United States
                Environmental Protection
                Agency
                Office of Health and
                Environmental Assessment
                Washington DC 20460
EPA-bUU/8-83-032A
October 1983
External Review Draft
                Research and Development
v>EPA
Health Assessment
Document for
Epichlorohydrin
Review
Draft
(Do Not
Cite or Quote)
                               NOTICE

                This document is a preliminary draft It has not been formally
                released by EPA and should not at this stage be construed to
                represent Agency policy It is being circulated for comment on its
                technical accuracy and policy implications

-------
                                    EPA-600/8-83-032A
                                           October 1983
                                            Review Draft
Health Assessment Document
                      for
           Epichlorohydrin
                     NOTICE

    This document is a preliminary draft. It has not been formally
    released by the U.S. Environmental Protection Agency and
    should not at this stage be construed to represent Agency policy.
    It is being circulated for comment on its technical accuracy and
    policy implications.
          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

-------
                               DISCLAIMER

     This report is an external draft for review purposes only and does
not constitute Agency policy.   Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
                                     11

-------
                                 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.
                                    iii

-------
                               CONTENTS
                                                                      Page
DISCLAIMER	    ii
PREFACE 	    iii
LIST OF TABLES 	    x
LIST OF FIGURES 	    xlii
AUTHORS, CONTRIBUTORS, AND REVIEWERS 	    xiv
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
          1.1.5  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	    2-1
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
                                   iv

-------
                           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

-------
                      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 	     4-9
     4.5  SUMMARY	     4-9
5.  EFFECTS ON HUMANS 	     5-1
     5.1  EPIDEMIOLOGIC STUDIES 	     5-1
     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  Inununocompetence	     5-5
     5.4  EFFECTS ON THE LIVER	     5-5
     5.5  EFFECTS ON THE SKIN	     5-6
          5.5.1  Case Studies	     5-6
          5.5.2  Sensitization	     5-8
     5.6  EFFECTS ON MALE FERTILITY	     5-9
     5.7  SUMMARY	     5-10
6.   ANIMAL TOXICOLOGY 	     6-1
     6.1  SPECIES SENSITIVITY 	     6-1
          6.1.1  Acute Toxicity	     6-1
                 6.1.1.1  Inhalation 	     6-1
                 6.1.1.2  Oral	     6-9
                 6.1.1.3  Subcutaneous Injection 	     6-10
                 6.1.1.4  IntrapeHtoneal 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  IntrapeHtoneal Injection	     6-31
                 6.1.2.4  Dermal 	     6-32
     6.2  EFFECTS ON THE LIVER, KIDNEYS, AND LUNGS 	     6-33
                                   vi

-------
                      CONTENTS  (Cont'd.)


          6.2.1  Liver 	     6-33
          6.2.2  Kidneys 	     6-34
          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 	     6-38

     6.5  SUMMARY 	     6-38

7.   CARCINOGENICITY, MUTAGENICITY, AND REPRODUCTIVE
     AND TERATOGENIC EFFECTS 	     7-1

     7.1  CARCINOGENICITY 	     7-1

          7.1.1  Introduction 	     7-1
          7.1.2  Animal Studies	     7-1
                 7.1.2.1  Inhalation Exposure:   Rat 	     7-1
                 7.1.2.2  Oral  Administration:   Rat 	     7-6
                 7.1.2.3  Dermal Exposure:  Mouse 	     7-12
                 7.1.2.4  Initiation - Promotion:  Mouse 	     7-12
                 7.1.2.5  Subcutaneous or Intraperitoneal
                           Administration:  Mouse	     7-13
          7.1.3  Epidemiologic Studies 	     7-14
          7.1.4  Quantitative Estimation 	     7-21
                 7.1.4.1  Procedures for Determination of Unit
                           Risk	     7-21
                 7.1.4.2  Description of the Low-Dose
                           Extrapolation Model  	     7-22
                 7.1.4.3  Selection of Data	     7-24
                 7.1.4.4  Calculation of Human Equivalent
                           Dosages from Animal  Data 	     7-25
                 7.1.4.5  Inhalation 	     7-27
                 7.1.4.6  Calculation of the Unit Risk from
                          Animal Studies 	     7-29
                 7.1.4.7  Adjustment for less than Natural
                           Lifetime Experiment 	     7-30
                 7.1.4.8  Interpretation of Quantitative
                           Estimates 	     7-30
                 7.1.4.9  Alternative Methodological Approaches..     7-31
                 7.1.4.10 Estimation of Unit Risk Based on
                           Human Data	     7-32
          7.1.5  Interpretation of Quantitative Estimates 	     7-33
                 7.1.5.1  Unit Risk Estimate Based on Human
                           Studies 	     7-33
                                   vii

-------
                      CONTENTS  (Cont'd.)


                 7.1.5.2  Unit  Risk Based on Animal Studies 	    7-36
                 7.1.5.3  Summary of Unit Risks 	    7-40
                 7.1.5.4  Relative Potency 	    7-41
          7.1.6  Summary	    7-46
                 7.1.6.1  Qualitative Assessment	    7-46
                 7.1.6.2  Quantitative Assessment 	    7-47
          7.1.7  Conclusions  	    7-48

