the low-dose long-term exposure
of the other groups. In order to extrapolate to low environmental levels, the
CAG used the multistage model on the controls and the two lower doses to provide
a 95 percent upper limit of risk. The linearized multistage model was used for
low-dose extrapolation in order to give an upper-bound estimate of lifetime
cancer risk, recognizing that uncertainties in both the qualitative evaluation
and the quantitative extrapolation method can yield a lower bound of risk
approaching zero. The plausibility of the upper bound is enhanced when there
is sufficient evidence for genotoxicity, which is the case with epichlorohydrin.
Using this procedure, the plausible upper bound of the individual life-
time cancer risk resulting from continuous exposure to air with an epichloro-
hydrin level of 1 ug/m3 is 1.2 x 10-6. por high exposures of short duration
7-68
-------
early in life, the same study, using the 100 ppm exposure level, estimates
lifetime risks ten times as high.
The study of employees from the Shell Oil Company was used to provide a
unit risk estimate for exposure to epichlorohydrin based on human epidemiologic
data. This study showed an increase in respiratory cancer deaths in an other-
wise healthy population. While this increase was not statistically significant,
the evidence for the carcinogenicity of epichlorohydrin in animals suggests that
a comparison of estimated risk levels should be made. On the basis of these
human data, the 95 percent upper-limit cancer risk resulting from continuous
exposure to air with an epichlorohydrin level of 1 ug/m3 is estimated to be 3.9
"x 10-5. Since this estimate, which is based on a nonsignificant increase, is
higher than that from the animal inhalation study, the lower animal-based esti-
mate of 1.2 x 10-6 (ug/m3)-1 will be used as the 95 percent upper-limit unit risk
estimate for inhalation exposure.
7.1.6 Conclusions
The animal evidence for the carcinogenicity of epichlorohydrin includes
nasal carcinomas in rats, local sarcomas in mice, and forestomach neoplasms in
rats. Applying the International Agency for Research on Cancer (IARC) classi-
fication system (Appendix F), this level of evidence would be considered suffi-
cient for concluding that epichlorohydrin is carcinogenic in experimental
animals. Although the human epidemiologic evidence with regard to epichlorohy-
drin alone is negative, sequential exposure to the IPA process and epichloro-
hydrin produced evidence for the carcinogenicity of epichlorohydrin. However,
this evidence has only marginal'statistical significance (P < 0.1). Using the
IARC system for describing the overall evidence for carcinogenicity, epichloro-
hydrin would be classed as a 2B chemical.
As described in the mutagenicity section, epichlorohydrin has been demon-
7-69
-------
strated to be mutagenic in both prokaryotic and eukaryotic systems. Epichloro-
hydpin 1s a direct-acting alkylating agent, and therefore does not require
metabolic activation to attack biological macromolecules.
Quantitative estimates of the carcinogenic potency of epichlorohydrin have
been made by the CAG for both drinking water and inhalation. On the basis of
forestomach tumors in male Wistar rats exposed to epichlorohydrin in drinking
water, a lifetime exposure to 1 ug/L of epichlorohydrin in drinking water is
estimated to present an upper-limit risk of 2.8 x 10-7.
The CAG's inhalation estimates were made on the basis of animal data on
nasal carcinomas and human data on a non-statistically significant increase in
'respiratory cancers. These two unit risk estimates are not very close; the
upper-limit estimate from nonsignificant human data is 3.9 x 10~5 (ug/m3)-1; the
upper-limit estimate from positive animal data is 1.2 x 10~6 (ug/m3)-1. The
estimate from the animal data is chosen. The estimate based on the nonsigni-*
ficant increase in the human data based would have been chosen only if it had
been lower than the animal-based estimate. The animal data estimate does not
use the nasal carcinoma response to short-term high exposures because the CAG
feels that such an exposure does not reflect environmental experience and is
not consistent with the long-term lower-dose response. Under conditions of
short-term high exposures, lifetime risks can be extrapolated and are estimated
to be greater than from long-term exposures by a factor of ten.
The carcinogenic potency of epichlorohydrin lies in the fourth quartile
among the 53 suspect carcinogens evaluated by the CAG. This places epichloro-
hydrin among the weakest of the substances that the CAG has evaluated as suspect
carcinogens.
7-70
-------
7.2 MUTAGENICITY
7.2.1 Introduction
Chemicals that induce gene mutations and chromosomal aberrations have
been regarded as a potential risk to human health. If mutations occur in
human germ cells, they may be passed on to future generations, causing
deleterious effects. On the other hand, if mutations occur in somatic
cells, they may lead to the onset of diseases such as cancer. The aim of
mutagenicity risk determination is to assess the risk that particular
chemicals pose to human well-being.
Epichlorohydrin has been tested for its ability to cause genetic
damage in both prokaryotic and eukaryotic systems. The prokaryotic systems
include assays for gene mutations and reparable genetic damage in bacteria.
The eukaryotic systems include gene mutation studies in yeast, Drosophila.
and mammalian cells, and chromosomal aberration studies in human and other
mammalian cells exposed to epichlorohydrin both in vitro and in vivo.
Positive findings in most of these mutagenicity assays clearly indicate
that epichlorohydrin is a mutagen. The following is an analysis of the
literature pertaining to the mutagenic effects of epichlorohydrin.
7.2.2 Gene Mutations in Bacteria
7.2.2.1 Salmonella Assay—The potential of epichlorohydrin to induce
reverse mutations in Salmonella typhimurium has been documented by many
investigators. Sram et al. (1976) tested epichlorohydrin for the induction
of back mutations (revertants) in S. typhimurium using both the spot test
and suspension assay. In the spot test, S. typhimurium strains hi.s646,
TA100, TA1950, TA1951, TA1952, TA1534, TA1537, and TA1538 were used.
Epichlorohydrin (purity not given) concentrations of 1 percent (0.05 pinole),
5 percent (0.27 umole), 10 percent (0.54 (jmole), and 100 percent (1.10
(jmole) were employed. Epichlorohydrin was applied in 50 pi quantities on
to the center of the bacterial petri dishes. Positive (+) results were
noted only in the strains hisG46 and TA100. The positive results with
hisG46 and TA100 together with the negative result with the other strains
indicate that epichlorohydrin is acting as a base-pair substitution mutagen
in Salmonella. Furthermore, epichlorohydrin is active without metabolism
by mammalian enzymes and that this is consistent with its known activity as
an alkylating agent and ability to react directly with DNA bases (Sram et
al., 1981).
7-71
-------
In the suspension assay, only strains hisG46 and TA100 were used. The
cells of the strain hisG46 were treated with epichlorohydrin at concentrations
of 1.08 x 10~4, 1.08 x 10~3, 5.40 x 10~3, 1.08 x 10~2, 2.70 x 10~2, 5.40 x
10""2, 1.08 x lO"1, and 5.40 x 10"1 M for 60 minutes without a metabolic activa-
tion system (S-9) and assayed for revertant colonies. The concentration of
5.40 x 10 M was toxic and produced 100 percent cell killing. Numbers of
revertants/109 survivors were 6, 4, 9, 18, 15, 1.68 x 10s, 3.18 x 107 for the
above concentrations, respectively. In TA100, epichlorohydrin concentrations
of 1.08 x 103, 5.40 x 10"3, 1.08 x 10~2, 5.40 x 10"2, and 1.08 x lO"1 M were
used. The revertant frequencies obtained respectively for these concentrations,
except for the concentration 1.08 x 10 M, which was toxic for 100 percent of
the cells, were 2.25 x 10l6, 9.64 x 101, 2.85 x 105, 3.44 x 105, and 5.00 x
106. The spontaneous revertant frequencies were 6 in hisG46 and 2.25 x 10 in
TA100. A clear dose-response relationship was evident in the experimental
groups. These results indicate that epichlorohydrin is mutagenic in S.
typhimurium strains hisG46 and TA100.
Andersen et al. (1978) tested epichlorohydrin for its mutagenic poten-
tial in the plate incorporation assay using S. typhimurium strains TA100
and TA1535 as described by Ames et al. (1975). In strain TA100, epichloro-
hydrin doses of 0.5, 1.0, 1.5, 2.0, and 2.5 jumoles/plate (in 100 jil DMSO)
gave 352, 500, 650, 800, and 1,200 revertants/plate, respectively, indica-
ting a clear dose-response relationship (Figure 7-8). The experiment was
carried out in the absence of an S-9 mix. The spontaneous frequency of
revertants for TA100 was 250 revertants/plate. No solvent control data
were given in this paper. In strain TA1535, the number of revertants/plate
were 216, 474, and 1,418 at epichlorohydrin concentrations of 0.254, 0.635,
and 2.54 umoles/plate in the absence of an S-9 mix. In the presence of an
S-9 mix, doses of 2.54, 6.35, and 25.4 jjmoles/plate induced 34, 152, and
911 revertants/plate, respectively. The spontaneous frequency in the
strain TA1535 was 36 revertants/plate. These results indicate that epi-
chlorohydrin is mutagenic in Salmonella strains TA100 and TA1535.
7-72
-------
1200
UJ
§
Q.
OC
Ul
a.
to
UJ
1
O
O
I-
I
UJ
ui
ec
O
z
300 —
1.0 1.5
DOSE PER PLATE, p mole
2.0
2.5
Figure 7-8. Mutagenicity of
epichlorohydrin for S. typhimurium TA100 (revert by
base-pair substitution).
Source: Andersen et al. (1978).
Planche et al. (1979) also investigated the mutagenic potential of
epichlorohydrin in S. typhimurium strain TA100 using the plate incorpora-
tion assay (Ames et al. 1975). Epichlorohydrin (Merck-Schuchardt, Darmstadt,
F.R.G.) concentrations of 10, 100,.1,000, and 5,000 umoles in 0.1 ml actone/
plate were employed in the study. Epichlorohydrin was mutagenic with a
clear dose-response relationship (Figure 7-9).
Stolzenberg and Hine (1979) also detected positive results with a
clear dose-response relationship in S. typhimurium strains TA100 and TA1535
using epichlorohydrin (purity 99+%, Aldrich Chemicals) concentrations of 2,
4, 6, 8, and 10 umoles/plate both in the presence and absence of an S-9 mix
(Figure 7-10).
7-73
-------
3500
I
a.
« 1000
z
3
o
o
I
UJ
cc
100
10
100
CONCENTRATION.
100°
5000
Figure 7-9. Mutagenicity of epichlorohydrin { • ), at
various concentrations (nmol/ml of soft agar) in
S. typhimuriumTMQQ3.
aThe compounds were added as an acetone solution
0.1 ml/plate). Solvent control assays (-----). The
number of spontaneous his revertants/plate has not
been subtracted. Mean values from 3 to 6 plates are
plotted.
Source: Planche et al. (1979).
7-74
-------
ui
s
WITH S-9
WITHOUT S-9
200 400 600 800
COMPOUND IN TOP AGAR, M mote
Figure 7-10. Dose-response curves for epichlorohydrin.
Source: Stolzenberg and Nine (1979).
1000
7-75
-------
Eder et al. (1980) investigated the mutagenic potential of epichloro-
hydrin in the S. typhimurium strain TA100 both in the presence and absence of
S-9 mix using the suspension assay. Epichlorohydrin (purity 99.5%) induced
275 revertants/umoles and in the presence of S-9 mix there were 70 revertants/
umoles. There was a clear linear dose-response relationship between the
number of revertants/plate and the concentrations of the test compound (Figure
7-11).
Bartsch et al. (1980), tested epichlorohydrin for its mutagenic potential
in S. typhimurium strains TA100 and TA1535 in the absence of an S-9 mix using
the plate incorporation assay of Ames et al. (1975). These investigators used
an epichlorohydrin concentration of 1.1 x 10 ^moles/plate and found epichloro-
hydrin to be highly mutagenic in the strain TA100 and mutagenic in the strain
TA1535. These investigators also mention that the dose-response curve is
linear. Detailed tabulated data are not given in this report. Based on the
other reports available on epichlorohydrin mutagenicity, the report of Bartsch
et al. (1980) is regarded as an indication of positive mutagenic response of
epichlorohydrin.
Elmore et al. (1976) and Voogd et al. (1981) detected epichlorohydrin to
be mutagenic in S. typhimurium strain TA100 with a dose-response relationship.
However, the test compound was more mutagenic in the absence of an S-9 mix.
Bridges (1978) detected 600 revertant colonies/plate with a concentration
of 2 ug/ml of epichlorohydrin in agar in S. typhimurium strain TA1535 when
plates were incubated in sealed airtight jars. When the same concentration
was added externally to the plates and allowed to evaporate into the sealed
jar, only 300 colonies per plate were induced. This indicates that epichloro-
hydrin can freely penetrate into aqueous media. If agar plates containing
epichlorohydrin are .not incubated in a sealed incubator, the activity may be
lost as indicated by fewer induced mutants. Simmon, as quoted by Bridges
(1978), was able to detect revertants in S. typhimurium strain TA1535 at a
concentration of 3 ug/1 air, which happens to be the U.S. OSHA maximally
allowed concentration for a 10-hour occupational exposure period. In strain
TA100, Bridges (1978) detected that a concentration of 1.25 pg/1 air to be
mutagenic when the plates were incubated in sealed airtight containers.
7-76
-------
WITH S-9 MIX
O WITHOUT S-9 MIX
10 20
MOLES/INCUBATION VOLUME (2 ml)
Figure 7-11. Mutagenicity of epichlorohydrin with \ 9 )
and without ( O ) S-9 mix.
Source: Eder et al. (1980).
7-77
-------
Laumbach et al. (1977) reported epichlorohydrin to be mutagenic for S.
typhimurium strain TA100. Epichlorohydrin at a concentration of 4.746
umoles/plate induced 2,856 revertants/plate. No dose-response data were
available in this report.
It is clear from the foregoing account that epichlorohydrin is mutagenic
in the S. typhimurium reverse mutation assay with a clear-cut dose-response
relationship. Epichlorohydrin was particularly active in S. typhimurium
strains TA100, TA1535, and G46 indicating that it is an inducer of mutations
through base-pair substitution. Furthermore, the mutagenic activity is
expressed in the absence of a metabolic activation system indicating that
epichlorohydrin is a direct-acting mutagen.
7.2.2.2 Mutations in Klebsiella—In an abstract published by Voogd (1973),
epichlorohydrin was reported to be mutagenic in Klebsiella pneumom'ae by
inducing streptomycin-resistant mutations. No details regarding the con-
centrations of the test compound or the frequencies of mutations in the
experimental and control groups are provided in this abstract. Consequently,
this report cannot be critically evaluated.
7.2.2.3 Host-Mediated Assay—Epichlorohydrin was tested for its ability to
induce reverse mutations in the host-mediated assay (Sram et al. 1976).
Female ICR mice, aged 10-12 weeks and weighing 35 g each, were injected
intraperitoneally with tester strains of S. typhimurium G46, TA100, TA1950,
TA1951, and TA1952. The test compound (purity not given) in the concentra-
tions of 50 and 100 mg/kg, dissolved in 0.2 ml of DMSO, was administered to
groups of five mice intramuscularly. The control group consisted of five
animals and received the tester strains and 0.2 ml DMSO. Mice were killed 3
hours postinjection of epichlorohydrin, and their peritoneal fluid containing
the bacteria was assayed for revertants. The result was expressed as C, which
is a relative mutagenicity; i.e., the ratio between the mutation frequency in
the experimental groups and the mutation frequency in the control group. C
greater than 2 was considered to be a significant increase. A significant
increase (C greater than 2) in the frequency of revertants was noted for
strains G46, TA100, and TA1950. In strains TA1951 and TA1952, the revertant
frequency was similar to the control frequency (C less than 2). These results
indicate that epichlorohydrin is mutagenic in the host-mediated assay employing
Salmonella strains G46, TA100, and TA1950.
7-78
-------
7.2.2.4 Body Fluid Analysis—In an abstract published by Kilian et al.
(1978), urine samples of two industrial workers exposed to epichlorohydrin
(25 ppm), as a result of accidental spill, induced a twofold increase in
the revertant frequency in S. typhimurium TA1535 over the control value.
In six workers exposed to low levels (0.8-4.0 ppm) of epichlorohydrin, the
urine samples showed no increase in the revertant frequency over the control
value. Urine samples of mice orally exposed to 200-400 mg/kg of epichloro-
hydrin also exhibited mutagenic activity in S. typhimurium TA1535. This
abstract contained no information on the number of revertants in the experi-
mental and control groups and thus critical evaluation cannot be made.
7.2.3 Bacterial DNA Repair Tests
Epichlorohydrin was tested for its ability to damage the DNA using the
PolA plate assay of Rosenkranz and the Rec-assay of Kada (Bridges 1981:
El more et al. 1976). These tests revealed that epichlorohydrin produced
reparable DNA damage similar to that of an alkylating agent in the absence
of metabolic activation. Epichlorohydrin at a concentration of 0.01 ug/ml
produced a positive response in the PolA assay. In the Rec-assay, a con-
centration of 0.1 ug/ml produced a positive result.
7.2.4 Gene Mutations in Neurospora
Epichlorohydrin was tested for its ability to induce point mutations
(reverse mutations) in the mold Neurospora crassa (Kolmark and Giles 1955).
The purple adenineless mutant, 38701 strain, of N. crassa was used in this
experiment.