     7.2  MUTAGENICITY	    7-49

          7.2.1  Introduction	    7-49
          7.2.2  Gene Mutations in Bacteria	    7-49
                 7.2.2.1  Salmonella Assay	    7-49
                 7.2.2.2  Mutations in Klebsiella 	    7-56
                 7.2.2.3  Host-Mediated Assay	    7-56
                 7.2.2.4  Body  Fluid Analysis	    7-57
          7.2.3  Bacterial DNA  Repair Tests 	    7-57
          7.2.4  Gene Mutations in Neurospora	    7-57
          7.2.5  Gene Mutations in Yeast	    7-58
          7.2.6  Gene Mutations in Mammalian Cell Cultures 	    7-58
          7.2.7  Sex-Linked Recessive Lethal Test in Drosophila ..    7-59
          7.2.8  Chromosomal Aberrations in Human and Other
                 Mammalian Systems 	    7-61
                 7.2.8.1  Studies on Human Chromosomes in
                           Vitro	    7-61
                 7.2.8.2  Studies on Rodent Chromosomes jn Vitro .    7-63
                 7.2.8.3  Studies on Human Chromosomes in Vivo ...    7-63
                 7.2.8.4  Studies on Rodent ChromosomesTn Vivo ..    7-64
                 7.2.8.5  Micronucleus Assay 	    7-66
                 7.2.8.6  Dominant Lethal Assay 	    7-66
                 7.2.8.7  Sister-Chromatid Exchange Assay 	    7-67
          7.2.9  Conclusions  	    7-68

     7.3  REPRODUCTIVE AND TERATOGENIC EFFECTS 	    7-69

          7.3.1  Reproductive Effects 	    7-69
                 7.3.1.1  Male  Clinical-Epidemiologic
                           Investigations 	    7-73
          7.3.2  Teratogenic Effects 	    7-74
          7.3.3  Summary and Conclusions	    7-75

8.  SYNERGISM AND ANTAGONISM AT THE PHYSIOLOGICAL LEVEL 	    8-1

9.  ECOSYSTEM CONSIDERATIONS	    9-1

     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 	    9-2

-------
                           CONTENTS  (Cont'd.)

     9.3  EFFECTS ON AQUATIC ANIMALS 	     9-3
          9.3.1  Freshwater Fish	     9-3
          9.3.2  Freshwater Invertebrates  	     9-6
          9.3.3  Saltwater Fish	     9-6
     9.4  SUMMARY	     9-6
10.   REGULATIONS AND STANDARDS	     10-1
     10.1  OCCUPATIONAL STANDARDS 	     10-1
     10. Z  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-l
APPENDICES
     A.  Evaporation Rate of Epichlorohydrin Calculated
         According to the Method of Dill ing (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) 	     C-l
     D.  Biconcentration Factors Calculated for
         Epichlorohydrin by Four Methods  	     D-l
     E.  Comparison of Results by Various Extrapolation Models ..      E-l
                                        IX

-------
                                 LIST OF TABLES
Table                                                                 Page
3-1       Physical and Chemical Properties of Epichlorohydrin 	     3-2
3-2       Typical Reactions of Epichlorohydrin 	     3-5
3-3       Rate Constants for Hydrolysis of Epichlorohydrin as a
          Function of Temperature 	     3-7
3-4       Rate Constants for Epichlorohydrin Reaction with
          Various Anions 	     3-8
3-5       Rate of Reaction of Epichlorohydrin (ECH) under
          Neutral and Buffered Conditions at 37°C 	     3-9
3-6       Estimation of Epichlorohydrin Production, 1978-1980,
          in Millions of Pounds 	     3-14
3-7       Domestic Consumption of Epichlorohydrin for 1977 	     3-15
4-1       Distribution of 14C-Radioactivity in Rat Tissue
          Following a 10 mg/kg Oral Dose of 14C-Epichlorohydrin ..     4-5
4-2       Tissue Distribution of Radioactive 14C-Epichlorohydrin
          and Metabolites in Rats 	     4-6
5-1       Illness Episodes in Epichlorohydrin Workers 	,     5-2
5-2       Chromosomal Aberration Frequency in Lymphocytes from
          Workers Exposed to Synthetic Resin ED-20 	     5-4
6-1       Acute Effects of Epichlorohydrin 	     6-2
6-2       Summary of Mortality Findings in Rats and Mice after
          Acute Inhalation Exposure to Epichlorohydrin 	     6-8
6-3       Acute Intraperitoneal Toxicity of Epichlorohydrin 	     6-12
6-4       Dermal Irritation Scores for Solutions of
          Epichlorohydrin in Cottonseed Oil 	     6-15
6-5       Subchronic Effects of Epichlorohydrin 	     6-19
6-6       Mortality in Mice Exposed to 2,500 ppm Epichlorohydrin .     6-25
6-7       Mortality of Mice Administered Epichlorohydrin Orally ..     6-31
6-8       Lethality Following Repeated Dermal Application of
          Epichlorohydrin in Rats 	     5-33
7-1       Squamous Cell Carcinomas of the Nasal Cavity Following
          Thirty 6-Hour Exposures to 100 ppm Epichlorohydrin  	     7-2

-------
                       LIST OF TABLES (Cont'd.)
Table
Page
7-2       Dose-response for Induction of Squamous Cell  Carcinomas
          in the Nasal Cavity of Male Wistar Rats Exposed to
          Epichlorohydrin Vapor 	     7-4

7-3       Kidney Weights and Kidney/Body Weight Ratios  in Male
          Wistar Rats Given Epichlorohydrin in Drinking Water
          for 81 Weeks 	     7-11

7-4       Comparison of Mortality in Enter!ine's Epichlorohydrin
          Study Updates by Cause and by Latency (1978 versus
          1981)	     7-15

7-5       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 	     7-16

7-6       Comparison of Mortality in Epichlorohydrin (ECH)
          Alone and Combined Exposure Groups in Deer Park,
          Texas 	     7-18

7-7       Relative Carcinogenic Potencies Among 53 Chemicals
          Evaluated by the Carcinogen Assessment Group as
          Suspect Human Carci nogens	     7-43

7-8       Induction of Sex-linked Recessive Lethals in
          Drosophila by Epichlorohydrin 	     7-60

7-9       The Effects of Epichlorohydrin on the Fertility of
          Wistar Rats 	     7-70

7-10      The Effects of Inhaled Epichlorohydrin on the Semen
          of Rabbits and on the Fertility of Male and Female
          Rats  	     7-71

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 LC5Q 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 Fish 	     9-5

                                   xi

-------
Table No.                                                            £§S§

10-1      Occupational Standards for Epichlorohydrin 	    1°"2

 E-l      Estimates of Epichlorohydrin Low-Dose Risk in Male
          Wistar Rats Derived from Four Different Models 	     t-3
                                    XI1

-------
                             LIST OF FIGURES
Figure                                                                page
4-1       Blood Concentrations of Epichlorohydrin in Mice
          after Intraperitoneal Injection of 200 mg/kg 	     4-3
4-2       Proposed Metabolic Pathways for Epichlorohydrin 	     4-7
7-1       Mortality of Rats Following Exposure to 100 ppm of
          Epichlorohydrin 	     7-3
7-2       Mortality of Rats Following Lifetime Exposure to
          Epichlorohydrin 	     7-5
7-3       Growth of Rats Following Chronic Exposure to
          Epichlorohydrin	     7-5
7-4       Patterns of Epichlorohydrin Administration in Male
          Wistar Rats 	     7-7
7-5       Intake of Epichlorohydrin in Drinking Water by Male
          Wistar Rats 	     7-9
7-6       Effect of Epichlorohydrin Treatment of Body-Weight in
          Male Wistar Rats 	     7-10
7-7       Histogram Representing Frequency Distribution of the
          Potency Indices of 53 Suspect Carcinogens 	     7-42
7-8       Mutagenicity of Aromatic Epoxy Resins and
          Epichlorohydrin for S. typhimurium TA100  	     7-51
7-9       Mutagenicity of Epichlorohydrin, Styrene Oxide, and
          DDNU-oxide at Various Concentrations  in S.
          typhimurium TA100 	     7-52
7-10      Dose-response Curves for Epichlorohydrin  	     7-53
7-11      Mutagenicity of Epichlorohydrin With  and Without S-9
          Mix	     7-54
7-12      Dose-Response Curve for Epichlorohydrin-Treated
          Cultures 	     7-59
                                  X111

-------
                   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 carcinogenic!ty.   The  principal authors of
 the carcinogenicity material  were Larry Anderson,  Ph.D. and
 Steven Bayard, 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  Pierano, 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-Salem, N.C.
   Derek Hodgson, Ph.D.
   Chemistry Department
   University of North Carolina
   Chapel  Hill, N.C.
                                    xiv

-------
P.O. Lotilaker, Ph.D.
Pels Research Institute
Temple University Medical Center
Philadelphia, PA

P.G. Watanabe, Ph.D.
Toxicology Research  Laboratory
Health and Environmental Sciences
DOW Chemical USA
Midland, MI
                                 xv

-------
                             1.   EXECUTIVE SUMMARY

1.1  BACKGROUND INFORMATION
1.1.1  Properties
     Eplchlorohydrin  (l-ch1oro-2,3-epoxypropane)  is  a colorless liquid with
a characteristic  chloroform-like,  irritating odor.   It  is  slightly soluble
in 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
     Epichlorohydrin  is  produced commercially  by high  temperature  chlori-
nation of propylene to allyl chloride,  followed by chlorohydration  with hypo-
chlorous acid to  form a mixture of  isomeric  glycerol dichlorohydrins.   The
mixture  is  subsequently dehydrochlorinated with  alkali  to  yield epichloro-
hydrin.   Epichlorohydrin  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)  (Blackford,  1978).  In  1980,  300 million pounds were
produced (U.S. EPA, 1983).
1.1.3  Use
     Epichlorohydrin's 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 is
                                    1-1

-------
not expected  to  persist in air, water,  or soil  because of  its  tendency to
nydrolyze  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 these 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  ANIMAL TOXICITY
     Epichlorohydrin is well  absorbed  and moderately toxic by oral, dermal,
and inhalation routes.  The acute oral dose lethal to  50% of rats exposed
(LD50) to epichlorohydrin  was approximately 250 mg/kg  body weight.  The in-
halation 6-hours LC50  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 inhala-
tion, oral,  and  intraperitoneal  injection routes studies caused severe renal
toxicity, which can be reversed on cessation of exposure.  Epichlorohydrin was
intensely irritating to skin, nasal mucosa, and eyes; in addition, it can cause
skin sensitization in laboratory 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  Careinogenicity
     Results of long-term animal  studies provide some evidence of the carcino-
genic potential of epichlorohydrin.  There are no epidemiologic studies avail-
able which have demonstrated  epichlorohydrin to be carcinogenic to humans.  In
view of increases in nasal  carcinomas seen in rat inhalation tests, the increased
local sarcomas produced in  mice after subcutaneous injection of epichlorohydrin,
and the chromosomal aberrations found in the peripheral  lymphocytes of exposed
workers, a  recommendation  that epichlorohydrin be considered  as  a potential
human carcinogen appears to be prudent.  Due to the unknown length of the latent
period that  may  precede  the  appearance of tumors, there  are grounds for con-
tinuing to  follow  the exposed workers, giving attention  to lifestyle, ethnic
origin, and the possible contributions of occupational  exposure to other agents.
1.5.2  Mutagem'city
     Substantial evidence   is available  demonstrating  that  epichlorohydrin
causes gene  and  chromosomal mutations  in several  experimental  systems  both jn
vitro and in animals.  Cytogenetic studies of workers exposed to epichlorohydrin
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
                                    1-3