Epichlorohydrin at a concentration of 0.15 M (14 mg/ml) was added to
the suspension of microconidia at 25°C and allowed to incubate for 15, 30,
45,, and 60 minutes. The microconidia were washed free of the test compound
and plated on minimal agar plates. The number of viable and of surviving
conidia in the treated and control series was determined by plating diluted
samples on minimal medium supplemented with adenine. The revertant fre-
quencies in the experimental groups were 8.5 (94.7% survival), 13.0 (87.8%
survival), 135.2 (41.5% survival), and 411.0 (0.72% survival), respectively,
for the above treatment periods per 106 survivors. The control frequency
was 0/10 survivors. The positive mutagenic effect of epichlorohydrin in
N. crassa was also confirmed by Westergard (1957).
7-79
-------
7.2.5 Gene Mutations In Yeast
Epichlorohydrin was found to induce gene mutations and other types of
genetic damage in yeast. Vashishat et al. (1980) investigated the ability
of epichlorohydrin to induce reverse gene mutation, mitotic crossing-over,
and gene conversion in a diploid strain of the yeast Saccharomyces cerevisiae
D7. Two batches of cultures were used and assayed for cross-overs, rever-
tants, and convertants: one batch was treated with 0.065 M epichlorohydrin
for 0, 5, 10, 15, and 20 minutes, and another batch of cultures was treated
with 0.13 M epichlorohydrin for 0, 5, and 10 minutes. In the first batch,
the cross-over frequencies were 0.13, 0.40, 0.59, 1.68, and 2.17 percent,
revertants/107 survivors were 30, 190, 366, 427, and 297; and convertants/10
were 27, 59, 113, 183, and 297, respectively, for 0, 5, 10, 15, and 20
minutes of treatment. In the second batch, the cross-over frequencies were
0.15, 0.55, and 1.39 percent revertants/10 ; survivors were 46, 326, and
547;'and convertants/106 were 33, 109, and 330, respectively, for 0, 5, and
10 minutes of treatment, indicating that epichlorohydrin was mutagenic in
the yeast. Sora et al. (1979) also reported the induction of gene muta-
tions both of base-pair substitution and insertion/deletion-type, mitotic
crossing-over, and mitotic gene conversion in the yeast. However, these
investigators did not provide data to support their claim. Heslot (1962)
reported (abstract) the induction of Arg mutations in Schizosaccharomyces
pombe by epichlorohydrin.
7.2.6 Gene Mutations in Mammalian Cell Cultures
Moore-Brown and Clive (1979) demonstrated the induction of gene muta-
tions at the thymidine kinase (TK) locus in mouse lymphoma cells in vitro
by 0.21, 0.42, and 0.65 umoles of epichlorohydrin. Two types of mutant
colonies (TK-/-), large and smalK were induced by epichlorohydrin. The
large mutant colonies followed a linear dose-response relationship indica-
ting a typical one hit point mutational mechanism. However, the dose
induction curve for the small colonies indicated a multihit mutational
mechanism. From the shape of the dose-response curve (Figure 7-12), there
appears to be little doubt about the mutagenic potential of epichlorohydrin
in cultured mammalian cells.
7-80
-------
0 20 40 60
CONG. OF EPICHLOROHYDRIN l^g/ml)
Figure 7-12. Dose-response curve for
epichlorohydrin-treated cultures. The total
mutant frequency is divided to show TK"'"
mutant induction of large colony ( « ) and
small-colony ( O ) mutants versus the
concentration of epichlorohydrin.
7.2.7 Sex-Linked Recessive Lethal Test in Drosophila
Epichlorohydrin was tested for the induction of sex-linked recessive
lethal mutations in Drosophila.
Rapoport (1948) analyzed 526 chromosomes from the experimental and 887
chromosomes from the control groups. The frequencies of sex-linked reces-
sive lethal mutations were 0.7 percent and 0 percent, respectively, in the
experimental and control groups, indicating epichlorohydrin was mutagenic
in Drosophila. Details about the concentration of the test compound and
experimental conditions were not given in the paper.
The observations of Knapp et al. (1982) indicated that epichlorohydrin
induced sex-linked recessive lethal mutations in Drosophila (Table 7-9).
Flies were exposed to epichlorohydrin by injection and feeding methods. In
the injection method, 4-day-old male flies (Oregon-K) were given 2.6, 5.1,
25.5 [jmoles of epichlorohydrin and individually mated to 3 Base females per
7-81
-------
u
1
I
a
i
c
z:
r- i
Q-
IC
s
s
CJ
HH
UJ
s
5
1
§
o.
*^
g
UJ
UJ
I— i *
1/1
UJ
a
C£
a.
2=
^
«
u«
0
s
S
1
cn
i4-
LiJ
t
3
1—
.
*J
j
3 to
3 +
c
«o
+J
s^*»
= 1
o
4->
01
cLz
UJ
in cn CM ^
. . . 1 '
O CD CD CD 1 OCD
loiOrH 1 rHCO
rH rH
rHCM^r^ CO
0000000
t CO O ^" *3* CO CD
jrHin3-oog
r-H CM
arH cn r*^ co i**«
CO^t-CO CMrH ,
O CD CD O CD OCD
rH t**- CM rH O CM r^
M in rH O CO CD ^~
in cn «* co m r^ CM
cor^- co ID CM
cn*sj- co cn CM
O O i-H CD O CD C3
r^- r^- co to CM rH r-*
CO in CM CO T^- CZ> CM
in CD m co r-- ^ CD
CM r^ ^ co r*«
tn ID r*~ rH rH
CDOO rHO OO
U3CO CO CM CO CO ID
CM tn **• CM CM r^ rH
in cn «T cn UD r^ CM
m
oicn
c c
c c c= e~5*o
o o o o cu a>
*4j !p '^ 4^ <2i <£i 1
o u o u
CO O) 03 CU J= JZ
c c c c ^* ^t*
HH I— 1 1— 1 1— I CM CM
ID rH m in U3 i — 1
CMintnincMin i
CM CM
rH CM CO ^J- in IO O
||«
o in c=
•r- O O
-p CU'r- -P
.S,5l£-
c in o m
•r- «j in in
3: in
c co ai
«r- U)-C U
Qj5 '1
T3 O -P
T3 * O
fas: in c
m cu m
to to c ra
CM ^
QJ U P
O O -P
I- in J=
(J Ol O)
3 C 5- 3
m-r- a> o
5| QJ.r?^—
£_
0 ai * C
o»*- t- ^
rd id co o
^lir^-.^
.is nj
o o J=&^
o-PLrt
.«£- CO •
"O ,*"! r-" ^
§J=4->»t-
f- O) C O
JQ 3 CO
o u in
t_ S- S- C
o
CXP Q.*£
* — . CO CD fd
Ulr-O S-
*5)t- in o>
> o
*^s O) E U
t- s
in a) r—
co s m nj
r- « Ct
fd in m «r-
E Ot CM 1—
if,r^ O> -P
14- r— fd
in o> J= -P
our s c
CO-P CU
- >
Cy q_ U) r—
O> O S- O
s- 3 w
o o ^- tn
4-1 f_ CM *i—
CU r—
CU °"T- X
4-> in J= 3
^1-^ s
r— a) in
3 . "O
•om *— N
«r-CM W VD
.S^.S+J
•a Oi— c:
•Si='<~ E
"2*rc 5 nj
rd U ex
QJ 4- X
-o-»-3 o a
4-3-r- -P C
co $- ai *" .
S- CO 0 +^ tj
ax ex o (d
a> o a>
^ >>r^"'*^*r^
O CU a> in ••-
4J cd'r- > 3 S-
c E-P cu "a
° ¥ CXJ= S "0
>» c in t-
C 0 »r— S-
O) Cd O O) CT
-t-> -o in c: cd c:
f5 1 -r- -r-
r- S- OiT3 T3 X
3 3 C CU CU*r-
E 0'*- CU >>E
=3 U-T3M- cd
o cu cn'cxo
^ ..4— C O)
11 fel'ilfe
CJ 2= «d T3 J= CX
CM
§
v— ' '
r—
rd
cu
s-
cd
c
^
o>
u
3
O
to
7-82
-------
brood. A concentration of 2.6 mM epichlorohydrin induced sex-linked reces-
sive lethals in broods B and C but not in broods A, D, and E. However, the
concentrations of 5.1 and 25.5 mM induced sex-linked recessive lethal
mutations in all five broods. In the feeding method at the concentration
of 2.6 mM, no sex-linked recessive lethals were found. The other two
concentrations, 5.1 and 25.5 mM, were toxic to the flies, resulting in 70
percent and 100 percent mortality, respectively. This study also indicates
that the negative results with the feeding study may be due to the fact
.that flies did not consume the test compound in sufficient quantities or
that the test compound was unable to reach germ cells probably because of
its rapid metabolism by the other drgans in the body.
,Wurgler and Graf (1981) tested epichlorohydrin in the Drosophila
sex-linked recessive lethal test and found it to be negative. Flies,
Berlin-K 2-day-old males, were fed with 0.2 percent of epichlorohydrin for
3 days on glass filters. Epichlorohydrin was dissolved in 2 percent DMSO
and then diluted in buffered 5 percent sucrose solution (pH 6.8) before
feeding. In 2,209 chromosomes tested in broods, days 1-3, there were 7 or
0.32 percent recessive lethals. In water and solvent controls, the reces-
sive lethal frequencies were 0.28 percent and 0.40 percent, respectively.
The negative response in this experiment is probably due to the problems
found in feeding studies such as the flies not eating sufficient amounts of
the test compound, or the rapid metabolism and distribution of the compound
in other tissues so that it was unable to reach the germ cells. If injec-
tion studies were performed in this experiment, a positive response may
have been obtained. The injection studies of Kramers (quoted by Vogel et
al. 1981) conclusively show that epichlorohydrin (purity not given) is
mutagenic in Drosophila sex-linked recessive lethal mutation tests.
7.2.8 Chromosomal Aberrations in Human and Other Mammalian Systems
7.2.8.1 Studies on Human Chromosomes in Vitro—Kucerova et al. (1976)
investigated the cytogenetic*> effects of epichlorohydrin in cultured human
blood lymphocytes. Blood samples were obtained from two healthy donors
(one male and one female) and cultured for 56 hours. Two series of experi-
ments were conducted. In the first series of experiments, epichlorohydrin
7-83
-------
(Czechoslovak Chemical Industry) was added for the last 24 hours of culti-
vation at concentrations of 10"5, 10"6, 10~7, 10"8, 10"9, 10"10, 10 , and
1 M. In the second series of experiments, cells were exposed to 10
and 10~5 M concentrations of epichlorohydrin in three ways: (1) for 1 hour
before the beginning of cultivation (GQ); (2) for 1 hour between the 24th
and 25th hour of cultivation (G-^; and (3) for the last 24 hours of culti-
vation (6). Chromosome preparations were stained with Giemsa and 100
metaphases were scored for each dose. Aberrations were classified as
chromatid breaks, chromatid exchanges, chromosome breaks, and chromosome
exchanges. Gaps were scored separately. In the first series of experi-
ments, a dose-related response of chromosomal aberrations was obtained.
The aberration frequencies were 8.9, 3.3, 1.3, 1.0, 1.7, and 0.7/100 meta-
phases, respectively, for the above doses. The control frequency was 1
aberration/100 metaphases. Chromosome and chromatid breaks were the most
common type of aberrations found. In the second series of experiments, no
differences were found between the treated (GQ and G-^ and control groups.
However, in the 24-hour treatment group, the concentration of 10 M was
too toxic; only 10"5 M increased the number of aberrations (9.2/100 meta-
phases) compared to the control value of 1.9 aberrations in 100 metaphases.
Appropriate positive (TEPA) and solvent (DMSO) controls were employed in
this study.
Kucerova and Polivkova (1976) tested the clastogenic effect of 10 M
epichlorohydrin in cultured human lymphocytes 1 hour before initiation (GQ)
and 24 hours after the initiation of DNA synthesis (S), using conventional
and Giemsa-banding (G-banding) procedures. One hundred metaphases from
conventional staining and 100 metaphases from G-banding procedures were
analyzed for chromosomal aberrations. The banded preparations according to
these authors exhibited higher incidence of aberrations, 6 percent at 1 and
18 percent at 28 hours of treatment, as compared to frequencies of 2.5
percent and 1.5 percent aberrations for the same periods of treatment in
conventionally stained chromosome preparations. In the negative controls,
there were 0.7 percent aberrations in conventionally stained, and 3 percent
in banded preparations. Solvent and positive controls were also used in
these studies. Even though increased aberration frequencies were noted in
the banded chromosomes by these investigators, the banding technique in
general has been very rarely used in screening chemicals for mutagenic
7-84
-------
activity. These investigators studied only one concentration of the chemical
and consequently no dose-response data are available. Due to these short-
comings, this report cannot be evaluated critically.
Norppa et al. (1981) also demonstrated the clastogenic effects of
epichlorohydrin in cultured human lymphocytes. Concentrations of 0.05,
0.20, 0.40 mM epichlorohydrin induced 0 percent, 18 percent, and 13.3
percent aberrations. In controls, the frequency of aberration was 0.5
percent. Two hundred metaphases per dose from replicate cultures were
scored. Epichlorohydrin was dissolved in acetone prior to use. Fisher's
exact probability test (one-tailed) was used to analyze the data.
7.2.8.2 Studies on Rodent Chromosomes in Vitro—Negative results on the
clastogenic effects of epichlorohydrin in cultured rat liver cell line
(RL1) were obtained by Dean and Hodson-Walker (1979). Epichlorohydrin at
concentrations of 5, 10, 15, 20 ug/ml induced 0.8, 0.9, 0.8, and 1.5 percent
aberrations, respectively, as compared to the controls (0.6%). The median
lethal concentration (LC5Q) was 40 |jg/ml. It appears rat liver cells are
resistant to the clastogenic action of epichlorohydrin through detoxifica-
tion.
7.2.8.3 Studies on Human Chromosomes in Vivo—Humans occupationally exposed
to epichlorohydrin have been examined for chromosomal abnormalities in
their blood lymphocytes. Positive results were reported by many investi-
gators (Kucerova et al. 1977; Sram et al. 1976; Picciano 1979).
Kucerova et al. (1977) examined 35 workers 23 to 54 years of age
before they started work, 1 year after they started work, and 2 years after
they started work in a newly established chemical plant manufacturing
epichlorohydrin. The workers were not previously exposed to either radia-
tion or drugs. According to these investigators, the concentration of
epichlorohydrin to which the workers were exposed exceeded the limits (1
2
mg/m ) of acceptable concentration in Czechoslovakia. Chromosome prepara-
tions were made from blood samples cultured for 56-58 hours and stained
with Giemsa. Slides were coded and scored blind by two collaborating
laboratories. Two hundred metaphases were scored from each worker for each
o
of three intervals. Data were analyzed with the X test. Before the
workers started to work in the epichlorohydrin manufacturing plant, they
had an average frequency of 1.37 percent + aberrations. One year after
7-85
-------
they started to work, the aberration frequency increased to 1.91 percent; 2
years later the aberration frequency increased to 2.69%. Statistical
analysis revealed that the aberration frequency in workers exposed for 2
years was highly significant (P<0.0001) as compared to controls. The
aberrations were mostly in the form of chromatid and chromosomal breaks.
The results of Kucerova et al. (1977) were confirmed by Picciano
(1979) in the United States. Picciano (1979) examined the blood lympho-
cytes of 93 workers (20-62 years of age) exposed to epichlorohydrin and 75
matching controls (20-49 years of age). Two hundred cells from each indi-
vidual were analyzed by five independent laboratories. Picciano indicated
that the details of exposure data were not available to him. The aberra-
tion .data were analyzed using the X test. In the exposed workers, there
were 4.34 percent chromatid breaks, 0.96 percent chromosome breaks, 0.13
percent marker chromosomes, 0.12 percent severely damaged cells, and 4.25
percent abnormal cells with a total of 9.80 percent aberrations. In the
controls there were 2.15 percent chromatid breaks, 0.51 percent chromosome
breaks, 0.08 percent marker chromosomes, 0.01 percent severely damaged
cells, and 2.38 percent abnormal cells with a total of 5.13 percent aberra-
tions.
Sram et al. (1980) examined 28 workers, 34 matching controls, and 21
general population subjects. None of the subjects was previously exposed
to radiation or other mutagenic chemicals according to these investigators.
Epichlorohydrin concentration in the exposed workers ranged over the maxim-
ally permitted concentration limits (1 mg/m3) in the last 2 years prior to
chromosome analysis. The following frequencies of aberrant cells bearing
chromosome and chromatid breaks were detected in the various groups. In
the epichlorohydrin exposed group, the aberration frequency was 3.12 percent,
in matching controls the aberration frequency was 2.06 percent, and in the
general population control group the frequency was 1.33 percent. Statis-
tical analysis revealed significant difference between the epichlorohydrin
group and the matching control group (P<0.05). Similarly, significant
differences were found between the epichlorohydrin exposed group and the
general population control group (P<0.01).
7.2.8.4 Studies on Rodent Chromosomes in Vivo—Clastogenic effects of
epichlorohydrin in vivo have been reported in the bone marrow cells of
laboratory rodents (Sram et al. 1976, 1981; Dabney et al. 1979).
7-86
-------
Sram et al. (1976) studied the in vivo clastogenic effects of epichlor-
ohydrin in mouse bone marrow cells. Epichlorohydrin (l-D™ was 100 mg/kg)
was administered by both intraperitoneal and oral routes. Chromosome
analysis from bone marrow cells of mice injected intraperitoneally with 1,
3, 5, 10, 20, and 50 mg/kg for 24 hours revealed 2.8, 6.0, 10.0, 27.2, and
20.4 percent aberrant cells, respectively, with a clear dose-response
relationship. In DMSO control animals, 4.0 percent of the bone marrow cells
exhibited chromosomal aberrations. High frequencies of aberrant cells were
also noted in mice injected with subacute doses (five daily injections of
5, 10, and 20 mg/kg of epichlorohydrin). The incidence of cells with
chromosome aberrations were 38.0, 36.0, and 80.4 percent, respectively, for
these doses. When epichlorohydrin was given orally at 5, 20, 40, and 100
mg/kg, dose-related increases in the incidence of aberrations (6, 24.0,
22.4, and 29.5 percent, respectively) were noted; these were mainly in the
form of chromatid breaks.
Sram et al. (1981), in a review article, refers to a paper by these
same authors (1976) in which they reportedly investigated the clastogenic
effect of epichlorohydrin in the bone marrow cells of the Chinese hamster.
The incidence of aberrations at 5-20 mg/kg of epichlorohydrin was 2.0-2.4
percent compared to the .control frequency of 0.6 percent aberrations.
However, Sram et al. (1976) does not reveal such a report of chromosomal
studies in the Chinese hamster.
Dabney et al. (1979), as cited by Sram et al. (1981), failed to detect
chromosomal aberrations in groups of 10 male and 10 female rats exposed to
epichlorohydrin at 0, 5, 25, or 50 ppm for 6 hours/day, days/week for 4
weeks in air. The aberration frequencies in treated groups were 0.1-0.4
percent. It appears that rat bone marrow cells are relatively insensitive
to epichlorohydrin compared to mice. See also earlier reference to the rat
liver cell line (RL1).