-------
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  in  vivo  and |n yitro 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, but recovery  followed  termination of exposure.  The effect seems to be
related  to  immobilization of  spermatozoa in the  epididymis.   However, no
detrimental  effects  were  observed  on  the  fertility of male workers exposed to
epichlorohydrin in the manufacture of glycerine.
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
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.
                                    1-4

-------
1.8  REGULATIONS AND STANDARDS
     ipichlorohydrin 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
     Based on health and exposure-related data, the U.S. Environmental Protec-
tion Agency  is  considering listing epichlorohydrin as  a hazardous pollutant
under  Sections  111 and  112 of the Clean Air  Act.   Epichlorohydrin is an
appropriate candidate for such an assessment because of its alkylating proper-
ties, its mutagenicity in a variety of systems, and its carcinogenic potential
in mammals.  Moreover, increased chromosomal aberrations have been reported in
peripheral lymphocytes of workers exposed to epichlorohydrin.  Recent epidemi-
ologic  studies  on exposed  workers  indicate that epichlorohydrin  should  be
considered a potential carcinogen.  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
epichlorohydrin 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.
          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.
                                    1-5

-------
          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 route.

          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.

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 DMA in target organs.

                                    1-6

-------
                                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, p-propylene oxide and has
the formula:
                                 A
                                   - 1CH - CH2C1
     It is a clear, colorless, unstable 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:
     ECH                                2-chloromethyl oxyrane
     ECHH                               glycidyl chloride
     l-chloro-2,3-epoxypropane          (chloromethyl) oxirane
     3-chloro-l,2-epoxypropane          3-chloro-l,2-propylene oxide
     (chloromethyl) ethylene oxide      crepichlorohydrin
     2-(chloromethyl) oxirane           (DL)-crepichlorohydrin
     chloropropylene oxide              SKEKhG
     T-chloropropylene oxide            l,2-epoxy-3-chloropropane
     3-chloropropene 1,2-oxide          2,3-epoxypropyl chloride
                                        glycerol epichlorohydrine
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:  C3HKOC1
     Molecular Weight:   92.53
     Elemental
     Composition:        C = 38.94X
                         H«  5.45%
                        Cl = 38.3Z%
                         0 = 17.29%
PHYSICAL PROPERTIES
Melting Point (Weast 1978)
Freezing Point (Shell 1969)
               (Dow 1980)
Boiling Point (Shell 1969)
               (Dow 1980)
Density (g/«l, 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)
1 ppm (25° C, 760 mmHg) (Hine et al. 1981)
1 mg/1 (25° C, 760 mmHg) (Hine et al. 1981)
Heat Capacity (25° C) (Dow 1980)
              (100° C) (Oow 1980)
Heat of Formation (25° C) (Oow 1980)
Explosive Limits (volume % in Air) (Oow 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 mnHg)
  20          S6
dL,  1.1812; d,  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
nD1.4358
3.78 mg/m3
265 ppm
31.5 cal/mol0 C
40.0 cal/mol0 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).
     No ultraviolet spectrum was found listed in the literature for epichloro-
hydrin.  Based upon its structure, which combines alkyl halide and alky! 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 charac-
ter 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 mole-
cule 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
additive 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 weeks1 time under
laboratory conditions  (Brfinsted 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 3.1.5.3).
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 and
electronic 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 (Bronsted 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 am'on, or other
nucleophile at the C-l carbon (Brdnsted et al. 1929; Kwart and Goodman 1960; Long
and Pritchard 1956; Addy and Parker 1965):
                                    3-4

-------
                     TABLE 3-2.  TYPICAL REACTIONS OF EPICHLOROHYDRIN
Monohydric

 Alcohols
                              0
  /\

CH2-CH-CH2C1 + ROM
                                                  Catalyst
                                    > R-0-CH2-CHOH-CH2C1
Organic Acids
Acyl Chlorides
  /\

CH2-CH-CH2C1 + RCOOH
  A
CH2-CH-CH2C1
               RCOC1
                                      RCOO-CH2-CHOH-CH2C1



                                      HO-CH2-CHOOCR-CH2C1







                                      CH2C1-CH(OOCR)-CH2C1
Aldehydes
                              .0
CH2-CH-CH2C1 + RCHO
                                            V
                                            ox o
                                            I  1
                                      C1CH2-CH-CH2
Ami nes
   w





CH2-CH-(
         »C1 + RHNH
                                      RHNCH2-CHOH-CH2C1
Grignard Reagents
CH2-CH-CH2C1 + RMgBr
                                                   HoO
CH2R-CHOH-CH2C1
                           CH2OH-CHR-CH2C1 + HgBr2
                                    •*  CH2R-CHOMgBr-CH2Cl
                                                                  CH2OMgBr-CHR-CH2Cl
Water
   0


  A
 %if — PLI—I
 *no Un I
               HOH
                                    •* CH2OH-CHOH-CH2C1
Inorganic Acids
   0

  /\
CH2-CH-CH2C1
               HC1
                                    •*  CH2C1-CHOH-CH2C1
Source:  DOW (1980)
                                           3-5

-------
            u
           A
                                       0"
                                                   OH
     C1CH2CH-CH2 *  A
                        C1CH2CH-CH2 A —2	»  C1CH2CHCH2A  * OH~
The acid-catalyzed reactions have been Identified as having an A-2 type mechanism.
This mechanism may depend on the strength of the acid.   In strong add the reaction
could be unimolecular with the opening at the C-2 and anion.   In these, epichlo-
rohydrin 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):
       u
     A
C1CH2CH-CH2
                 H
                0
              +
          C1CHCH-CH
              A
               i
          C1CH2CHCH2OH
                                          H
                                          0
                             ,-       A
                                   C1CH2CH-CH2
                                 I
                                 H
                                 0
                           C1CH2CH-CH21
                          '      I
                                OH
                           C1CH2CHCH2
     Most of the information necessary for the product and half-life calculations
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 (kj is for the uncatalyzed addition of water,
and k2 is for the acid-catalyzed addition of water).  Table 3-4 lists the experi-
mentally derived rate constants, k3 and k^, for anion reactions with epichlorohy-
drin (k3 is for the uncatalyzed addition of an anion, and k4 is the acid-catalyzed
addition of an anion).
                                    3-6

-------
       Table 3-3.   Rate Constants for Hydrolysis of Epichlorohydrin  as  a
                           Function of Temperature
Temperature
0.0

20.0


25.0


35.0
37.0

45.0
50.0
75.0
85.0
(s"1)


0.97





5.9
5.3

13.8
20.4
129
246
5 C
10 k2 Reference
(liter mol s~ )
6.91 Pritchard and Siddiqui
(1973)
Pritchard and Siddiqui
43.4 (1973)
BrSnsted et al. (1929)
1e Noble and Duffy (1964)
68.2, 77 Pritchard and Siddiqui
(1973)
Shvets and Aleksanyan (1973)
Ross (1962 as cited in
NIOSH 1976a)
Shvets and Aleksanyan (1973)
Shvets and Aleksanyan (1973)
Shvets and Aleksanyan (1973)
Shvets and Aleksanyan (1973)
aReaction conditions are indicated in each reference.
 Rate for uncatalyzed addition of water.
GRate for acid-catalyzed addition of water.
                                    3-7

-------
  Table 3-4.  Rate Constants for Epichlorohydrin Reaction with Various Anions
Anion
Chloride, Cl"1

Iodide, l"1
Thiosulfate, S«0,*2
* 2 3
Formate, HCO«
— T
Benzoate, CgHgCOp
Acetate, CH-CO/1
3 2
Nitrate, NO,"
Bicarbonate, HC03


-2
Carbonate, CO,
3


oC
20
40
20
20

20
20
20
37
20
65
75
80
85
35
45
50
60
105k3a 10<"k4b
.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
0.309
0.529,
0.689
0.429
0 83
i:««
2.5a
a Rate for uncatalyzed addition of anion.
 Rate for acid-catalyzed addition of anion.
cBr8nsted 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 (BrSnsted et al.  1929).   Ross (1950) reported the rates
of flfit-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
Neutral Condition
Duration of
Reaction (h)
24
48
72
100
169
% ECH
Reacting
36.5
59.0
72.5
84.5
96.5
kx (h"1)
0.0190
0.0185
0.0180
0.0185
0.0200
Buffered Condition
% ECH
Reacting
52.0
76.6
88.0
96.0
100.0
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:
                      f\
                    CH2-CHCH2C1
HO
                                     H2°
     8
C1CH2C-OH
                    CH2CHCH2C1
                                      HI
                                     3-9
       -* C1CH2CH2CH3

-------
     Epichlorohydrin can be oxidized by free radical processes in liquid (Dobbs
et al. 1976; Beckwith 1982) or gas phases (Dilling 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, (CH,)3CO, 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, Billing 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   (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.
     A hydrochloric  acid-in-dioxane  method (Dobinson  et  al.  1969)  can be used
either by direct  sampling  into  the reagent mixture  or by sampling  aliquots  of
a bottle-collected specimen.   Infrared absorption  in the frequency range of
1,240 to 1,260 cm   will identify the characteristic  oxirane group of epichlo-
rohydrin.  A method  for determination  (in  the  range  of 10 to  1,000 ppm in
aqueous solution) involves  sampling  air through the absorbers containing 0.5 N
alcoholic potassium  hydroxide.    The absorbent solutions  are  combined and
refluxed and the resulting  chloride  is  titrated potentiometrically with silver
nitrate (Dobinson et al. 1969).
     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
                                     3-11

-------
vapor 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
         3
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/m  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
(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 dichloro-methane eluent was between 99 and
100 percent,  with a range of standard deviation 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 ug 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.
                                    3-12

-------
     Epichlorohydrin may also be determined in water samples by extracting the
samples with carbon tetrachloride (Adamek and Peterka 1971).  The carbon tetra-
cnlorlde 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.
3.3  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.  The third  step  is the hydrolysis of
epichlorohydrin to glycerine.  Crude epichlorohydrin may be transferred direc-
tly 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;
                                    3-13

-------
Norco, Louisiana) and Dow Chemical U.S.A. (Freeport, Texas) by the chlorohydri-
nation of ally! 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).
     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).

              Table 3-6  Estimation of Epichlorohydrin Production,
                        1978-1980, in Millions of Pounds.