In vivo chromosomal aberration studies indicate that epichlorohydrin
is mutagenic in mouse bone marrow cells but not in rat bone marrow cells.
The difference is probably due to the fact that rats are resistant to the
effects of epichlorohydrin. Such an observation was also made by Norppa et
al. (1981) in cultured rat liver cells.
7-87
-------
7.2.8.5 Micronucleus Assay—Epichlorohvdrin was tested for its ability to
induce micronuclei in the bone marrow cells of mice (Kirkhart 1981; Tsuchi-
moto and Matter 1981). Micronuclei are formed when chromosome fragments
that lack centromeres fail to incorporate into daughter nuclei and these
can be detected in polychromatic erythrocytes (PCEs) of the bone marrow
under the microscope.
Kirkhart (1981) demonstrated the negative response of epichlorohydrin
in mice. Mice were treated i.p. twice, once at 0 and the other at 24
hours, with the test compound at concentrations of 0.0225 mg/kg (12.5%
LD5Q), 0.045 mg/kg (25% LD5Q), and 0.09 mg/kg (50% LD5Q). They were sacri-
ficed 6 hours after the second injection. Bone marrow smears were made for
each .concentration and 1,000 polychromatic erythrocytes (PCEs) were examined
for the .presence of micronuclei. The frequencies of micronuclei were 10,
15, and 12 per 1,000 PCEs, respectively, for the above doses. In the
negative control, the frequency of micronuclei was 4 per 1,000 PCEs and in
the positive (TMP) controls the frequency was 113 per 1,000 PCEs. Statis-
tical analysis (Mackey and MacGregor 1979) revealed no significant differ-
ences between negative controls and experimental groups.
Tsuchimoto and Matter (1981) also reported the negative response of
epichlorohydrin in the micronucleus assay. At concentrations of 0.02
(12-5% of LD5Q), 0.04 (25% of LD50), and 0.08 (50% of LD5Q) mg/kg, the test
compound induced 0.10, 0.08, and 0.10 percent micronuclei as compared to
0.05% micronuclei in the negative control. The criteria set for positive
conclusion were: (1) two or more mice per group with micronucleated poly-
chromatic erythrocyte frequencies above 0.40 percent, (2) one or more
treated groups with mean polychromatic erythrocytes frequencies above 0.30
percent, and (3) statistical significance (Kastenbaum and Bowman 1970) in
one or more, treated groups. Epichlorohydrin did not meet any of these
criteria and thus concluded as negative by these investigators.
It should be noted that the failure of epichlorohydrin to induce
micronuclei does not necessarily mean that the test compound is not mutagenic.
It may be that chromosome aberrations that were induced were probably of
reciprocal exchange type and consequently no micronuclei were formed.
7.2.8.6 Dominant Lethal Assay—The dominant lethal assay detects dominant
lethal effects induced by chemical mutagens in parental germ cells. The
germ cells carrying dominant mutations when they fertilize normal counter-
parts result in the death of the fetuses during development, which can be
7-*
-------
scored and evaluated. The dominant lethal assay generally involves treat-
ment of the male parent with single or multiple doses of the mutagen and
breeding the treated males with virgin females for eight weeks. Mated
females are sacrificed at mid-pregnancy and uterine analysis is made for
total implants, live implants, and dead implants, and compared with controls
to determine the incidence of dominant lethals.
Epstein et al. (1972), in a survey of 174 chemicals, tested epichloro-
hydrin for the induction of dominant lethal effects in mice. Male mice
(group of 10) were treated intraperitoneally with 150 mg/kg of epichloro-
hydrin (purity not given) and bred with untreated females for 8 consecutive
weeks (number of female mice per week not given). The dominant lethal
analysis revealed no differences in total implants and fetal deaths between
the experimental and control groups. However, details were not provided in
this report.
Sram et al. (1976) also reported negative results with epichlorohydrin
in the dominant lethal assay in mice. The test compound at concentrations
of 5, 10, and 20 mg/kg was injected intraperitoneally and at concentrations
of 20 and 40 mg/kg administered orally with acute (single dose) and subacute
(1 dose/day for 5 days) doses. No differences in the frequency of dominant
lethal mutations were noted between the experimental and control groups.
It should be noted that the dominant lethal assay may not be sensitive
to epichlorohydrin. The negative results in this assay may not necessarily
mean that the test compound is not mutagenic; it is possible that epichlor-
ohydrin is unable to reach mammalian germ cells in sufficient quantity to
cause dominant lethal effects or it may not reach the germ cells at all.
However, more information is needed before reaching such a conclusion that
epichlorohydrin is not a germ cell mutagen.
7.2.8.7 Sister-Chromatid Exchange Assay—The sister-chromatid exchange
(SCE) assay detects reciprocal exchanges induced by mutagenic agents between
sister chromatids of chromosomes. Epichlorohydrin was found to induce SCEs
in cultured human lymphocytes (White 1980; Carbone et al. 1981; Norppa et
al. 1981).
White (1980) studied the effects of epichlorohydrin on the frequencies
of sister-chromatid exchanges in the lymphocytes of two female healthy
adult donors. The lymphocyte cultures were exposed to epichlorohydrin as
follows: (1) cultures were exposed to 1 x 10"3, 4 x 10"4, 2 x 10~4, 1 x
7-89
-------
~4 ~4 -5
10 , 8 x 10 , and 4 x 10 M concentrations of epichlorohydrin for the
entire culture period of 73 hours, (2) cultures were exposed for the final
25 hours of cultivation with above concentrations of epichlorohydrin, and
(3) cultures were exposed for 2 hours (48-50 hours of cultivation) with 1 x
~3 ~4 -4 -4
10 , 4 x 10 , 2 x 10 , and 1 x 10 M concentrations of epichlorohydrin.
Chromosome preparations were stained with the fluorescence plus Giemsa
(FPG) technique of Perry and Wolff (1974) to differentiate sister chromatids.
Twenty to 30 metaphases were scored and the frequency was expressed as
SCE/cell. In cultures exposed for 73 hours, there were 14.8 + 0.73, 12.6 ±
0.79, 10.1 + 0.53 SCEs/cell indicating a dose-related response. The other
three concentrations, 1 x 10~3, 8 x 10~5, 4 x 10~4, and 2 x 10~4 M, yielded
no mitoses. The control frequency was 8.2 + 0.53 SCE/mM. Similar dose-
related .increases in SCEs were also noted for cultures treated for 25 hours
of cultivation was threefold higher (19.5 + 1.01/cell) at the concentration
—4
of 4 x 10 M, as compared to control frequency (6.6 + 0.49/cell) in the
absence of metabolic activation.
Carbone et al. (1981) demonstrated the induction of SCEs in cultured
_c — o
human blood lymphcytes with low concentrations of 1 x 10 , 1 x 10 , and
1 x 10 M of epichlorohydrin in the absence of metabolic activation. The
frequencies of SCEs were analyzed with the Student's t-test, and the results
— R —A
were significant at concentrations of 1 x 10 M (p<0.001) and 1 x 10 M
(p<0.05) compared to controls.
Norppa et al. (1981) also demonstrated the induction of SCEs in human
blood lymphocytes. Epichlorohydrin at concentrations of 0.05, 0.20, and
0.40 mM induced 8.4 + 0.4, 30.5 + 1.2, and 56.3 +2.8 SCEs/cell, respec-
tively. The solvent control frequency was 7.0 + 0.3/cell. There is a
clear dose-response relationship between the number of SCEs and the concen-
trations of epichlorohydrin. The data were analyzed using the Student's
t-test and found that the experimental groups exhibited statistical signi-
ficance over the control value.
7.2.9 Conclusions
Epichlorohydrin has been demonstrated to be mutagenic in both prokary-
otic and eukaryotic systems. This compound has been shown to be an active
inducer of gene mutations in bacteria, Neurospora, yeast, cultured mammalian
cells, and Drosophila. Epichlorohydrin was also effective in causing
sister-chromatid exchanges in human cells in vitro and preferential cell
7-90
-------
killing of repair-deficient bacteria. Chromosomal effects induced by
epichlorohydrin were detected both in vivo and in vitro mammalian assays.
The micronucleus assay, however, indicated a negative response of epichlor-
ohydrin presumably because the aberrations induced were reciprocal exchanges
that segregated without forming micronuclei. The dominant lethal assay in
mice also produced negative results. However, the assay may not be sensitive
enough to detect mutations other than gross chromosomal aberrations. Based
on the above weight-of-evidence, epichlorohydrin should be regarded as
mutagenic, thus having the potential to cause somatic mutations, which may
be involved in the etiology of cancer in humans. The concern is also
raised that epichlorohydrin may reach germ cells; however, additional
studies are required before concluding that epichlorohydrin is not a germ
cell mutagen in mammals.
7.3 'REPRODUCTIVE AND TERATOGENIC EFFECTS
7.3.1 Reproductive Effects
A qualitative assessment of the available data was conducted to deter-
mine whether epichlorohydrin has the potential to cause adverse reproduc-
tive or developmental effects. Six studies have been reviewed concerning
the effect of epichlorohydrin on the reproductive ability of male and
female rats and rabbits; on the development of offspring in the rat, mouse,
and rabbit; and on the semen of workers exposed to epichlorohydrin.
Hahn (1970) was the first to investigate antifertility effects in male rats
due to epichlorohydrin. Male Sprague-Dawley rats (quantity not stated)
were administered 15 mg/kg epichlorohydrin orally for 1-2 days. There was
no observed histologic change in the testes, epididymis, prostate, or
seminal vesicles after 12 days of exposure, nor was sexual libido or ejacu-
latory ability affected. However, temporary sterility was produced in the
males. After 1 week of exposure, the male rats were unable to impregnate
proestrous female rats, with the effect reversed 1 week after discontinua-
tion of treatment.
Cooper et al. (1974) studied the effects of epichlorohydrin and several
related compounds. Adult Wistar rats (five per group) were given epichlor-
ohydrin orally in suspensions of arachis oil at doses of 20-100 mg/kg and
then were sequentially mated to unexposed females for 10 consecutive weeks.
When given at 50 mg/kg/day for 5 consecutive days, epichlorohydrin rendered
male rats totally incapable of impregnating unexposed female rats. When
7-91
-------
animals were exposed to only a single dose of epichlorohydrin at 100 mg/kg,
fertility was reduced but not completely abolished (see Table 7-10). With a
single dose of 100 mg/kg epichlorhydrin, no histologic effects were observed
after 8 weeks. However, after 12 weeks, lesions were observed in the
efferent ductus, and large retention cysts were present in the ductuli
efferentes and proximal caput in 4 of 5 animals.
TABLE 7-10, THE EFFECTS OF EPICHLOROHYDRIN ON THE FERTILITY OF WISTAR RATS
Compound
al pha-Chl orohydri n
t
Epichlorohydrin
No. of Days
of Exposure
5
1
5
5
1
Daily
Dose
(mg/kg) Weeks:!
20
10
20
50
100
0
0
0
0
0
Average Weekly
2
0
0
0
0
4
3
7
0
0
0
3
4
9
0
11
0
4
b
3
0
11
0
4
6
0
0
2
Litter
7
0
0
2
8
0
0
4
Size
9
0
0
2
10
0
0
3
Five Wistar rats used for each dose level.
Source: Cooper et al. (1974)
The Toxicology Research Laboratory, Dow Chemical Company (John et al.
1979) conducted a three-part study evaluating the reproductive ability of
the male rabbit, the male rat, and female rat after exposure to epichloro-
hydrin. Groups of 10 male rabbits (New Zealand), 30 male rats (Sprague-
Dawley), and 30 female rats (Sprague-Dawley) were exposed for 10 weeks by
inhalation (6 hours/day, 5 days/week) to 0, 5, 25, or 50 ppm production
grade epichlorohydrin supplied by Dow Chemical Company (analyzed as 98.8%
pure by weight with 0.03% propylene dichloride, 0.08% cis-l,3-dichloropropane,
0.07% 2,3-dichloropropene, and 0.01% beta-chloroalkyl alcohol). The quality
and quantity of rabbit semen'was evaluated every week for 2 weeks prior to
exposure, every week for 10 weeks during exposure, and every other week for
10 weeks after exposure (see Table 7-11). After 10 weeks of exposure, the
male rabbits were mated to untreated female rabbits in estrus; the females
7-92
-------
(0
00
cS
LU
g
UJ
a
z
^
to
a
i
a>
ZJ
O)
to
t.
Q_
•o
C
to
. — s
If)
01
"rO
s
If)
to
•3
*(—
T3
to
to
CU
Ivf
Z
m
OJ
u
a>
a.
oo
ol
01
01
3
O
CM
••
01
cu
V)
to
•a
ID
•>
to
•a
LO
(_
13
0
ID
C
O
^3
to
"io
c:
hH
01
5-
O
CL
X
UJ
g_
o.
o
LO
•a
c
• IO
trt"
CM
Lf)
O
"ai
01
Ul
O
Q.
X
CU
t-
o
cu
01
S-
01
CL
U)
10
s-
01
to
E
CU
o
ro
•a
c:
cu
E
O
CO
^
w
JZ)
ja
to
01
"to
o
i—i
t/i
cu
Nl
LO
a.
0
L
O
CM X X
S
r-.
J IO
Qi rH X
00
0
CL ID
LU
£
C
C
CO
3 X X
rH
rH X
O
rH X X XX
CT> X
00 X
r~* x x
UJ
§D
OO UO X
o
LU ID 3<
* X X
n x
CM X X
iH X
OX X
rH
LU I X
=
oo
O OJ
Q- 1 X
LU
LU
O_
U)
^
Ol
cu
^-x
U)
/-^ O)
w cz
O) T-
£Z 4J
•r- to
•P E
to ^— '
*— ' a>
01
W I — •
cz - cu to
0 r- E
•.- ro cu
•(-> S u-
to
Z3
1 J?
00 CU -^
1— 0) r-
1-1 CZ C ••-
QQ 01 -r- OO 4J
OQ E 4-> H- !_
a: oo s: cz. u_
to
a.
E
0
o
_
to
u
£
cu
r—
O
Q
„
£
0
to
s_
o
fO
J
.c:
u
s_
ttt
0)
to
cu
C£,
§s
o
o
u
X
0
J—
to
CD
OT
rH
'(O
-P
CU
C
.C
0
f-D
-------
were subsequently induced to ovulate with human chorionic gonadotrophin.
On day 28 of gestation, the female rabbits were sacrificed, and the number
of corpora lutea, implantation sites, and resorbed fetuses were recorded.
To evaluate fertility in male rats, males were exposed to epichlorohydrin
for 10 weeks; then the exposed male rats were mated to two unexposed female
rats for 1 week of cohabitation initiated on the 2nd, 4th, 7th, 10th, 12th,
and 20th week of study. The untreated female rats were sacrificed 12 days
after the last day of cohabitation, and examined for the number of corpora
lutea, implantation sites, and resorption sites. To evaluate fertility in
female rats the animals were exposed for 10 weeks and then were allowed to
mate with two different unexposed male rats for 2 consecutive 5-day periods.
The stage of estrus was evaluated in daily vaginal smears until sperm was
observed in the vagina. The date of delivery, the number of live and dead
pups, and observations of external abnormalities were recorded at birth.
In this study (John et al. 1979), the body weights, clinical chemistry,
number of corpora lutea, and semen parameters were evaluated statistically
by one-way analysis of variance and Dunnett's test. Preimplantation loss
and numbers of resorptions were analyzed by the Wilcoxan test modified by
Haseman and Hoel. The fertility index was analyzed by Fisher's exact
probability test.
In this study, exposure to epichlorohydrin produced signs of toxicity
in rats. Male and female rats exposed to 50 ppm epichlorohydrin but not 5
or 25 ppm, gained significantly less weight during the 10-week exposure
period than the controls. Male and female rats exposed to 25 ppm had
slight increases in both absolute and relative kidney weights, whereas
those exposed to 50 ppm had significant increases. The livers of males at
all exposure levels and the livers of females exposed to 50 ppm were slightly
but not statistically heavier than the controls. In addition, histopatho-
logic changes were observed in the nasal turbinates of male and female rats
exposed to 25 and 50 ppm epichlorohydrin. It is possible that both control
and experimental rats were ill prior to treatment, since white blood cell
counts were elevated during the preexposure period with symptoms of sialoda-
cryoadenitis observed during the first 2 weeks of exposure in all groups.
In male rats, 25 and 50 ppm epichlorohydrin markedly affected the
ability of the animals to impregnate unexposed female rats. After the
females were mated with the exposed males, there were significantly fewer
7-94
-------
implantation sites in female rats mated to males exposed to 25 and 50 ppm, but
not 5 ppm epichlorohydrin. The matings conducted during weeks 12-20 of the
experiment (2-10 weeks after discontinuation of exposure) did not result in
significant reduction in implantation sites, suggesting that this effect was
reversible. There was an increase in preimplantation loss in females mated to
males exposed to 25 and 50 ppm epichlorohydrin. This preimplantation loss was
observed in matings (at 25 and 50 ppm) conducted during the exposure period,
while preimplantation losses in matings conducted during the recovery period
were observed only in groups exposed to 50 ppm. A reduction in the number of
corpora lutea was observed in females mated with males (i.e., those exposed to
50 ppm epichlorohydrin) during weeks 2, 4, 7, and 10 but not weeks 12-20.
In studies on the female rat (John et al. 1979), exposure to 5, 25, or 50
ppm epichlorhydrin did not affect the animal's ability to become pregnant, the
length of gestation, litter size, survival indices, sex ratios, or .the incidence
of malformations in the offspring.
In the study using male rabbits, signs of toxicity were observed in
animals exposed to 50 ppm epichlorohydrin (John et al. 1979). Male rabbits at
this dose level gained significantly less weight than controls during the
10-week exposure period, with two rabbits dying spontaneously or sacrificed
due to moribund conditions. Epichlorohydrin exposure apparently did not alter
semen volume, sperm concentrations, motility, or morphology in rabbits.