Glycerine Feed
Refined Feed
Total
1978
-"60
265
325
1979
—49
310
350
1980
-
-
300
Source:  U.S. EPA 1983

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).
                                    3-14

-------
          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
3.3.3.1  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-hydroxyphenyl) 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   Textiles—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
                                     3-15

-------
preparation  of  dyeable propylene  fibers and  in  the  dyeing of polyolefin,
polyacrylonitrile, polyvinyl  chloride,  polyvinyl  alcohol, and other fibers.
It has  also  been used to impart wrinkle resistance and to prepare 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
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 caustic soda 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
                                    3-16

-------
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   Epichlorohydrin-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
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 jackers, 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.
                                    3-17

-------
3.3.3.10   Other Current Uses—A  variety of other products are  produced  from
epichlorohjtfirin,  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
           -   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 Kaila 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).
                                    3-18

-------
     The unique properties of epoxy resins and epichlorohydrin elastomers are
difficult to replace, especially if the use of closely related chemicals such
as epibromohydHn 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
                                                       5                 4
facilities in the United States totaled about 1.47 x 10  pounds (6.7 x 10  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
United States during  the  production of epoxy resins; these emissions totaled
               55                                   4
about 2.5 x 10  pounds/year (1.1 x  10   kg/year).   An additional  8.1 x  10
                     A
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.
                                    3-19

-------
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.
3.4  ENVIRONMENTAL TRANSPORT AND FATE
     The ultimate environmental fate of epichlorohydrin depends  on its release,
transport,  and persistance  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;
                                    3-20

-------
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 Oil ling 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 (Koc) 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 K   values range
from 10.28 x lO"1'23 (0.61) to 10.28 x 101'23 (174) at a solubility  of 60,000
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 K   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  KOC of  100, is pre-
dicted 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
                                    3-21

-------
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 3.1.5.1 and 3.1.5.2.
     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
3.1.5.3.
3.4.2.2   Bi ological 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).
                                    3-22

-------
     Information on microbial  biodegradation  of epichlorohydrin is limited.
Epichlorohydrin was identified as an intermediate in the enzymatic hydrolysis
of 2,3"d1bromo-l-propanol when chloride ions were present (Castro and Bartinicki
1968).  Unspecified cultures  of Flavobacterium 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  as follows:
                                 0                      OH

                H + Cl" + BrCHCH-CH^         r BrCH2CHCH2C1
                                                 A
                                                CH2-CHCH2C1 + Br~ + H*
     The biochemical  degradation of epichlorohydrin was  studied by Bridie at
al.  (1979a)  using unidentified seed cultures.   They reported the theoretical
oxygen  demand  (TOD),  biochemical  oxygen demand (BOD),  and the chemical  oxygen
(COD).  Five-day  BOD's were measured using  the American Public Health Associa-
tion's  Standard  Method No.  219 published in 1971.   The method was  modified by
adding  0.5 mg/1  allylthiourea  to  prevent nitrification.   The seed cultures
were obtained from a biological sanitary waste treatment  plant, and duplicates
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

                                    3-23

-------
epichlorohydrin using unadapted and adapted seed culture were 3 and 14 percent
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
                               5                 4
were estimated to  be 1.47 x 10  pounds (6.7 x 10   kg).  Emissions from epoxy
                                                      5                 5
resin production in 1978 were estimated to be 2.5 x 10  pounds (1.1 x 10  kg).
                      4                 4
An additional 8.1 x 10  pounds (3.7 x 10  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.
                                    3-24

-------
4.  COMPOUND DISTRIBUTION  AND  RELATED PHARMACOKINETICS IN HUMANS AND ANIMALS

     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-epichlorohy-
drin 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 concentra-
tions of radioactivity  in tissues, fluids, and excreta were measured.  Epichloro-
hydrin was rapidly absorbed from  the  gastrointestinal  tract.  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 was 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 in-
halation routes and studied the pharmacokinetics  of absorption,  distribution, and
excretion.  Single oral doses  of  1 mg/kg  or 100 mg/kg  l,3-14C-epichlorohydrin 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  containing 1 ppm or 100 ppm
l,3-l4C-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 percent was  exhaled as carbon dioxide.   These
experiments indicate that epichlorohydrin is absorbed  well from  the gut  or the
                                    4-1

-------
lungs, is  rapidly distributed to  other tissues, and much of the administered
epichlorohytrin 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 ijn  vivo half-life of epichlorohydrin is extremely short,
being only just detectable after 15 minutes.
               10.00
            "5»
            a.
            Z
            O
            i
            cc
            Z
            Ul
            O
            O
            O
1.00
0.10
                   E' i I i  [ I •  I i  I  I i  i  i I r I i  i| I I I I | l  I 11=
               0.01
                LIMIT OF ASSAY SENSITIVITY •
     I  I I  I  I I  I 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.
     Epichlorohydrin  is  readily absorbed  by the dermal  route.   Kremneva and
Tolgskaya (1961) immersed  the tails of 10 mice (to 75 percent of their lengths)
in epichlorohydrin for 1 hour.   Seven mice died within 3 days of the single  expo-
sure.  Similarly, when exposure was for 20-30 minutes on 2-3 successive days, all
10 of the mice died.
     Comparably toxic doses  for the oral,  subcutaneous, and 'inhalation routes were
also reported by Kremneva  and Tolgskaya (1961).  A single oral dose (by gastric
                                     4-2

-------
intubation) of 325 mg/kg was lethal for all of a group of 10 mice, whereas none
of a group of 10 mice receiving 250 mg/kg died or showed signs of toxicity within
2 weeks.   By  the  subcutaneous route, 10 of 10 mice  receiving 375 mg/kg as a
single dose died; 7 out of 10 mice given 250 mg/kg died; and a dose of 125 mg/kg
was tolerated by a group of 10 mice.  Groups of 10 mice were exposed to epichlo-
rohydrin vapors for a single 2-hour exposure.   All animals exposed to 6-9 mg/1
died, whereas 40 percent of the animals exposed to 2.5-4.0 mg/1  died.  None of
the animals exposed to concentrations of 1.2 mg/1 or lower showed signs of toxi-
city.
     It can be concluded from this study that epichlorohydrin was absorbed well
when administered to mice by oral, dermal, subcutaneous, or the inhalation routes.
There were, however, no quantitative data on the rates or efficiency of absorption
by oral, dermal, or subcutaneous routes.  Other studies examining the acute toxi-
city of epichlorohydrin  by various routes of exposure are discussed in other
sections of this document.
4.2  DISTRIBUTION
     The tissue distribution  of 14C in  rats  receiving a  single  oral  dose of
14C-epich1orohydrin (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  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.
In the  course of the 72-hour experiment,  exhaled radioactive carbon dioxide
accounted for 18 and 21 percent of the radioactive dose in female and male rats,
respectively.   Excretion in urine over the 72 hours  of the study accounted for
38-40 percent of the radioactivity, and  fecal excretion was less than 4 percent
of the administered dose.  This study seems to indicate a rapid biotransformation
of epichlorohydrin.
     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
                                    4-3