During the 10th week of the experiment, each male was mated to unexposed
females. There were no dose-related alterations in fertility, implantations,
corpora lutea, or resorptions in the unexposed female rabbits.
7-3.1.1 Male Clinical-Epidemiologic Investigations—Milbv et al. (1981)
conducted a clinical-epidemiologic investigation of testicular function in
chemical plant workers occupationally exposed to epichlorohydrin. Men working
on the Shell Chemical Corporation plants in Deer Park, Texas (plant A, epichlo-
rohydrin production since 1948), and Norcos Louisiana (plant B, production
since 1955), was included in this study. Semen samples were obtained as well
as blood samples for measurement of follicle stimulating hormone (FSH) and
luteinizing hormone (LH). There was some attempt at evaluating the intensity
of exposure (exposure estimates from industrial hygiene survey, personal
exposure knowledge, plant employment records) from the men in plants A and B.
7-95
-------
It was not possible to evaluate unexposed workers for use as controls
from either of the two plants. The control population was selected from
chemical plant workers previously evaluated by the authors in other studies.
The men used as controls had no history of exposure to testicular toxicants
(90 men were used as controls).
The results of this study indicated that the frequency distribution of
sperm in the semen collected from 44 men in plant A and 87 men in plant B
did not significantly differ from that of the control populations. There
was no association suggestive of deleterious effect with either duration or
intensity of exposure. In addition, there were no significant differences
in hormone concentrations (FSH, LH, testosterone). However, these results
do not conclusively determine whether epichlorohydrin produces adverse
testicular effects in humans. It is recognized that this type of study has
major inherent weaknesses due to confounding factors related to the parti-
cipation.of all men potentially exposed to epichlorohydrin. In addition,
data such as marital status of the men, age, and number of children were
not available to the authors. Therefore, it was difficult to assess whether
the population studied was a true representative population of all male
workers exposed to epichlorohydrin or whether the control population repre-
sented a true distribution of all fertile men. The most critical weakness
with this study was that there were no actual exposure measurements; the
exposure intensity and duration was estimated by job category or by a
combination of judgments based upon industrial hygiene sampling and the
investigator's appraisal of the work situation.
7.3.2 Teratogenic Effects
The teratogenic potential of epichlorohydrin has been evaluated in two
studies (Pilny et al. 1979; Marks et al. 1982). Pilney et al. (1979)
evaluated a small number of rats and rabbits to establish a dose for maternal
toxic effects (tolerance study) and then conducted a teratology study using
larger numbers of rats (Sprague-Dawley) and rabbits (New Zealand). For the
tolerance study, five or six pregnant rats and five pregnant rabbits v/ere
exposed to 0, 25, 50, or 100 ppm epichlorohydrin (containing by weight
99.8 percent epichlorohydrin, 0.11 percent 2,3-dichloropropene, 0.03 percent
cis-l,3-dichloropropene, and 0.01 percent beta-chloroallyl alcohol) admin-
istered by inhalation and analyzed by the Dow Chemical Company, Freeport,
7-96
-------
Texas. Concurrent controls were exposed to filtered air. Exposure to 50
and 100 ppm epichlorohydrin in the tolerance study produced signs of maternal
toxicity (decrease in maternal weight gain and decrease in intra-abdominal
adipose tissue). In the groups exposed to 100 ppm epichlorohydrin, three
of the six animals had only resorption sites, two animals had no implanta-
tion sites, and the one remaining had normal appearing fetuses. The rabbits
exposed to 50 and 100 ppm epichlorohydrin had signs of maternal toxicity
(decrease in maternal weight gains, increased respiratory tract infections),
but no fetal loss. Based upon the results of the tolerance study, the
teratology study was conducted with doses of 2.5 and 25 ppm epichlorohydrin
to avoid problems with severe maternal toxicity. The teratology study
utilized 43-66 rats and 20-25 rabbits. The rats and rabbits were exposed
for 7 hours/day on days 6-15 or 6-18 of gestation, respectively.
The data in this study (Pilney et al. 1979) were analyzed statistic-
ally using the Wilcoxan test modified by Haseman and Hoel for evaluating
frequency of resorption among litters and fetuses. Analysis of percent
pregnant, maternal survival rate, and other incidence data were made by
Fisher's exact probability test. Analyses of fetal body weight, body
length, maternal weight gain, and maternal organ weights were made by
analysis of variance. Group means were compared to control values using
Dunnett's test. The level of significance was chosen at p<0.05.
Pilney et al. (1979) reported signs of maternal toxicity in rats
exposed to 25 ppm epichlorohydrin, but not to 2.5 ppm epichlorohydrin.
Rats exposed to 25 ppm weighed less (statistically significant) than the
control animals throughout the exposure period, and consumed significantly
more water. There were no signs of maternal toxicity observed in rabbits.
There were no alterations in pregnancy rates, number of litters, corpora
lutea, implantation sites, resorption site, numbers of dead fetuses, fetal
body weight or crown-rump length, or incidence of malformation in either
rats or rabbits.
Marks et al. (1982) evaluated the teratogenic potential of epichloro-
hydrin administered to rats and mice. Epichlorohydrin (laboratory grade
Fisher Scientific Co.) was administered by gastric intubation in doses of
0, 40, 80, and 160 mg/kg/day to 14-35 outbred albino rats (CD, 176-200 g).
Epichlorohydrin was administered by gastric intubation in doses of 80, 120,
7-97
-------
and 160 mg/kg/day to 25-49 outbred albino mice (CD-I, 60-90 days old). The
chemical was dissolved in cottonseed oil and administered to both rats and
mice on days 6-15 of gestation.
In this study (Marks et al. 1982), the data were analyzed statistic-
ally to evaluate differences between the groups using the Mann-Whitney
U-test or Student's t-test. Differences in the dose-response relationship
were evaluated using Jonckheere's test. Two-tailed analysis was performed
and the level of significance was chosen at p<0.05.
In this study (Marks et al. 1982), both the rats and mice showed signs
of maternal toxicity at the two highest doses administered. Rats admin-
istered 160 mg/kg/day had significantly greater increases in liver weight
and 3 of 27 rats died; and at 80 mg/kg/day the epichlorohydrin caused a
significant reduction in the average weight gain during pregnancy. In mice
administered 160 mg/kg/day, 3 of 32 mice died, and there was a significant
decrease'in fetal weights. In addition, there was a significant increase
in the average maternal liver weight. There were no dose-related increases
in soft tissue or skeletal malformations.
7.3.3 Summary and Conclusions
Epichlorohydrin has been evaluated in six studies for its potential
for causing (1) adverse reproductive effects in female rats, (2) adverse
reproductive effects in male rats and rabbits, (3) adverse spermatogenic
effects in humans occupationally exposed to epichlorohydrin, and (4) adverse
developmental effects in rat, mouse, and rabbit concepti.
In females, epichlorohydrin has not been adequately investigated to
determine if there is a potential for reproductive hazard. Only one study
(John et al. 1979) has investigated reproductive effects in female rats.
Animals were exposed for 10 weeks and the possible effect on future genera-
tions was not investigated. No adverse reproductive effects were observed
in this study (no alteration in pregnancy rate, gestation length, litter
size, survival indices, sex ratio, or external alterations). However,
additional studies in the future should be conducted to firmly establish
that there is no potential for harmful effects.
The data on males indicate that epichlorohydrin possesses the ability
to alter male fertility. Three investigations using rats demonstrate that
epichlorohydrin can cause sterility (Hahn 1970, Cooper et al. 1974, John et
al. 1979). In most cases this effect is reversible (Hahn 1970; John 1979);
7-98
-------
but with longer durations of exposure or higher concentrations, this effect
may be irreversible (Cooper et al. 1974). The effect on fertility was
observed with (John et al. 1979) and without, (Hahn 1970; Cooper et al.
1974) other signs of toxicity (i.e., losses in the animal's body weight).
In males occupationally exposed to epichlorohydrin no alterations in
sperm concentration were observed when the frequency distribution of sperm
of the exposed population was compared with that of a control population
(Milby et al. 1981). In addition there was no association between inten-
sity or duration (LH, FSH, testosterone). However, since there was no
concurrent control, information on reproductive history of men, or measure-
ment of actual exposures, the full potential for adverse reproductive
effects in humans cannot be adequately assessed from this study. The
sensitivity of this type of clinical-epidemiologic study in detecting
potential reproductive toxins has not yet been determined. Although this
type of study has been used to establish a correlation between dibromo-
chloropropane (DBCP) exposure and male sterility, it should be noted that
the success of this type of study in establishing an association was depen-
dent upon the severity of the effect. In the case of DBCP, the effects on
male fertility were quite severe, with some men unable to produce sperm
(azospermia) 4 years after the discontinuation of exposure.
Epichlorohyrin has been investigated for its potential to alter the
development of the conceptus in rats, mice, and rabbits in two studies
(Pilney et al. 1979, Marks et al. 1982). No malformations were produced
even at maternally toxic doses. A reduction in fetal weights in mice were
reported; however, this was observed only at doses that caused increases in
liver weights in the dams (Marks et al. 1982).
In conclusion, the data available to date indicate that epichlorohydrin
has the potential to produce adverse reproductive effects in the male, but
not in the developing conceptus. Epichlorohydrin1s ability to affect
adversely male reproduction might be expected since its metabolite, alpha-
chlorohydrin, is known for its antifertility properties. Alpha-chlorohydrin
is thought to be produced from epichlorohydrin by the action of epoxide
hydratase (Jones and O'Brien 1980) (see also section on metabolism). The
antifertility effects of alpha-chlorohydrin have been studied extensively,
7-99
-------
and alpha-chlorohydrin has been shown to cause adverse reproductive effects
in a number of animal species (small laboratory rodents, large domestic
animals, and nonhuman primates) (Gomes 1977). Both epichlorohydrin and
alpha-chlorohydrin produce the same urinary metabolites in rats (Jones
et al. 1969); however, alpha-chlorohydrin appears to be more potent than
epichlorohydrin in producing adverse male reproductive effects.
7-100
-------
8. SYNERGISM AND ANTAGONISM'AT THE PHYSIOLOGICAL LEVEL
No studies on the synergistic or antagonistic effects of epichlorohydrin
in combination with other chemicals or conditions in humans were found in
the available literature. There are few animal studies on synergistic or
antagonistic effects and the studies examined are limited in scope.
Lukaneva and Rodionov (1978) studied the combined effects of epichlor-
ohydrin and cholesterol on the development of atherosclerosis in rabbits.
Five groups of rabbits (experimental details not provided) were used. The
first two groups were administered epichlorohydrin orally at concentrations
of 3.44 mg/kg and 17.2 mg/kg, daily for 7 months and 3.5 months, respectively.
The second 'two groups received these two doses of epichlorohydrin plus 200
mg/kg of cholesterol according to the same schedule. The fifth group
received 200 mg/kg of cholesterol for 3.5 months. Electrocardiographic
monitoring of the treated animals was performed; however, no information
was provided as to which animals were monitored. Serum lipid levels were
assayed at 3.5 and 7 months. Rabbits were sacrificed at 3.5 and 7 months
(number of animals in each sacrifice not provided), and the hearts were
examined grossly and microscopically for changes.
The authors stated that the animals administered epichlorohydrin alone
at doses of 17.2 mg/kg for 3.5 months had "only a few individual" electro-
cardiographic changes. These were not described. However, epichlorohydrin
administered alone at doses of 3.44 mg/kg for 7 months led to increased
atrioventricular conductivity and evidence of metabolic and functional
changes in the myocardium including myocardial hypoxia.
Combined administration of epichlorohydrin at either dose and cholesterol
led to a number of electrocardiographic changes characteristic of stage I
atrioventricular block, and other disorders such as abnormal conductivity
of the right atrium (deformation of the P wave), and increased electrical
potential of the left atrium. These changes were more evident as exposure
continued for longer periods (i.e., 7 months).
8-1
-------
Epichlorohydrin administration led to an increase in the concentration
of free cholesterol in the blood and hyperlipidemia (excess lipids in the
blood). The administration of both epichlorohydrin at 17.2 mg/kg and
cholesterol for 3.5 months led to higher concentrations of phospholipids
and esterified cholesterol than did both epichlorohydrin at 3.44 mg/kg and
cholesterol for 7 months; however, 'levels of free cholesterol and free
fatty acids were higher in the animals treated with epichlorohydrin at 3.44
mg/kg for 7 months.
Shumskaya et al. (1971) examined the interaction between inhalation
exposure to epichlorohydrin and subsequent exposure to cold. Three groups
of 60 albino male rats (weighing approximately 180 g) were exposed to
epichlorohydrin for a single exposure at concentrations of 0.007, 0.02, and
0.35 mg/1 (1.85, 5.28, and 92.5 ppm) for 4 hours. A fourth group was not
exposed and served as a control. After exposure, half of the animals in
each group were placed in a cold chamber at 5°C for 2 hours. Several
parameters were examined immediately after exposure and then within 24
hours after exposure. These included body temperature, oxygen demand,
bromsulphalein retention, and levels of blood urea nitrogen, serum sulf-
hydryl, and liver glycogen. In addition, urinalysis was done. No descrip-
tion of the clinical methods of analysis was provided by the authors.
Terminal organ weights were recorded 24 hours after exposure. Those animals
exposed to cold stress showed fewer deviations from normal values than
animals not exposed to cold stress. Body temperature decreased in rats
immediately after exposure to epichlorohydrin and exposure both to epichlor-
ohydrin and cold. After 24 hours, body temperatures were normal. Oxygen
demand decreased in both cold-stressed and noncold-stressed animals. For
some measurements such as bromsulphalein retention, the increases were
larger in the cold-stressed rats. Urine volumes and chlorides increased
and specific gravity decreased in the noncold-stressed animals only. No
significant changes were observed in blood urea nitrogen, serum sulfhydryl,
and liver glycogen levels in either cold-stressed or noncold-stressed
animals. The parameters that showed significant variations are shown in
Table 8-1.
8-2
-------
TABLE 8-1. SUMMARY OF STUDY MEASUREMENTS AFTER EXPOSURE TO EPICHLOROHYDRIN
AND SUBSEQUENT EXPOSURE TO COLD3
Concentration (DDID)
Measurement
Control
1.84
5.28
92.5
Ambient Cold
Body temperature, °C
After exposure
At 24 hb
Oxygen demand, ml/h
After exposure
At 24 hb
Kidney weight, g
At 24 h°
Bromosulphalein
Retention after
exposure
Uri ne
24- h volume, ml
Total protein, g/100 ml
After exposure
36.6/36.7
36.9/36.9
426/414
410/398
0.83/0.86
0.1/6.5
2.6/4.3
7.08/6.81
36.2/36.3
37.1/37.2
331/348
318/350
0.87/0.83
1.4/4.2
4.9/2.6
6.78/7.10
35.5/36.1
36.7/36.9
275/264
281/278
0.87/0.84
2.5/3.5
4.0/3.0
6.90/7.44
33.4/33.5
36.8/36.6
235/251
244/247
1.02/1.01
8.6/18.9
4.4/4.6
7.80/8.01
Source: Shumskaya et al. (1971).
°0ne day after exposure.
After completion of dosage or cold exposure.
Grigorowa et al. (1977) investigated the effects of repeated epichlor-
ohydrin inhalation exposure followed by repeated exposure to elevated
ambient temperatures. Four groups of 30 male albino rats (strain and age
unspecified) that weighed 220-260 g each were treated as follows:
Group 1
Group 2
Group 3
Group 4
2
30 mg/m epichlorohydrin--4 hours/day at 20°C
30 mg/m epichlorohydrin—4 hours/day at 20°C followed
by 2 hours/day exposure to heat stress at 35°C and 50
percent relative humidity.
no exposure to epichlorohydrin, placed in chamber~4
hours/day at 20°C
no exposure to epichlorohydrin, placed in chamber—4
hours/day at 20°C followed by 2 hours/day exposure to
heat stress at 35°C and 50 percent relative humidity.
8-3
-------
The experimental animals were exposed daily for a 4-week period.
Following exposure, the animals were maintained for an additional 2 weeks
for observation. After the first day of exposure, 10 rats from each group
were killed on days 10, 30, and 45, and the tissues and organs were necrop-
sied and examined for abnormalities. The liver, kidneys, lungs, adrenals,
testes, and thymus were sectioned and examined for microscopic changes.
Urine was examined for aminopeptidase, albumin, volumes eliminated, excretion
of phenol red, concentrations of phenol red, and sodium and potassium
levels. Catalase activities in whole blood and in kidney hemogenates were
determined. Aminopeptidase, glutamic-pyruvic transaminase and glutamic-
oxaloacetic transaminase activities were measured in serum. Frozen sections
of tissue were also examined for peroxidase, succinate dehydrogenase, and
alpha-naphthylacetate activities.
In the group exposed only to epichlorohydrin (group 1), there was a
slight increase in catalase activity in blood, a decrease both in the
concentration and excretion of albumin in the urine, and a marked decrease
in the production of urine 10 days following the initial exposure. Urine
production returned to relatively normal levels in all groups 30 and 45
days following the initial treatment. Significant decreases (p <0.01) in
phenol red concentrations were observed in the animals exposed to both
epichlorohydrin and heat stress (group 2). The phenol red concentrations
for the other dosed groups were comparable to those for the controls.
There was also a decrease in the rate of phenol red elimination in the
urine of group 2 animals, 30 days following the initial exposure. Histo-
pathologic examinations of the lungs, kidneys, and liver showed no sig-
nificant differences between heat-stressed and nonheat-stressed animals.
The animals exposed to epichlorohydrin (groups 1 and 2) showed renal toxicity
including edema, degeneration of the tubular epithelium, and glomerular
changes. These changes were more evident after 4 weeks of exposure. The
authors concluded that healt enhanced the toxicity of epichlorohydrin.