-------
    TABLE 4-1.  DISTRIBUTION OF 14C-RADIOACTIVITY IN RAT TISSUES FOLLOWING
                 A 10 mg/kg ORAL DOSE OF 14C-EPICHLOROHYDRIN
Tissue
Kidneys
Liver
Pancreas
Adrenals
Spleen
Others6
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.
GFour-hour sample.
 Peak value was at 8 hours.
ePeak value was at 4 hours.
 Lungs, heart, brain, fat, muscle, skin, ovaries, and/or testes.
Source:  Weigel et al.  (1978).

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,
                                    4-4

-------
              TABLE 4-2.   TISSUE DISTRIBUTION OF RADIOACTIVE  14OEPICHLOROHYDRIN AND METABOLITES IN RATSC
ORAL
100 mg/kgb,.
Timed
Selected
Tissues
Stomach
Small instestine
Kidneys
Large instestine
Lacrimal gland
Liver
Lungs
Brain
Plasma
Testes
Adrenal s
Heart
Nasal turbi nates
Muscle
Fat
3- hour
Post**


100
0-hour
Tissue
ug Eq/g
1047.99
154. 53
110.12
95.61
89.01
84.07
64.89
38.17
36.12
34.69
32.05
30.92
28.91
23.71
10.30
Plasma
(29.
( 4.
( 3.
( 2.
( 2.
( 2.
( 1.
( 1.

( 0.
( o.
( o.
( o.
( o.
( o.
01>d
28)d
05).
65)d
46)
33).
80)d
06)

96)
89)
86>d
80)d
66)
29)
ug
6
20
60
75
70
51
21
19
18
16
32
17
94
11
4
Eq/g
.20
.19
.40
.17
.92
.34
.30
.54
.29
.59
.09
.57
.01
.67
.36
INHALATION 6-HOUR EXPOSURE
b
PPm r
Postc
Tissue
Plasma
(0.34)
(1.10)
(3.30)
(4.11)
(3.88)
(2.81)
(1.16)
(1.07)

(0.91)
(1.75)
(0.96)
(5.14)
(0.64)
(0.24)
100 ppm „
3-hour

ug
9.
28.
42.
37.
70.
.42.
13.
10.
10.
8.
14.

19.
5.


Eq/g
59
99
76
24
44
82
69
48
39
72
03

79
91

Post**
Tissue
Plasma
(0.92)
(2.79)
(4.12)
(3.58)
(6.78)
(4.12)
(1.32)
(1.01)

(0.84)
(1.35)

(1.90)
(0.57)

100 ppmC
24- hour Post

ug Eq/g
5.00
15.04
26.18
11.81
95.50
24.29
8.30
5.23
7.35
4.43
13.10

13.63
6.12

Tissue
Plasma
( 0.68)
( 2.05)
( 3.56)
( 1.61)
(12.99)
( 3.30)
( 1.13)
( 0.71)

( 0.60)
( 1.78)

( 1.85)
( 0.83)

 Adult male Fischer 344 rats weighing 190-220 g;  two rats for oral  exposure and three  for  each  of the
 other exposures.
 Dose level administered or inhalation concentration for 6 hours.
 Time rats were killed after dosage or after 6-hour inhalation exposure.
 One rat only.
Source:  Smith et al.  (1979).

-------
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
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 dehydrochlon"native cyclization of  3-chloro-l,2-propanediol.
                                    4-6

-------
                                                                        HjC
                                                   GSHb
                          CH2SCH2CHCOOH
                         CHOH


                         CH2CI
     CHOH


     CH^C^CHCOOH

              NH2



 1.3-(bis-cyipropan2-ol


             ACETYLATION


      CH2SCH2CHCOOH


     CHOH   NHCOCH3

      I
^   CH^C^CHCOOH

 ^            NHCOCH3


 1,3-(bls-N-acatyl cysteinyllpropan 2-ol
                                  I
                                             02
                                                                          CH2CI


                                                                  . EPICHLOROHYDRIN .
                                   S-(2.3-dihydroxypropyl)cysteine
                                                                                  EPOXIDATION
:H2SCH2CHCOOH
1
NH2
:HOH
;H2OH2
• GSH
- glutamate
- glycine
/
CH2OH
1
CH
CH2
glycidol
                                                               ACETYLATION

                                                               •*.
                                                               CH2SCH2CHCOOH


                                                               CHOH   NHCOCH3

                                                                I
                                                               CH2OH





                                                         N acetyl S42.3dihydro«ypropnnolk:y5IBine
HYDRATION
"*<•»• ATP ADP
CH2OH V__^T
I
Cftn" >
I
CH2CI
p— —


CHOH
1
_CH2CI
                                                                                                   3-chloro 1.2 propanediol
                                          3-chloroglycerophoiphate
       COOH


       CHOH


       CH2CI





beta chlorolactic acid



         02
                             COOH

                             I
                             COOH
                           oxalic acid
Compounds In brackets are hypothesized

bGlutathiona
                                         Figure 4.2 Proposed metabolic pathways for epichlorohydrin.