In an additional study, Grigorowa et al. (1977) examined the effect of
heat stress on the lethal concentration of epichlorohydrin. Groups of 20
male mice (weighing 18-26 g) and 20 male rats weighing (230-270 g) were
exposed to epichlorohydrin by inhalation for either 2 hours (mice) or 4
hours (rats). The strains of rodents used were unspecified. Half of the
8-4
-------
animals in each group were then placed in a heated chamber for 45 minutes
at 35°C. The remaining animals were kept at room temperature (18°C). Two
similar, nonexposed control groups of mice and rats were either placed in
the heated chamber or kept at room temperature. The heated chamber had a
relative humidity of 35-50 percent. The authors reported that the rats
were more sensitive to heat stress than were the mice. However, care
should be used in evaluating this study as the confidence limits of the
respective LC5Q values are large and overlap considerably. The LC5Q values
obtained for epichlorohydrin are shown in Table 8-2. Information on
syngergism and antagonism at the physiological level was limited to a few
fragmentary animal studies. Epichlorohydrin administered orally in combination
with cholesterol affected heart functions and blood lipid levels in rabbits
(Lukaneva and Rodinov 1978). Rats that were administered epichlorohydrin
by inhalation for 4 hours and subsequently cold-stressed, showed increased
bromsulphalein retention (measured 24 hours after treatment) compared with
noncold-stressed animals; there were no significant differences between the
two groups in body temperature, oxygen demand, kidney weight, urine volume,
and total blood protein, urea nitrogen, and serum sulfhydryl levels (Shumskaya
et a!., 1971). Another study of rats indicated that heat (35°C) enhanced the
toxicity of epichlorohydrin (Grigorowa et al. 1977).
TABLE 8-2.
Species
Rat
Mouse
THE EFFECT OF HEAT
IN THE
Condition
No heat
With heat
No heat
With heat
STRESS ON THE LD5o OF
RAT AND MOUSE3
(mg/1)
2.40
2.20
3.00
4.00
EPICHLOROHYDRIN
Confidence Limits
0.87-6.56
0.67-7.18
1.79-5.02
2.57-6.22
Source: Grigorowa et al. (1977).
8-5
-------
-------
9. ECOSYSTEM CONSIDERATIONS
9.1 EFFECTS ON MICROORGANISMS AND PLANTS
9.1.1 Effects on Microorganisms and Lower Plants
There were very few studies available .that reported the effects of
epichlorohydrin on microorganisms, but all indicate that toxic effects
would occur at concentrations in excess of a few milligrams per liter.
None of the studies, however, was performed in a closed system with a
constant, measured concentration of epichlorohydrin. Moreover, when released
into natural habitats, epichlorohydrin would not be expected to persist
because of its tendencies to hydrolyze and volatilize.
Toxicity threshold concentrations (i.e., lowest inhibitory concentra-
tions) of epichlorohydrin for Scenedesmus quadricauda (green alga), Microcystis
aeruginosa (blue-green alga), Entosiphon sulcatum (flagellated protozoan).
and Pseudomonas putida (aerobic bacterium) were 5.4, 6.0, 35.0, and 55.0
mg/1, respectively (Bringmann and Kuhn 1976, 1980). All of the species
were studied by comparable procedures to determine the minimum toxicant
levels that inhibited cell multiplication. Inhibition was measured by
turbidimeter for algae and bacteria and by electronic cell counter for
protozoa. For each organism, three parallel dilution series were prepared,
and the toxicity threshold was estimated graphically by plotting cell
numbers (per ml) against log concentration of epichlorohydrin (mg/1). Test
durations were 16 hours for bacteria, 72 hours for protozoa, and 168 hours
for the two algae. These studies may indicate toxicity thresholds but are
of limited usefulness because the epichlorohydrin concentrations were not
measured and chemical purity was not specified.
Kolmark and Giles (1955) investigated the mutagenicity of epichloro-
hydrin to an adenine-requiring strain of the purple fungus, Neurospora
crassa, and also observed toxic effects. Conidia were treated with 0.15 M
epichlorohydrin (13.88 g/1). Survival decreased to 40 percent and 0.7
percent with treatment periods of 45 and 60 minutes, respectively. Chemical
purity was not specified, and statistical analysis of the data was not
indicated.
Other studies indicate that epichlorohydrin is mutagenic to bacteria
and fungi at levels well above a few milligrams per liter (see Section 7.2).
9-1
-------
9.1.2 Effects on Higher Plants
The only study available for higher plants reported the effects of
treating Eucalyptus seeds with epichlorohydrin.
Epichlorohydrin was one of five chemical mutagens used to treat the
seeds of three species of Eucalyptus (Bandel 1971). Groups of 400 seeds
from each species were treated with 0.15 percent (1.5 g/1) and 0.30 percent
(3.0 g/1) solutions for 2 or 4 hours each. The treated seeds were then
sown in wooden boxes with soil and sterilized manure. Sixty days after
sowing, the number of live plants was determined (Table 9-1). Results
indicated decreased survival with increasing concentration or exposure
period. In addition, E. citriodora appeared to be more resistant than the
other two species. Chemical purity was not specified, and statistical
analyses were not provided.
TABLE 9-1. PERCENT SEEDLING SURVIVAL 60 DAYS AFTER SOWING EUCALYPTUS
SEEDS TREATED WITH EPICHLOROHYDRIN SOLUTION3
Treatment
Concentration (%)
0.00
0.15
0.15
0.30
0.30
Hours
-
2
4
2
4
Percent
E. tereticornis E.
100.00
95.07
26.91
28. 08
0.00
Survival
citriodora
100.00
91.84
60.45
77.85
0.88
E. maculata
100.00
98.84
41.87
33.26
0.00 •
aSource: Bandel (1971).
9.2 BIOCONCENTRATION, BIOACCUMULATION, AND BIOMAGNIFICATION
No experimental data were found in the literature on the bioconcentra-
tion (direct from the water), bioaccumulation (from food and/or water), or
biomagnification (through the food chain) of epichlorohydrin. However, the
properties of epichlorohydrin (including its octanol/water partition coef-
ficient (P), susceptibility to aqueous hydrolysis, and volatility) indicate
a low likelihood for accumulation in aquatic organisms or food chains.
Several workers have published methods that can be used to estimate
bi concentration. Biconcentration factors (BCF) can be derived from
either water solubility (Chiou et al. 1977, Lu and Metcalf 1975) or log P
9-2
-------
(Neely et al. 1974, Veith et al. 1980) (see Appendix D). Log P was estimated
to be 0.26 + 0.04 according to the method of Hansch and Leo (1979) (see
Appendix C). The log BCF values for epichlorohydrin estimated by these
methods range from -0.032 to 0.968. Log BCF values less than 2 indicate a
low bioconcentration potential (Kenaga 1980).
9.3 EFFECTS ON AQUATIC ANIMALS
Limited data were available to indicate the effects on aquatic biota.
The only aquatic toxicity studies found in the literature were laboratory
tests performed under static conditions, and the epichlorohydrin concentra-
tion was measured in only one of them. In assessing the available data,
one must consider the compound's environmental fate (see Section 3.5.1).
Epichlorodydrin that is released into natural waters is not expected to
persist beyond a few days because of its general reactivity and its tendency
to hydrolyze and/or volatilize. Additional factors affecting the results
of epichlorohydrin toxicity tests relate to experimental conditions (e.g.,
temperature, pH, water hardness, chemical synergism, dissolved oxygen, and
disease) (U.S. EPA 1975). Information on epichlorohydrin, however, is too
limited to examine the effects of these parameters.
The acute lethal effects of epichlorohydrin have been reported for
four fish and one invertebrate (see Table 9-2). Static median lethal values
ranging from 18 to 35 mg/1 were reported. In only one of the tests, however,
was the actual epichlorohydrin concentration measured. There was no information
on subchronic or chronic exposures or flow-through tests found in the literature.
9.3.1 Freshwater Fish
Toxicity information on epichlorohydrin was found for three warm water
fish; there was no toxicity information available on cold water fish. In a
bluegill study (Dawson et al. 1977), the test fish were obtained from
commercial hatcheries and held in 114-liter aquaria for 14 days at 23°C
prior to testing. During this period, the fish were fed an unspecified
"commercial fish food," treated to prevent disease, and'maintained in a
minimum water volume of 1 liter/gram of fish. Test fish were selected only
from those holding tanks showing less than 5 percent mortality. Aeration
was not used during the initial 24-hour test period. Dissolved oxygen was
measured daily, and dead fish were counted and removed daily. Toxicant
levels were not measured analytically in this static test. The acute
toxicity results for bluegill are summarized in Table 9-3. The death rate
for the control fish was low at 1.3 percent.
9-3
-------
UJ
i
CO
UJ
t—
5
UJ
1— 1
K—
*—**->
UJ
o
03
,3^
at
o
I—
ra
CO
I—
£
5
EG
g
i2
CO
»r™
U
CU
a.
CO
^-s
P en
ert en
^S ^*
• *
*ra ra
cu cu
C *0)
O •'-
CO "O
ra '£
O CO
03
C
re
ai
T3
T3
c
ra
cu
cz
3
•-3
J^-
CTl
^,
"ra
cu
c
o
CO
ra
CD
Sl .. D5 -r-
CU ___
"S/~> "S •-
W «^— ^iT
2 rain
r— re to s- co a. s
r— -CO 3i — ro CL-H
cu cj co cu -»J ra
S -ro ro >
en cj cu cu n
cz • *— ' E ?—
•r-p-
i co jz: c-
4->CO
-S co
1) IZCM
C •!-
r- CO
•acore +j.r-<4- -aco
CU * *r- £- CU CU CD
4->I»-^- ST3T3 -U C
u --. >> u -a
331 C3) >,J=?>, 3 S-
•a CXB -So*—? -a ra
^—
"(0
CO
re
CU
CO
cz
raco
ocz>
4->
o in +J o ra o c ||
o ..in coi — u ucx> ra
S- J=-
4-> CU CO >>U S -US.
CO -tJ CO 1— -f- CU CO
cu re cu cz Q..C cu 3:
h-ScZ C3CUU 1— Q.
cz> ro
rH CM
in JD.
ro co
CM
11 I.
S e
CJ —1
— J t—
ID >=«•
—» ^^
CO CO
rH
+1
CO £j
CM CM
CO
S- CO
JZ 4-*
u re
£ &
o ra
ra
S co
3
O£r- CO JZ-r-
UJr— V- CO CO
1 — «r- m T- CO
—>
4-) !_
CO O)
CZ
i— i O.
=: CO
CZ)
I— 1
°4
II
0
CJ
^
01
^
CO
CZ)
CM
cu
"O /•" "•*
•i- ra
CO CZ
J--r-
CUi—
.1^ S5
-s CO CU
n JD
3 £-
-> cu ra
D OS +J-r-
z LJJ ra TD
3 1 — S-i-
^ S'^S
.
u 3:
a.
cz
•a
CU !-
+-> 03
u +->
3 re
I!.
S4S
+J CU
CO CU
cu s-
1—1-
o
CM
a
CO
II
•o
IA
CJ
— 1
CM
ID
CO
CZ)
CM
/ — s
re
c
re'
H
re
uj cu ra
fec^'E
OS S- JZ
ca cu a.
LU •»-> re
UJ
^»
t— t
CO
cz
o
5
re
cz
cu
u
EZ
O
CJ
"D
3
CO
re
cu
II
• »,
CO
o
ra
i ^
c
cu
u
0
CJ
XI
s_
3
CO +J
re ••-
s **—
cz — J
cu
II u
Z3 ra
.. a>
•i- "o
2 e
co re
" "S
CO S
re -Q
9-4
-------
Table 9-3. The
Bluegi
Acute Toxicity of Epichlorohydrin to
ill and Tidewater Silverside Fish
Initial
Concentratio
Species (mg/1)
Bluegi 11
(Lepomis
macrochirus)
Tidewater
silverside
(Menidia
beryl! ina)
56
42
37
32
10
32
18
10
>n Percent Survival After
.24 h
0
50
100
100
100
100
100
90
48 h
—
0
90
90
100
30
90
90
72 h
—
—
80
80
75
0
70
90
96 h
—
__
60
70
75
*• »
50
90
Best Fit
96-h LC,n
(mg/1)50
35
18
Source: Dawson et al. (1977).
The goldfish study (Bridf et al. 1979b) was a static bioassay performed
using the methodology published by the American Public Health Association
(APHA 1976). The study was done without aeration using tapwater in 25-liter
aquaria. The chemical composition of the aged tapwater was determined, and
epichlorohydrin concentrations were measured before and after the test.
Juhnke and Ludemann (1978) reported the static acute toxicity of
epichlorohydrin to the ide (golden orfe), a species introduced from Europe
and established in U.S. waters. The 48-hour LC5Q value was 24 mg/1; 0 and
100 percent mortality occurred at 12 and 35 mg/1, respectively. The bio-
assay was conducted according to the method of Mann (1976). Ten fish (0.3
g, 5-7 cm) were exposed for 48 hours to epichlorohydrin (nominal levels) in
tapwater (pH 7-8, hardness 268 ±54 mg/1) at 20 ± 1°C. Although composi-
tion of the tapwater was not completely specified, the experimental pH,
hardness, and temperature values were within the range of values likely to
be found in the natural environment.
9.3.2 Freshwater Invertebrates
In the only available study reporting the effects of epichlorohydrin
on an aquatic invertebrate, the 24-hour LC5Q for Daphnia magna was deter-
mined to be 30 mg/1 (Bringmann and Kuhn 1977). In this static test, no
D. magna were killed at 20 mg/1, while all were killed at 44 mg/1. The
9-5
-------
study was conducted in chlorine-free tapwater at 20°C and pH 7.6. Epichloro-
hydrin levels were not actually measured. Three parallel dilution series
were studied using 10 D. magna in each culture vessel. The cessation of
swimming was considered to be equivalent to death.
9.3.3 Saltwater Fish
The only study found in the literature on the effects of epichloro-
hydrin on saltwater biota involved the tidewater silverside, Menidia beryllian,
an estuarine fish. Dawson et al. (1977) reported the 96-hour LC5Q to be
18 mg/1 (Table 9-3). Tidewater silversides were obtained from Horsehoe
Bay near Sandy Hook, New Jersey. They were acclimated for 14 days in
114-liter aquaria at 20°C and fed minced frozen shrimp. Dilution water
was obtained from a well in Passaic, New Jersey, and was the base for a
synthetic sea salt medium ("Instant Ocean"). A specific gravity of 1.018
was maintained. Water for testing was prepared 1 day in advance and placed
in 19-liter test aquaria. Tidewater silversides (40-100 mm in length) were
randomly selected for the assays. Continuous aeration was considered
necessary because of the activity and size of the fish. Only nominal
toxicant levels were reported for this static test. Mortality counts and
LCcA values were determined as for the bluegill (Section 9.3.1). The
bU
death rate of control fish during the experiment was acceptably low at 3.0
i
percent.
9.4 SUMMARY
The limited data found on the effects of epichlorohydrin on micro-
organisms and plants indicate that growth inhibition and toxicity would
occur at greater than 5 mg/1. Theoretical estimates of bioconcentration
suggest that epichlorohydrin would not accumulate substantially in food
chains.
Limited toxicity data for five aquatic animals indicated that exposure
to epichlorohydrin concentrations of more than 10 mg/1 for 1-4 days would
be harmful. In only one of, the tests, however, was the actual epichloro-
hydrin concentration measured. No data were found on the effects of longer
exposures.
9-6
-------
10. REGULATIONS AND STANDARDS
-. Epichlorohydrin is regulated under numerous U.S. and foreign statutes.
These have been grouped according to the type of activity or medium being
controlled.
10.1 OCCUPATIONAL STANDARDS
i
The current OSHA standard for epichlorohydrin levels in the workplace
is 19 mg/m3 (5 ppm) (29 CFR 1910.1000). This threshold limit value, expressed
as an 8-hour time-weighted average (TWA), was based on the known acute
health effects to humans from respiratory tract irritation and systemic
poisoning. After a comprehensive literature review, NIOSH (1976a) concluded
that human exposure risks may include carcinogenesis, mutagenesis, and
sterility. NIOSH (1976b) recommended that worker exposure to epichlorohydrin
be limited to 0.5 ppm (2 mg/m3) for a 40-hour workweek, with a ceiling
value of 15 ppm (15 minutes). At the time of this report, OSHA had not
adopted the lower, NIOSH-recommended TWA. Table 10-1 presents the accepted
occupational standards for epichlorohydrin exposure in seven countries.
10.2 FOOD TOLERANCES
FDA permits an epichlorohydrin-derived resin reacted with ammonia to
be used~as an ion-exchange resin in the treatment of food and potable water
(21 CFR 173.25) and use of molecular sieve resins cross-linked with epi-
chlorohydrin for processing foods and production of whey (21 CFR 173.40).
Industrial starch (21 CFR 178.3520) and food starch may be cross-linked by
treatment with epichlorohydrin (not to exceed 0.3 percent) alone or in
conjunction with propylene oxide, acetic anhydride, or succinic anhydride
(21 CFR 172.892). Traces of free epichlorohydrin have been found in resins
manufactured outside of the United States (NIOSH 1976a).
Various resins of epichlorohydrin can be used in the manufacture of
paper and paperboard that will be in contact with dry, aqueous, and fatty
foods (21 CFR 175.300, 175.390, and 175.320). In particular, 4,4'-
isopropylidene-diphenol-epichlorohydrin resins (minimum molecular weight
10,000) and 4,4'-isopropylidenediphenol-epichlorohydrin thermosetting epoxy
resins may be used as articles or components of articles intended for
food-related uses (21 CFR 177.1440 and 177.2280). Epichlorohydrin is also
regulated by the FDA as a component of adhesives (21 CFR 175.105).
10-1
-------
TABLE 10-1. OCCUPATIONAL STANDARDS FOR EPICHLOROHYDRIN
Standard
Country
Level (ppm)
MACC
MAC
MAC
MAC
MAC
MAC£
TWA€
Netherlands
U.S.S.R.
Czechoslovakia
Federal Republic of Germany
German Democratic Republic
Rumania
U.S.
2.0
0.26C
5.0
Maximum Allowable Concentration
DSource: IRPTC (1979).
Sources: Winell (1975), Sram et al. (1980).
^Source: Wexler (1971).
h"ime-weighted average
Source: 29 CFR 1910.1000
10.3 TRANSPORTATION REGULATIONS
Epichlorohydrin transport on both land and water is regulated. The
Department of Transportation (DOT) has designated epichlorohydrin as a
"hazardous material for the purpose of transportation" (49 CFR 172). This
requires container labeling for class 3 poisons as follows:
EPICHLOROHYDRIN
POISON! FLAMMABLE!
SKIN CONTACT CAUSES DELAYED BURNS
Avoid contact with eyes, skin, and clothing.
Avoid breathing vapor.
Use only with adequate ventilation.
Keep away from heat and open flame.
Keep container closed.
Do not take internally.
First Aid: NIn case of skin contact, immediately remove all
contaminated clothing, including footwear; wash skin with
plenty of water for at least 15 minutes; and call a physician.
In case of eye contact, flush eyes with water for
15 minutes and call a physician.
10-2
-------
The U.S. Coast Guard under 33 CFR 126, 46 CFR 153, and 46 CFR 151 also
has developed safe handling procedures for epichlorohydrin in waterfront
areas, in self-propelled vessels, and in unmanned barges. The required
warning labels are similar to that developed by DOT (above).
10.4 WATER REGULATIONS
Epichlorohydrin is not regulated under the Safe Drinking Water Act.
The Clean Water Act prohibits the discharge of more than 1,000 pounds (454
kg) of epichlorohydrin into navigable waters (40 CFR 116). Discharge at
this level may be harmful and must be reported.
10.5 SOLID WASTE REGULATIONS
Under the Resources Conservation Recovery Act, EPA has designated
epichlorohydrin as a "hazardous waste" (40 CFR 261). If quantities exceed
100.kg/month, disposal must be in a special landfill. Compliance with the
National Pollutant Discharge Elimination System (NPDES) is also required
(40 CFR 122).
10-3
-------
-------
REFERENCES
Abdel-Sayed, H. W.; Yousef S. H.; Tadros, F. E.; Soliman, N. Z. (1974) The
effect of epichlorohydrin to prevent the reaction between DDT lindane oil
emulsion and the inside walls of the containers. Agric. Res. Rev.
52:145-161.
Adamek, P.; Peterka, V. (1971) The determination of epichlorphydrin in aqueous
solutions in the presence of glycerin, monochlorohydrin and glycidol.
Analyst (London) 96:807-809.
Addy, J. K.; Parker, R.E. (1965) The mechanism of epoxide reactions. Part VII.
The^reactions of 1,2-epoxybutane, 3,4-epoxy-l-butene, l,2-epoxy-3-chloro-
propane, and l,2-epoxy-3-methoxypropane with chloride ion in water under
neutral and acidic conditions. J. Chem. Soc. Jan:644-649.
Attman, P. L. (ed); Dittmer, D. S. (ed). (1974) Biology Data Book, v. 3. 2nd
ed. Bethesda, MD: Federation of American Societies for Experimental
Biology.
American Conference of Governmental Industrial Hygienists. (1979) Documenta-
tion of threshold limit values for chemical-substances in workroom air.
4th ed. Cincinnati, OH: American Conference of Governmental Industrial
Hygienists; pp. 16-17, 34-37.
American Public Health Association. (1976) Standard methods for the examina-
tion of water and wastewater. 14th ed. Washington, DC: American Public
Health Association; p. 1076.
Ames, B. N.; McCann, J.; Yamasaki, E. (1975) Methods for detecting carcinogens
and mutagens with the Salmonella/mammalian-microsome mutagenicity tests.
Mutat. Res. 31:347-364.
Andersen, M.; Kiel, P.; Larsen, H.; Maxild, J. (1978) Mutagenic action of
aromatic epoxy resins. Nature (London) 276:391-392.
Anderson, G. E.; Liu, C. S.; Holman, H. Y.; Killus, J. P. (1980) Human exposure
to atmospheric concentrations of selected chemicals. Appendix A-13: A
summary of data on epichlorohydrin. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards.
Andersson, K.; Levin, J. 0.; Lindahl, R.; Nilsson, C. A. (1981) Sampling of
epichlorohydrin and ethylene chlorohydrin in workroom air using Amberlite
XAD-7 resin. Chemosphere 10:143-146.
Anonymous. (1961) Epichlorohydrin (l-chloro-2,3-epoxypropane). Am. Ind. Hyg.
Assoc. J. 22:511-513.
Anonymous. (1978) Derailment spills epi. Chem. Week 122:22.
Bandel, G. 1971. Methods for inducing mutations in Eucalyptus. Cienc. Cult.
(Sao Paulo) 23:241-246.
R-l
-------
Bartsch, H.; Malaveille, C.; Camus, A. M.; Martel-Planche, G.; Brun, G.;
Hautefeuille, A.; Sabadie, N.; Barbin, A.; Kuroki, T.; Drevon, C.;
Piccoli, C.; Montesano, R.; (1980) Validation and comparative studies on
180 chemicals with S. typhimurium strains and V79 Chinese hamster cells
in the presence of vari ous metabolizing systems. Mutat. Res. 76:1-50.
Beckwith, A. L. J. (1972) Phosphorous and hypophosphorous acid derived radicals
and their reactions - an E.P.R. study. Aust. J. Chem. 25:1887-1905.
Blackford, J. L. (1978) Marketing research report on epichlorohydrin. In:
Chemical economics handbook. Menlo Park, CA: SRI International; pp.
642.3021A-642.3022V.
Bridges, B. A. (1978) On the detection of volatile liquid mutagens with bac-
teria: experiments with dichloros and epichlorohydrin. Mutat. Res.
54:367-371.
Bridie, A. L.; Wolff, C. J. M.; Winter, M. (1979a) BOD and COD of some petro-
chemicals. Water Res. 13:627-630.
Bridie, A. L.; Wolff, C. J. M.; Winter, M. (1979b) The acute toxicity of some
petrochemicals to goldfish. Water Res. 13:623-626.
Briggs, G. G. (1973) A simple relationship between adsorption of organic
chemicals and their octane!/water partition coefficients. In: Proceedings
of the 7th British insecticide and fungicide conference; pp. 83-85.
Bringmann, G.; Kuhn, R. (1976) Comparative results of the harmful effects of
water pollutants on bacteria (Pseudomonas putida) and blue-green algae
(Microcystis aeruginosa). GWF Gas Wasserfach: Wasser Abwasser 117:410-413.
Bringmann, G.; Kuhn, R. (1980) Comparison of the toxicity thresholds of water
pollutants to bacteria, algae, and protozoa in the cell multiplication '
inhibition test. Water Res. 14:231-241.
Bringmann, G.; Kuhn, R. (1977) Findings on the damaging action of water pollu-
tants on Daphnia magna. Z. Wasser Abwasser Forsch. 10(5):161-166.
Bronsted, J. N.; Kilpatrick, M.; Kilpatrick, M. (1929) Kinetic studies on
ethylene oxides. J. Am. Chem. Soc. 51:428-461.
Bufalini, M. (1971) Oxidation of sulfur dioxide in polluted atmospheres - a
review. Environ. Sci. Techno!. 5(8):685-700.
Calvert, J. G.; Pitts, J. N., Jr. (1966) Photochemistry. New York, NY: John
Wiley & Sons, Inc.; pp. 522-528.
Carbone, P.; Barbata, G.; Margiptta, G.; Tomasino, A.; Granata, G. (1981) Low
epichlorohydrin concentrations induce sister chromatid exchanges in human
lymphocytes i_n vitro. Caryologia 34:261-266.
Carcinogen Assessment Group, U.S. Environmental Protection Agency. (1980)
Preliminary risk assessment of epichlorohydrin. Type I - air program.
Prepared for the Office of Air Quality Planning and Standards.
R-2
-------
Carpenter, C. P.; Smyth, H. F. , Jr.; Pozzani, U. C. (1949) The assay of acute
vapor toxicity, and the grading and interpretation of results on 96
chemical compounds. J. Ind. Hyg. Toxicol. 31:343-346 resuii;s on yb
E-' W' (968)
. Epoxidation of halo
by a gjavobacterium sp.
A'>; Jackson> H. (1974) Effects of alpha-chlorohydrin
°rgans and fenimy
Re9ression mode1 and 11fe
. Stat. Soc. B
°rUII1P35a57-169779' Dose"resP°nse
in carcinogenesis. Biometrics
Crump KS.; Watson, W. W. (1979) GLOBAL 79. A Fortran program to extra-
polate dichotomous animal carcinogenicity data to low dose National
Instrtute of Environmental Health Science Contract No l-ES-2123
Crump, K S. ; Howe, R. B._(1983) ADOLL1-83 and ADOLL2-83, FORTRAN programs for
ZnhT^hlH9^^11^-5^9!"10^1 Wlth a time-dependent dosePpaSn
Unpublished instructions for use of computer programs.
Crump K. S.; Howe, R B. (1984)The multistage model with a time-dependent
ati°nS t0 Carc1n°9en1c ^ assessment. Ri ^Analysis
' H' A'-; De51j L L (1977> Confidence intervals and test
inf—d *™ »1«1
.; Park, C. N. (1979) Epichloro-
A9ainst
'; Johnst°' R'
dete™1-«°" °f eplchlorohydrin in air.
Dawson, G. W. ; Jennings, A. L. ; Drozdowski, D. ; Rider, E
'
R-3
-------
DePetrocellis, L.; Tortoreto, M.; Paglialunga, S.; Paesam, R.; Airoldi, L.;
Ramos Castaneda, E.; Pantarotto, C. (1982) Determination of epichloro-
dydrin in blood by gas chromatography and selected ion monitoring. J.
Chromatogr. 240:218-23.
Oilling, W. L.
(1977) Interphase transfer processes. II. Evaporation rates of
chloro methanes, ethanes, ethylenes, propanes, and propylene from dilute
aqueous solutions. Comparisons with theoretical predictions. Environ.
Sci. Technol. 11:405-409.
Dilling, W. L.; Bredeweg, C. J.; Tefertiller, N. B. (1976) Organic photo-
chemistry. Simulated atmospheric photodecomposition rates of methylene
chloride, 1,1,1-trichloroethane, trichloroethylene, tetrachloroethylene,
and other compounds. Environ. Sci. Technol. 10:351-356.
Dobbs, A. J.; Gilbert, B. C.; Laue, H. A. H.; Norman, R. 0. C. (1976) Electron
spin resonance studies. Part IL. Radicals derived from oxirans in
aqueous solution. J. Chem. Soc. Perkin Trans. 2:1044-1047.
Dabinson, B.; Hofmann, W.; Stark, B. P. (1969) The determination of epoxide
' groups. Oxford, England: Pergamon Press; pp. 2-61.
Doll, R. (1971) Weibull distribution of cancer. Implications for models of
carcinogenesis. J. R. Stat. Soc. A 13:133-166.
Dow Chemical Company. (1980) Epichlorohydrin — properties, uses, storage,
handling. Midland, MI: Dow Chemical Company.
Dripps, R. D.; Eckenhoff, J. E.; Vandam, L. D. (1977) Introduction to anesthe-
sia, the principles of safe practice. 5th ed. Philadelphia, PA: W. B.
Saunders Company, pp. 121-123.
Eder, E.; Neudecker, T.; Lutz, D.; Henschler, D. (1980) Mutagenic potential of
ally! and allylic compounds: structure-activity relationship as determined
by alkylating and direct in vitro mutagenic properties. Biochem. Pharmacol
29:993-998.
Elmore, J. D.; Wong, J. L; Laumbach, A. D.; Streips, U. N. (1976) Vinyl
chloride mutagenicity via the metabolites chlorooxirane and chloroacetal-
dehyde monomer hydrate. Biochim. Biophys. Acta 442:405-419.
Enterline, P. E. (1978) Updated mortality in workers exposed to epichloro-
hydrin. Unpublished reports to Shell Oil Co. (Aug. 31, 1977- July 31,
1978) and letter to R.E. Joyner, Shell Oil Co. (July 31, 1978). Letter
from R.E. Joyner, corporate Medical Director, Shell Oil Co., to J.M.
Lane, Acting Director, NIOSH (Aug. 7, 1978).
Enterline, P. E. (1980) Cancer in men manufacturing isopropyl alcohol. Report
attached to a letter from M.J. Sloan (Nov. 11, 1980), Shell Oil Co., to
U.S. Environmental Protection Agency. EPA document control No. OTS-1180-
0069.
R-4
-------
Appl. Pharmacol. 23:288-325. .muus>e. IOXTCOI.
AtT.Y>hRaYm(19S7J?
and toxiclty in
Fjellstedt, T A.; Allen, R. H. ; Duncan, B. K. ; Jakoby, W. B (1973) Enzymatic
conjugation of epoxides with glutathione. J. Biol. Chen.. 248:3702-3707
Ga9e'l6:'llC-i4(1959)
toxi"city of epichlorohydrin vapour. Br. J. Ind. Med.
WVnm (19w7) DPha^acolo9i'cal agents and male fertility. In: Johnson A
d^TpreWss;%j,.Va6no^28k' "' L ' "*' ^^ ^ ^ New Yor^TY: ''
NeW
R- Weigmann, H. J
amb1ent
of ^ '- Pollution.
MAC value for
n Vn- V°pr°sy fl1^6^ atmosfernogo vozdukha i
naselennykh mest [Determination of latent time of reflex
reactions of laboratory animals acted upon by toxic agents in atmospheric
avr Uch. Zap. Mosk. Nauchno Issled. Inst. Gig. im F. F. Erismana:
R-5
-------
Hahn, J. D. (1970) Posttesticular antifertility effects of epichlorohydrin and
2,3-epoxypropanol. Nature (London) 226:87.
Hansch, C.; Leo, A. (1979) Substituent constants for correlation analysis in
chemistry and biology. New York, NY: John Wiley & Sons; p. 321.
Hawley, G. G. (ed). (1977) The condensed chemical dictionary. 9th ed. New
York, NY: Van Nostrand Reinhold Co.; p. 343.
Hayakawa, T.; Udenfriend, S.; Yagi, H.; Jerina, D. M. (1975) Substrates and
inhibitors of hepatic glutathione-s-epoxide transferase. Arch. Biochem.
Biophys. 170:438-451.
Heslot, H. (1962) Etude quantitative de reversions biochemiques chez la lerure
Schizosaccharomyces pombe par des substances radiomimetiques. Abh.
Dtsch. Akad. Wiss. BerTTn, K. L. Med. 1:193-228.
Mine, C.; Rowe, V. K.; White, E. R.; Darmer, K. I., Jr.; Youngblood, G. T.
(1981) Miscellaneous epoxy compounds. In: Clayton, G. D.; Clayton, F. E.,
eds. Patty's industrial hygiene and toxicology, 3rd ed., Vol. II-A. New
York, NY: John Wiley & Sons; pp. 2148-2154, 2242-2247.
Hushon, J.; Clerman, R.; Small, R.; Sood, S.; Taylor, A.; Thoman, D% (1980) An
assessment of potentially carcinogenic, energy-related contaminants in
water. McLean, VA: Mitre Corporation; p. 117 MTR-79W171.
International Commission on Radiological Protection (ICRP). (1977) Recom-
mendation of the International Commission on Radiological Protection,
adopted Jan. 17, 1977. Oxford, England: Pergamon Press. Publ. no. 26.
Ippen H.; Mathies, V. (1970) Protracted chemical burns (with special consider-
ation of dermatological damage due to epoxides and propansultone). Berufs
Dermatosen 18:144-165.
Jaraczewska, W.; Kaszper, W. (1967) Determination of epichlorohydrin in air.
Med. Pr. 18:168-176.
Jirasek, L.; Kalensky J. (1960) Epoxy resins. Prac. Lek. 12:67-74.
John JA; Quast JF; Murray FJ; Henck JW; Calhoun LG; Murray JS; Pilny MK Deacon
MS; Crawford AA; and Schwetz BA. 1979. The effects of inhaled epichlo-
rohydrin on the semen of rabbits and on the fertility of male and female
rats. Midland, MI: Dow Chemical USA, Toxicology Research Laboratory,
Health and Environmental Sciences; unpublished report.
Jones, A. R.; O'Brien, R. W. (1980) Metabolism of three active analogues of
the male antifertility agent alpha-chlorohydrin in the rat. Xenobiotica
10:356-370.
Jones, A. R.; Davies, P.; Edwards, K.; Jackson, H. (1969) Antifertility effects
and metabolism of alpha- and epichlorohydrins in the rat. Nature (London)
224(5214):83.
R-6
-------
Juhnke, I.; Ludemann, D. (1978) Results of the investigation of 200 chemical
compounds for acute fish toxicity with the Golden Orfe test. Z Wasser
Abwasser Forsch. 11:161-164.
Kalla, N. R. ; Bansal , M. P. (1977) In vivo & in vitro alkylation of testicular
cysteine by alpha-chlorohydrin. Ind. J. ExpTTJTol. 15:232-233.
Kastenbaum, M. A. ; Bowman, K. 0. (1970) Tables for determining the statistical
significance of mutation frequencies. Mutat. Res. 9:527-549.
Kawabata, A. (1981) Studies on the carcinogenic activity of epichlorohydrin by
oral administration in male Wistar rats. Nara Igaku Zasshi 32:270-280.
Kenaga, E. E. (1980) Predicted bioconcentration factors and soil sorption
coefficients of pesticides and other chemicals. Ecotoxi col. Environ. Saf.
* • &.Q OO*
Kenaga, ^E. E. ; Goring, C. A. I. (1980) Relationship between water solubility
soil sorption, octanol-water partitioning, and concentration of chemical's
in biota In: Eaton, J G. ; Parrish, P. R. ; Hendricks, A. C. , eds.
Aquatic toxicology. Philadelphia, PA: American Society for Testing and
Materials; pp. 78-115; ASTM STP 707.
?i'oM;.u?n.' T; G'; Conner> T- H-5 Le9ator, M. S. ; Edwards, H. N.
(1978) Mutagenicity of epichlorohydrin in the bacterial assay system-
Evaluation by direct in vitro activity and in vivo activity of urine
from exposed humans and "mice. Mutat. Res. 53:727 --
Kirkhart, B. (1981) Micronucleus test on 21 compounds. In: deSerres FJ Ashby
M'-UU-; Evaluation of short-term tests for carcinogens. Amsterdam, The
Netherlands: Elsevier/North Holland; pp. 698-704.
Knaap, A. G. A. C; Voogd, C. E. ; Kramers, P. G. N. (1982) Comparison of the
mutagemc potency of 2-chloroethanol , 2-bromoethanol , 1,2-epoxybutane
epichlorohydrih and glycidaldehyde in Klebsiella pneumpniae, Drosophila
melanogaster and L5178Y mouse lymphoma cells. Mutat. Res. 101:199-^08.
Kolmark, G. ; Giles, N. H. (1955) Comparative studies of monoepoxides as inducers
of reverse mutations in Neurospora. Genetics 40:890-902.
Konishi, T ; Kawabata A. ; Denda, A.; Ikeda, T. ; Katada, H. ; Maruyama, H. ;
mgasnigucni, R. (1980) Forestomach tumors induced by orally administered
epichlorohydrin in male Wistar rats. Gann 71:922-923.
Kotin, P.; Falk^H. L. (1963) Organic peroxides, hydrogen peroxide, epoxides,
and neoplasia. Radiat. Res. Supp. 3:193-211.
\
Kremneva, S. N. (1960) Investigation of toxicity of epichlorohydrin and dichlo-
rohydnn. Pr. Toksikol. Moscow Sb. 123-131.
Kremneva, S. N. ; Tolgskaya, M. S. (1961) Toxicology of epichlorohydrin. Toksikol
Nov. Prom. Khim. Veshchestv. 2:28-41.
R-7
-------
Kucerova, M.; Polivkova, Z. (1976) Banding technique used for the detection of
chromosomal aberrations induced by radiation and alkylating agents TEPA
and epichlorohydrin. Mutat. Res. 34:279-290.
Kucerova, M.; Polivkova, Z.; Sram, R.; Matousek, V. (1976) Mutagenic effect of
epichlorohydrin. I. Testing of human lymphocytes J_n vitro in comparison
with TEPA. Mutat. Res. 34:271-278.
Kucerova, M.; Zhurkov, V. S.; Polivkova, Z.; Ivanova, J. E. (1977) Mutagenic
effect of epichlorohydrin. II. Analysis of chromosomal aberrations in
lymphocytes of persons occupationally exposed to epichlorohydrin. Mutat.
Res. 48:355-360.
Kwart, H.; Goodman, A. L. (1960) The acid-catalyzed ring opening of epichloro-
hydrin. J. Am. Chem. Soc. 82:1947-1949.
Lambert, D.; Lacroix, M.; Ducombs, G.; Journet, F.; Chapuis, J. L. (1978)
Cutaneous allergy to epichlorohydrin. Ann. Dermatol. Venereol.
105:521-525.
Laskin, S.; Sellakumar, A. R.; Kuschner, M.; Nelson, N.; La Mendola, S.;
Rusch, G. M.; Katz, G. V.; Dulak, N. C.; Albert, R. E. (1980) Inhalation
carcinogenicity of epichlorohydrin in noninbred Sprague-Dawley rats. JNCI
J. Natl. Cancer Inst. 65:751-755.
Laumbach, A. D.; Lee, S.; Wong, J.; Streips, U. N. (1977) Studies on the
mutagenicity of vinyl chloride metabolites and related chemicals. In:
Proceedings of the 3rd international symposium on the prevention and
detection of cancer 1:155-170.
Lawrence, W. H.; Autian, J. (1972) Possible toxic effects from inhalation of
dental ingredients by alteration of drug biologic half-life. J. Dent.
Res. 51:878.
Lawrence WH; Malik M; Turner JE; Autian J. (1972) Toxicity profile of epichloro-
hydrin. J. Pharm. Sci. 61:1712-1717.
Lawrence, W. H.; Malik, M.; Autian, J. (1974) Development of a toxicity evalua-
tion program for dental materials and products. II. Screening for systemic
toxicity. J. Biomed. Mater. Res. 8:11-34.
le Noble, W. J.; Duffy, M. (1964) The effect of pressure on the rates of
acid-catalyzed hydrolysis of trimethylene oxide and of epichlorohydrin.
J. Phys. Chem. 68:619-622.
Long, F. A.; Pritchard, J. G. (1956) Hydrolysis of substituted ethylene oxides
in H^O solutions. J. Am. Chem. Soc. 78:2663-2667.
Lowenheim, F. A.; Moran, M. K. (1975) Faith, Keyes, and Clark's industrial
chemicals, 4th ed. New York, NY: John Wiley & Sons; pp. 335-338.
Lu, P-Y.; Metcalf, R. L. (1975) Environmental fate and biodegradability of
benzene derivatives as studied in a model aquatic ecosystem. EHP Environ.
Health Perspect. 10:269-284.
R-8
-------
Lukaneva, A.M.; Rodionov, G. A. (1978) Study of the influence of polychloro-
camphene and epichlorohydrin on the development of experimental athero-
sclerosis. Fiziol. (Kiev) 24:371-377.
Mackey, B. E.; MacGregor, J. T. (1979) The micronucleus test: statistical
design and analysis. Mutat. Res. 64:195-204.
Mann, H. (1976) Fischtest mit Goldorfen zur vergleichenden Prufung der akuten
Toxizitat yon Wasserinhaltsstoffen und Abwassern-Praktische Erfahrungen
aus drei Ringtesten. Z. Wasser Abwasser Forsch. 9:103-109.
Mantel, N.; Schneiderman, M. A. (1975) Estimating "safe" levels, a hazardous
undertaking. Cancer Res. 35:1379-1386.
Marks, T. A.; Gerling, F. S.; Staples, R. E. (1982) Teratogenic evaluation of
epichlorohydrin in the mouse and rat and glycidol in the mouse. J. Toxicol.
Environ. Health 9:87-96.
Mashford, M.; Jones, A. R. (1978) The antifertility action of alpha-chlorohydrin:
, Enzyme inhibition by alpha-chlorohydrin phosphate. Experientia 34:1267-1268.
Massiot, G.; Levy, J. (1981) Une reaction in attendue de 1'epichlorohydrine
avec les amines. Bull. Soc. Chim. France (11-12):11-463 - 11-467.
Milby, T. H.; Whorton, D. (1980) Epidemiological assessment of occupationally
related chemically induced sperm count suppression. JOM J. Occup. Med.
22:77-82.
Milby, T. H.; Whorton, M. D.; Stubbs, H. A.; Ross, C. E.; Joyner, R. E.;
Lipshultz, L. I. 1981. Testicular function among epichlorohydrin workers.
Br. J. Ind. Med. 38:372-377.
Moore-Brown, M. M.; Clive, D. (1979) The L5178Y/TK
situ results. Banbury Rep. 2:71-88.
mutagen assay system: In
National Institute for Occupational Safety and Health. (1976a) Criteria for a
recommended standard Occupational exposure to epichlorohydrin.
Washington, DC: Department of Health, Education, and Welfare. DHEW Publ.
NIOSH 76-206.
National Institute for Occupational Safety and Health. (1976b) A recommended
standard for occupational exposure to epichlorohydrin. Washington, DC:
Department of Health, Education, and Welfare.
National Institute for Occupational Safety and Health. (1978) Current Intelli-
gence Bulletin 30: Epichlorohydrin. Washington, DC: Department of Health,
Education and Welfare DHEW Publication No. 79-105.
Neeley, W. B.; Branson, D. R.; Blau, G. E. (1974) Partition coefficient to
measure bioconcentration potential of organic chemicals in fish. Environ.
Sci. Technol. 8:1113-1115.
R-9
-------
Norppa, H.; Hemminki, K.; Sorsa, M.; Vainio, H. (1981) Effect of monosubsti-
tuted epoxides on chromosome aberrations and SCE in cultured human
lymphocytes. Mutat. Res. 91:243-250.
Pallade, S.; Dorobantu, M.; Rotaru, G.; Gabrielescu, E. (1967) Experimental
studies of intoxication by epichlorohydrin. Arch. Mai. Prof. Med. Trav.
Secur. Soc. 28:505-516.
Perry, P.; Wolff, S. (1974) New Giemsa method for the differential staining of
sister chromatids. Nature (London) 251:156-158.
Petty, W. L.; Nichols, P. L., Jr. (1954) Rate of addition of the nitrate ion
to the ethylene oxide ring. J. Am. Chem. Soc. 76:4385-4389.
Picciano, D. (1979) Cytogenetic investigation of occupational exposure to
epichlorohydrin. Mutat. Res. 66:169-173.
Pilny, M. K.; Lederer, T. S.; Murray, J. S.; Deacon, M. S.; Hanley, T. R.;
Quast, J. F.; and John, J. A. (1979) Epichlorohydrin—subchronic studies.
IV. The effects of maternally inhaled epichlorohydrin on rat and rabbit
1 embryonal and fetal development. Midland, MI: Dow Chemical USA, Toxicology
Research Laboratory, Health and Environmental Sciencies; unpunlished
report.
Planche, G.; Croisy, A.; Malaveille, C.; Tomatis, L.; Bartsch, H. (1979)
Metabolic and mutagenicity studies on DDT and 15 derivatives. Detection
of l,l-bis(p-chlorophenyl)-2,2-dichloroethane and l,l-bis(p-chlorophenyl)-
2,2,2-trichloroethyl acetate (kelthane acetate) as mutagens in Salmonella
typhimurium and of l,l-bis(p-chlorophenyl) ethylene oxide, a likely
metabolite, as an alkylating agent. Chem. Biol. Interact. 25:157-175.
Pritchard, J. G.; Long, F. A. (1956) Kinetics and mechanisms of the acid-
catalyzed hydrolysis of substituted ethylene oxides. J. Am. Chem. Soc.
78:2667-2670.
Pritchard, J. G.; Siddiqui, I. A. (1973) Activation parameters and mechanism
of the acid-catalyzed hydrolysis of epoxides. J. Chem. Soc. Perkin
Trans. 2 4:452-457.
Quast, J. F.; Henck, J. W.; Postma, B. J.; Schuetz, D. J.; McKenna, M. J.
(1979a) Epichlorohydrin-- Subchronic studies. I. A 90-day inhalation
study in laboratory rodents (Fischer 344 rats, Sprague-Dawley rats and
B6C3F1 mice). Midland, MI: Dow Chemical U.S.A.
Quast, J. F; Lederer, T. S.; Postma, B. J.; Scheutz, D. J.; John, J. A.;
McKenna, M. J. (1979b) Epichlorohydrin—Subchronic studies. II. A 12-day
inhalation study in laboratory rodents (Fischer 344 rats, Sprague-Dawley
rats and B6C3F1 mice). Midland, MI: Dow Chemical U.S.A.
Rapoport, I. A. (1948) The effect of ethylene oxide, glycide and glycol on
genetic mutations. Dokl. Akad. Nauk SSR 60:469-472.
R-10
-------
Riesser G H. (1978) Epichlorohydrin. In: Kirk-Othmer encyclopedia of chemical
technology. V. 5. New York, NY: John Wiley & Sons; pp/ 858-864.
Of C6rta1n ep°Xldes in ac>ueous solutions.
Rotaru, G. ; . Pal lade, S. (1966) Experimental studies of histopathological
features in acute epichlorohydrin toxicity. Morfol. Norm. Patol. 11:155-
Rutledge, J. E ; Islam, M. N. (1973) Canning and pH stability of epichloro-
hydrin- treated parboiled rice. J. Agric. Food Chem. 21:458-460.
iJLV i,6^1?/5^09"96™115 ProPerties of industrial materials. 4th ed
New York, NY: Van Nostrand Reinhold Co.; p. 717.
Schultz, V. ^C. (1964) Fettleber und chronisch-asthmoide Bronchitis nach Inhala-
89°1342-1344a §Smitte1S (EP1chlo™hydrin). Dtsch. Ned. Wochenschr.
SheTl Chemical Company. (1969) Epichlorohydrin. Shell Oil Co., Industrial
Chemicals Division; technical bulletin no. IC-69-16.
Shellenberger R. J. ; McClimans C D. ; Ott, M. G. ; Flake, R. E. ; Daniel, R.
evaluatlon of the mortality experience of employees with
to ul
.GV H.owell> W" R'
technology-
Epoxy resins. In:
Shumskaya N I ; Karamzina, _N. M.^1966) An evaluation of the functional
NoT ?rUa\hkne"P0-9by 1ndUStrial SUbStanCeS' T°kSlko1
Shumskaya N I.; Karamzina, N. M. ; _Savina, M. Y. (1971) Evaluating the sensi-
intnvyr-a/- Inte9rf.fnd specific parameters in acute epichlorohydrin
intoxication. Toksikol. Nov. Prom. Khim. Veshchestv. 12:33-44.
A1eksanyaP' L v- (^3) Kinetics of the hydrolysis of 1-chloro
solutl°ns- Zh-
Khim
\
^?frH^ P-W.; Young, J. D. (1979) Pharmacokinetics of epichlo
nnwhv^ D administered to rats by gavage or inhalation. Midland, MI:
Dow Chemical USA, Toxicology Research Laboratory.
nF;V JV ;- CaTPer^r> c: p- (1948) Further experience with the range
30 63-68 ^ Industna1 toxicology laboratory. J. Ind. Hyg. Toxicol.
R-ll
-------
Smyth, H. F., Jr.; Carpenter, C. P.; Weil, C. S.; Pozzani, U. C.; Streigel, J.
A. (1962) Range-finding toxicity data, List VI. Ind. Hyg. J. 23:95-96.
Sora, S.; Zelaschi, D.; Ronzio, E. (1979') Activity of epichlorohydrin on
different genetic systems of Saccharomyces cerevisiae. Atti Assoc. Gen?t.
Ital. 24:274-276.
Sram, R. J.; Cerna, M.; Kucerova, M. (1976) The genetic risk of epichlorohy-
drin as related to the occupational exposure. Biol. Zentralbl. 95:451-462.
Sram, R. J.; Tomatis, L.; Clemmesen, J.; Bridges, B. A. (1981) An evaluation
of the genetic toxicity of epichlorohydrin. Mutat. Res. 87:299-319.
Sram, R. J.; Zudova, Z.; Kuleshov, N. P. (1980) Cytogenetic analysis of peri-
pheral lymphocytes in workers occupationally exposed to epichlorohydrin.
Mutat. Res. 70:115-120.
SRI International. (1981) Directory of Chemical Producers. Menlo Park, CA: SRI
International.
Sto'lzenberg, S. J.; Hine, C. H. (1979) Mutagenicity of halogenated and oxygenated
three-carbon compounds. J. Toxicol. Environ. Health 5:1149-1158.
Suskov, I. I.; Sazonova, L. A. (1982) Cytogenetic effects of epoxy, phenol-
formaldehyde and polyvinylchloride resins in man. Mutat. Res. 104:137-140.
Swan, J. D. (1954) Determination of epoxides with sodium sulfite. Anal. Chem.
26:878-880.
Tassignon. J. P.; Bos, G. D.; Craigen, A. A.; Jacquet, B.; Kueng, H. L.;
Lanouziere-Simon, C.; Pierre, C. (1983) Mortality in a European cohort
occupationally exposed to epichlorohydrin. Unpublished.
Thorgeirsson, A.; Fregert, S. (1977) Allergenicity of epoxy resins in the
guinea pig. Acta Derm. Venereol. 57:253-256.
Thurman, G. B.; Simms, B. G.; Goldstein, A. L.; Kilian, D. J. (1978) The
effects of organic compounds used in the manufacture of plastics on the
responsivity of murine and human lymphocytes. Toxicol. Appl. Pharmacol.
44:617-641.
Tsuchimoto, T.; Matter, B. E. (1981) Activity of coded compounds in the micro-
nucleus test. In: de Serres, F. J.; Ashby, J. eds. Evaluation of short-
term tests for carcinogens. Amsterdam, The Netherlands: Elsevier/ North
Holland, pp. 705-711.
U.S. Environmental Protection Agency. (1974) Addendum to development document
for proposed effluent limitations guidelines and new source performance
standards for the synthetic resins. Washington, DC: U.S. Environmental
Protection Agency, Office of Water and Hazardous Materials; pp. 17-24.
EPA report no. EPA-440/l-74/036-a.
R-12
-------
.S. Environmental Protection Agency. (1975) Methods for acute toxicity tests
with fish, macro-invertebrates, and amphibians. Corvallis, OR: U.S.
environmental Protection Agency, Office of Research.and Development; EPA
U.S. Environmental Protection Agency. (1978) Region III. Field investigation
emergency response action, and hydrogeological assistance at Chessie
P°1nt ""sant, WV,
U.S. Environmental Protection Agency. (1980) Fate of toxic and hazardous
materials in the air environment. Research Triangle Park: U.S. Environ-
no" EPA-eOO/SS-S-OM6"^' °ffiCS °f ReSearch and Devel°Pment; EPA report
U.S. Environmental Protection Agency. (1983) TSCA-plant and production (TSCAPP).
CICIS - Chemical Information Tape. EPA/DF-80/005.
Van Duuren, B. L. ; Katz, C. ; Goldschmidt, B. M. (1972a) Direct-acting alkylating
"C 6therS and re1ated con«Pounds. Toxicol. Appl. Pharmacol
Van Duuren, B. L. ; Katz, C. ; Goldschmidt, B. M. ; Frenkel, K. ; Livak, A. (1972b)
Carcinogen! city of halo-ethers. II. Structure-activity relationships of
a2ad°9S ° bls(cnlorometnyl)ether. JNCI J. Natl. Cancer Inst. 48:1431-
Van
DnUo!ft ?' L-; Go!dschmidt> B- M.-; Katz, C.; Seidman, I.; Paul, J. S.
Inst 53-a695-700niC act1v1ty °f alBating agents. JNCI J. Natl. Cancer
van Esch, G. J. (1982) Induction of preneoplastic lesions in the forestomach
of rats after oral administration of l-chloro-2,3-epoxpropane Prelimi-
nary report no. 627805-005. F rieniiu
van Lierop, J. B. H. (1978) Simple and rapid determination of epichlorohydrin
at the lower parts per billion level by gas chromatography-mass fragmen-
tography. J. Chromatogr. 166:609-610.
Vashishat^R. K.; Vasudeva, M.; Kakar, S. N. (1980) Induction of mitotic
crossing over, mitotic gene conversion and reverse mutation by epichlo-
rohydrin in Saccharomyces cerevisiae. Indian J. Exp. Biol. 18:1337-U38.
Veith G. D ; Macek, K. J.; Petrocelli, S. R.; Carroll, J. (1980) An evalua-
tion or using partition coefficients and water solubility to estimate
bioconcentration factors for organic chemicals in fish. In: Eaton J 'G •
Parnsh^P. R ; Hendricks, A. C., eds. Aquatic toxicology. Philadelphia '
PA: American Society for Testing and Materials; pp. 116-129 ASTM STP 707
Venable, J. R.; McClimans, C.
study of male employees
Occup. Med. 22:87-91.
D.; Flake, R. E.; Dimick, D. B. (1980) A fertility
engaged in the manufacture of glycerine. JOM J.
R-13
-------
Verschueren, K. (1977) Handbook of environmental data on organic chemicals.
New York, NY: Van Nostrand Reinhold Co.; pp. 293-294.
Vogel, E.; Schalet, A.; Lee, W. R.; Wurgler, F. (1981) Mutagenicity of selected
chemicals in Drosophila. In: de Serres, F. J., ed. Comparative Muta-
genesis.
Voogd, C. E. (1973) Mutagenic action of epoxy compounds and several alcohols.
Mutat. Res. 21:52-53.
Voogd, C. E.; van der Stet, J. J.; Jacobs, J. J. J. A. A. (1981) The mutagenic
action of aliphatic epoxides. Mutat. Res. 89:269-282.
Weast, R. C., ed. (1978) CRC handbook of chemistry and physics, 59th ed. West
Palm Beach, FL: CRC Press Inc.; p. C-447.
Weigle, W. W.; Plotnick, H. B.; Conner, W. L. (1978) Tissue distribution and
excretion Qf i4c-epichlorohydrin in male and female rats. Res. Cbmmun.
Chem. Pathol. Pharmacol. 20:275r288.
Weil, C. S.; Condra, N.; Hann, c!; Streigel, J. A. (1963) Experimental carcino-
genicity and acute toxicity of respresentative epoxides. Am. Ind. Hyg.
Assoc. J. 24:305-325.
Westergaard, M. (1957) Chemical mutagenesis in relation to the concept of
the gene. Experientia 13:224-234.
Wexler. B. (1971) Determination of epichlorohydrin contamination in an indus-
trial facility for the manufacturing of epoxy resins. Mater. Plast.
(Bucharest) 8:322-323.
White, A. D. (1980) In vitro induction of sister chromatid exchange in human
lymphocytes by epichlorohydrin with and without metabolic activation.
Mutat. Res. 78:171-176.
Winell, M. (1975) An international comparison of hygienic standards for chemi-
cals in the work environment. Ambio 4:34-36.
Wurgler, F. E.; Graf, U. (1981) Mutagenic activity of ten coded compounds in
the Drosophila sex-linked recessive lethal assay. In: de Serres, F. J. ,
Ashby, J., eds. Evaluation of short-term tests for carcinogens. Amsterdam,
The Netherlands Elsevier/ North Holland; pp. 666-672.
R-14
-------
Appendix A. Evaporation Rate of Epichlorohydrin Calculated According
to the Method of Dill ing (1977)
Equations:
1. H = (16.04)(P)(M)/(T)(S)
2. K-L = 221.1/[(1.042/HJ + 100](M)0-5
3. Half-life (days) = (0.6931/K-L)(d/l,440)
where:
H
P
M
T
S
K,
= Henry's law constant
= vapor pressure in mmHg at 20° C
= gram molecular weight of the solute
= temperature in °K (20° C = 293° K)
= solubility of the solute in water in mg/1 (ppm) at 20° C
= overall liquid exchange constant in cm/min
= solution depth in cm
Calculations:
H = (16.04)(12)(92.53)/(293)(60,000) = 1.01 x 10"3
K-L = 221.!/[(!.042/0.00101) + 100](92.53)°'5 = 2.03 x 10~2
Half-life (solution depth of 6.5 cm) = (0.6931/0.0203)(6.5/1,440) = 0.15 days
Half-life (solution depth of 100 cm) = (0.6931/0.0203)(100/1,440) = 2.37 days
A-l
-------
-------
Appendix B. Soil Adsorption Coefficient (KQC) and Soil Organic
Matter/Water Partition Coefficient (Q)
Equations
^ Io9 Koc = 3.64 - 0.55 (log WS) ± 1.23 orders of magnitude
(Kenaga and Goring 1980)
2. log Q = 0.618 - 0.524 (log P) (Briggs 1973)
where:
WS = water solubility
P = octanol/water partition coefficient
log WS
log P
oc
4.78
4.82
0.26
0.26
10.28 x 10±1>23
9,76 x 10±3"23
5.68
5.68
B-l
-------
-------
Appendix C. Calculation of the Log Octanol/Water Partition Coefficient
(log P) by the Method of Hansch and Leo (1979)
The 7i-constant system was used for calculating log octanol/water
partition coefficients of epichlorohydrin. Propylene oxide was used as a
parent molecule and properly restructured. The appropriate n-constaht
value was used together with its "uncertainty units." The 7i-constant is
an indication of hydrophobicity and is additive. Relative to hydrogen, a
positive value indicates that the substituent favors the octanol phase,
whereas a negative value indicates the water phase is favored.
log P = log P CH3-CH-CH2 + Cl
= -0.13 + 0.39 [± 0.04]*
= 0.26 [± 0.04]*
Uncertainty units.
C-l
-------
-------
Appendix D. Bioconcentration Factors Calculated for Epichlorohydrin
by Four Methods
1. log BCF = 0.124 + 0.542 (log P) (Neely et al. 1974) = 0.265
2. log BCF = 0.23 + 0.76 (log) (Veith et al. 1980) = -0.032
3. log BCF = 3.995 - 0.3891 (log WS) (Lu and Metcalf 1975) = 0.968
(WS = 6.0 x 107 ppb)
4. log BCF = 3.41 - 0.508 (log WS) (Chiou et al. 1977) = 0.458
(WS = 6.48 x 105 umole/1)
where:
BCF = bioconcentration factor
P = octanol/water partition coefficient
WS = water solubility
D-l
-------
-------
APPENDIX E. COMPARISON OF RESULTS BY VARIOUS EXTRAPOLATION MODELS
The estimates of unit risk from animals presented in the body of this
document are all calculated by the use of the linearized multistage model. The
reasons for its use have been detailed therein. Essentially, it is part of a
methodology that estimates a conservative linear slope at low extrapolations
doses and is consistent with the data at all dose levels of the experiment. It
is a nonthreshold model holding that the upper limit of risk predicted by a
linear extrapolation to low levels of the dose-response relationshop is the most
plausible upper limit for the risk.
Other models have also been used for risk extrapolation. Three nonthres-
hold models are presented here: the one-hit, the log-probit, and the Weibull.
/•
The one-hit model is characterized by a continuous downward curvature but is
linear at low doses. It can be considered the linear form or first stage of
the multistage model because of its functional form. Because of this and its
downward curvature, it will always yield estimates of low-level risk that are
at least as large as those of the multistage model. Further, whenever the
data can be fitted adequately by the one-hit model, estimates from the two
procedures will be comparable.
The other two models, the log-probit and the Weibull, are often used to
fit toxicological data in the observable range, because of their general "S"
curvature. The Tow-dose upward curvatures of these two models usually yield
lower low-dose risk estimates than those of the one-hit or multistage models.
The log-probit model was originally proposed for use in problems of
biological assay, such as the assessment of potency of toxicants and drugs, and
is generally used to estimate such values as percentile lethal dose or percen-
tile effective dose. Its development was strictly empirical in that it was
E-l
-------
observed that several log dose-response relationships followed the cumula-
tive normal probability distribution function, $. In fitting the cancer
bioassay data, assuming an independent background, this becomes:
P(D;a,b,xc) = c + (1-c)$ (a+blog10 D) a,b > 0 <_ c < 1
where P is the proportion responding at dose D, c is an estimate of the back-
round rate, a is an estimate of the standardized mean of individuals tolerances,
and b is an estimate of the log dose-probit response slope.
The one-hit model arises from the theory that a single molecule of a
carcinogen has a probability of transforming a single noncarcinogenic cell into
a carcinogenic one. It has the probability distribution function:
P(D;a,b) = l-exp-(a+bd) a,b > 0
where a and b are the parameter estimates. The estimate a represents the
background or zero dose rate, and the parameter estimated by b represents the
linear component or slope of the dose-response model. In discussing the added
risk over background, incorporation of Abbott's correction leads to
P(D;b) = l-exp-(bd) b > 0
Finally, a model from the theory of carcinogenesis arises from the multihit
model applied to multiple target cells. This model has been termed here the
Wei bull model. It is of the form
P(D;b,k) = l-exp-(bdk) b,k > 0
E-2
-------
For the power of dose only, the restriction k > 0 has been placed on this
model. When k > 1, this model yields low-dose estimates of risks usually
significantly lower than either the linear multistage or one-hit models, which
are linear at low doses. All three of these models usually project risk
estimates significantly higher at the low exposure levels than those obtained
with the log-probit model.
The estimates of added risk for low doses for the above models are given
in Table E-l for the epichlorohydrin drinking water study (Konishi et al.,
1980). Both maximum likelihood estimates and 95 percent upper confidence limits
are presented. Since all models estimate the background rate as 0, there is no
need to incorporate Abbott's correction for independent background rate.
The results (Table E-l) show, in order of descending risk, the one-hit,
multistage, Weibull, and log-probit models. The best fit of the data with the
multistage model is a cubic with zero linear component, which accounts for its
nonlinear behavior at low doses.
E-3
-------
CO
cc
OL
^
CO
3
UJ
_]
z
1— <
^r* CO
CO— 1
3 D
UJS
CO
Q 5
1 UJ
OLD
_JU-
U_
Zt-i
ZC O
ou-
§§
wJ Vo'
0:0:
ou.
Eo
LU UJ
^
feS
UJ
co n
UJ
I—
1—4
fc
UJ
•
r-4
t
UJ
UJ
_J
CO
H-
o
_p
•r-
E
•r—
0
0
c in
CO ^
T3 tn
<£ '£
c
O r-
o
•!•"• ^^
I ^ (/)
If) •!-
CO C-
-a r—
O (O
o c=
-C 0
•r- -r—
r— 4-3
CO -r-
j^ -a
•r- -CJ
i— (O
£
'x
(O
yr
•I-5 T
1 •!- 0)
^JJ JQ L^j
O 0 0
_I ^ E
a.
[H ,
3 r—
U) CO
•i— T3
, — g
S
1 -I— "CO
en JD ~o
o o o
_J (- E
a.
,— ,_
3 CO
'S 1
3:
•r— r—
i -a
CO O
o
CO
en
+-> i—
v> co
•r- T3
i — E
S
CO
t/)
o
o
o
o
CO
0
__l
X
co
CO
1
o
CO
CM
0
o
CO
1
0
I— 1
X
T-H
CM
c>
_J
en
3
1 1
o
o
o
1
o
f.^
X
CO
r-.
i
o
X
CO
CM
0
O
r—
i
o
t-H
X
1— 1
CM
O
en
3
T— 1
O
O
<£>
O
, — i
X
CO
<£>
1
O
CO
CM
O
O
<£>
1
O
t— <
X
1 — 1
CM
r-
i
o
i-H
X
t-H
_J
en
3
O
t— 1
o
CO
«— i
i
o
1 — 1
X
r-t
•
CM
LO
1
0
X
CO
LO
1
0
X
CO
CM
0
•*
1" "1
o
1 — 1
X
r-H
t— i
LO
1
o
1 — 1
X
t — 1
CM
, — 1
1
o
r- 1
X
ID
CM
en
3
o
0
1— 1
o
0
t — 1
1
0
1 — 1
X
0
•
CM
o
,—<
X
•=*
CO
<5f-
1
o
X
CO
•
CM
o
r— 1
1
0
1 — 1
X
CO
r1 1
<3-
1
0
t— 1
X
T-H
CM
T-l
1
o
t— 1
X
to
CM
_J
en
E
t— 1
CO
a>
o
co
E-4
-------
APPENDIX F
INTERNATIONAL AGENCY FOR RESEARCH ON CANCER CLASSIFICATION SYSTEM
FOR THE EVALUATION OF THE CARCINOGENIC RISK OF CHEMICALS TO HUMANS*
ASSESSMENT OF EVIDENCE FOR CARCINOGENICITY FROM STUDIES IN HUMANS
Evidence of cardnogenicity from human studies comes from three main
sources:
1. Case reports of individual cancer patients who were exposed to the
chemical or process.
2. Descriptive epidemiological studies in which the incidence of cancer
in human populations was found to vary in space or time with exposure to the
agents.
3. Analytical epidemiological (case-control and cohort) studies in
which individual exposure to the chemical or group of chemicals was found to
be associated with an increased risk of cancer.
Three criteria must be met before a causal association can be inferred
between exposure and cancer in humans:
1. There is no identified bias which could explain the association.
2. The possibility of confounding has been considered and ruled out as
explaining the association.
3. The association is unlikely to be due to chance.
In general, although a single study may be indicative of a cause-effect
relationship, confidence in inferring a causal association is increased when
several independent studies are concordant in showing the association, when
the association is strong, when there is a dose-response relationship, or when
a reduction in exposure is followed by a reduction in the incidence of cancer.
The degrees of evidence for carcinogenicity from studies in humans are
categorized as: ,
1. Sufficient evidence of carcinogenicity, which indicates that these
is a causal relationship between the agent and human cancer.
Adapted from International Agency for Research on Cancer Monographs Supple-
ment 4, Evaluation of the Carcinogenic Risk of Chemicals to Humans, 1982,
pp. 11-14.
F-l
-------
2. Limited evidence of carcinogenicity, which indicates that a causal
interpretation is credible, but that alternative explanations, such as chance,
bias, or confounding, could not adequately be excluded.
3. Inadequate evidence, which indicates that one of three conditions
prevailed: (a) there were few pertinent data; (b) the available studies, while
showing evidence of association, did not exclude chance, bias, or confounding;
(c) studies were available which do not show evidence of carcinogenicity.
ASSESSMENT OF EVIDENCE FOR CARCINOGENICITY FROM STUDIES IN EXPERIMENTAL ANIMALS
These assessments are classified into four groups:
1. Sufficient evidence of carcinogenicity, which indicates that there .is
an increased incidence of malignant tumors: (a) in multiple species or strains;
or (b) in multiple experiments (preferably with different routes of administra-
tion or using different dose levels); or (c) to an unusual degree with regard
to incidence, site or.type of tumor, or age at onset. Additional evidence may
be provided by data on dose-response effects, as well as information from
short-term tests or on chemical structure.
2. Limited evidence of carcinogenicity, which means that the data sug-
gest a carcinogenic effect but are limited because: (a) the studies involve a
single species, strain, or experiment; or (b) the experiments are restricted
by inadequate dosage levels, inadequate duration of exposure to the agent,
inadequate period of follow-up, poor survival, too few animals, or inadequate
reporting; or (c) the neoplasms produced often occur spontaneously and, in the
past, have been difficult to classify as malignant by histological criteria
alone (e.g., lung and liver tumors in mice).
3. Inadequate evidence, which indicates that because of major qualita-
tive or quantitative limitations, the studies cannot be interpreted as showing
either the presence or absence of a carcinogenic effect; or that within the
limits of these tests used, the chemical is not carcinogenic. The number of
negative studies is small, since, in general, studies that show no effect are
less likely to be published than those suggesting carcinogenicity.
4. No data indicate that data were not available to the Working Group.
The categories, sufficient evidence and limited evidence, refer only to
the strength of the experimental evidence that these chemicals are carcinogenic
and not to the extent of their carcinogenic activity nor to the mechanism
involved. The classification of any chemical may change as new information
becomes available.
F-2
-------
EVALUATION OF CARCINOGENIC RISK TO HUMANS
At present, no objective criteria exist to interpret data from studies in
experimental animals or from short-term tests directly in terms of human risk
Thus, -m the absence of sufficient evidence from human studies, evaluation of
the carcmogenic risk to humans was based on consideration of both the epide-
miological and experimental evidence. The breadth of the categories of evi-
dence defined above allows substantial variation within each. The decisions
reached by the Working Group regarding overall risk incorporated these differ-
ences, even though they could not always be reflected adequately in the
placement of an exposure into a particular category.
The chemicals, groups of chemicals, industrial processes, or occupational
exposures were thus put into one of three groups:
Group 1
The chemical, group of chemicals, industrial process, or occupational
exposure is carcinogenic to humans. This category was used only when there
was sufficient evidence from epidemiological studies to support a causal
association between the exposure and cancer.
Group 2
The chemical, group of chemicals, industrial process, or occupational
exposure is probably carcinogenic to humans. This category includes exposures
for which, at one extreme, the evidence of human carcinogenicity is almost
sufficient," as well as exposures for which, at the other extreme it is
inadequate. To reflect this range, the category was divided into higher
(Group A) and lower (Group B) degrees of evidence. Usually, category 2A was
reserved for exposures for which there was at least limited evidence of car-
cinogenicity to humans. The data from studies in experimental animals played
an important role in assigning studies to category 2, and particularly those
in Group B; thus, the combination of sufficient evidence in animals and inade-
quate data in humans usually resulted in a classification of 2B.
In some cases, the Working Group considered that the known chemical
properties of a compound and the results from short-term tests allowed its
transfer from Group 3 to 2B or 'from Group 2B to 2A.
Group 3
The chemical, group of chemicals, industrial process, or occupational
exposure cannot be classified as to its carcinogenicity to humans.
F-3
*USGPO: 1984-559-111-10737
-------
-------
-------
cn
i
o
o
oo
i
oo
oo
o
IS)
~n
C 0. =S
2 2-c
ou do not wis
ach. or copy I
jer left-hand
O 3- D"
S 5 o
2 o -.
2 o »
r1 < o
a a
23'
-f
Is
M
51
o n
cu X
o. m
So
Si
* m
3 2
fm
P
— Q. -rj
^* S 5"
ase make all
ach or copy,
;-hand corner
™ 3
O. *
a v)
35
° §
53
IS
£ o
3 =
S 5
" 0
3 U
5 g
C
II
If
§1
ml
ol
§
2?
i!
Q) Ml
=f. ol
O3'
X
(71
O>
CO
------- |