                                         Source: Adapted from Fakhouri and Jones (1979) and Jones and O'Brien (1980).

-------
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-ch1oro-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.
4.4  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 l^-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 1*C02 was initially rapid; by 4 hours, 8 percent of thdL
dose and by 8 hours^, 14 percent of the dose appeared as expired  14CO«.   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-epich1orohydrin administered orally  or by inhalation was excreted as
14C02, while 46-56 percent of the radioactivity was excreted  in  the urine by

                                    4-8

-------
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 semilogarithmic 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,
high tissue levels were found in the  nasal turbinates,  lacrimal glands, kidneys,
large intestine, and liver.  The major routes  of excretion 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  accounted  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
                                    4-9

-------
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.
                                    4-10

-------
                             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, cytogenic changes  in  heraatopoietic tissues,  and liver dysfunction.
Epichlorohydrin is  known to cause delayed skin  burns, and it  may  cause skin
sensitization reactions.
5.1  EPIOEMIOLOGIC 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 Kill an (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

                              Minimal Exposure              Moderate Exposure
No. of Employees                     213                           49
Total Episodes of Illness          1,343                          193
Episodes/Employee                    6.3                          4.0
Respiratory Illnesses
  In Exposure Areas                  254 (19%)                     57 (30%)
  In Nonexposure Areas               231 (17%)                     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
        5
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
                                                                       •3
on alpha-rhythm  bursts.  Two subjects were exposed to 0.2 and 0.3 mg/m  epi-
chlorohydrin,  and the  cerebral  biopotentials were recorded using an electro-
encephalograph (EEG).  An  epichlorohydrin concentration of 0.3 mg/m   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  10/1)  in  five workers,  decreased hemoglobin concentrations  (10.8-13.2
                                                                      9
g/100 ml)  in 16  workers,  and decreased leukocyte counts (3.4-4.4 x 10 /I) 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-
pationally exposed to epichlorohydrin for 2 years (estimated air concentration
0.5-5 mg/m3).  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  epichldrohydrin  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
                  Average Chromosomal
                 Aberration Frequency
Aberrant Chromosomes
     per Cell
 Breaks per
Group Aberrant
  Chromosomes
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 jn vitro effects of epichlorohydrin on
human lymphocytes.   Both T-cell and  B-cell  responses were studied.  Human
peripheral blood lymphocytes were separated by Ficoll-Hypaque gradient centri-
fugatlon, suspended in medium plus serum, and distributed in wells of microtiter
plates.   Mitogenic response was measured by incorporation of tritiated thymidine
  3
([ H]TdR) into  DNA.   A  variety  of  mitogens were used:  phytohemagglutinin
(PGA-P,   0.053>)  stimulated mainly mature T-cells;  Concanavalin A (Con A, 5
ug/well) stimulated both immature  and mature T-cells; pokeweed  mitogen  (PWM
tf») 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 iji  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 days 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
urob111nogen 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, esterified  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 incident, 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 (sec
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  EFFECTS ON MALE FERTILITY
     Venable et  al. (1980)  studied  the fertility status of male  employees
engaged in  the manufacture  of glycerine, who were exposed  to epichlorohydrin,
ally! chloride,  and 1,3-dichloropropane at the  Freeport,  Texas, Division  of
Dow Chemical Company.  Sixty-four exposed workers were compared with 63 control
volunteers.   Reproductive medical histories were  taken,  and  the  laboratory
studies included blood hormone analysis and analysis of semen specimens (volume,
viscosity,  percent  progressive sperm,  percent motile  sperm,  sperm count,
percent viable sperm,  and percent normal sperm forms).   The results showed no
detrimental  effects on fertility from exposure to the chlorinated  three-carbon
compounds including  epichlorohydrin.   Milby and Whorton (1980) also found no
sperm count suppression  among  workers exposed to epichlorohydrin  in contrast
to parallel  observations made with l,2-dibromo-3-chloropropane workers.
                                    5-9

-------
5.7  SUMMARY
     Epiclrlorohydrin 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.
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 epidemiclogic 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 metered, 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

-------
                                            TABLE 6-1.  ACUTE EFFECTS OF EPICHLOROHYORIN
       Route
               Species
Dose
Effect
Reference
rss
Inhalation     Rat         250 ppm for 4 h
Inhalation     Rat         250 ppm for 8 h
Inhalation     Rat         273-316 ppm for 2 h
Inhalation     Rat         360 ppm for 6 h
Inhalation     Rat         590-944 ppm for 2 h
Inhalation     Rat         631 ppm for 4 h
Inhalation     Rat         1,062-1,416 ppm for 2 h
Inhalation     Rat         1,416-2,124 ppm for 2 h
Inhalation     Mouse       237-316 ppm for 2 h
Inhalation     Mouse       590-594 ppm for 2 hr
Inhalation     Mouse       789 ppm for 2 hr
Inhalation     Mouse       1,062-1,416 ppm for 2 hr
Inhalation     Mouse       1,416-2,124 ppm for 2 hr
Inhalation     Mouse       2,370 ppm for 1 hr
Inhalation     Mouse       7,414-16,600 ppm for 0.5 hr
Inhalation     Mouse       18,097 ppm
                       2-4/6 dead
                       4/6 dead
                       Not lethal
                       50% mortality
                       62% mortality
                       50% mortality
                       80% mortality
                       Lethal concentration
                       Not lethal
                       40% mortality
                       50% mortality
                       93% mortality
                       Lethal concentration
                       Not lethal
                       Lethal concentration
                       50% mortality in 9.13 min
                          Carpenter et al.  (1949)
                          Weil et al.  (1963)
                          Kremneva (1960)
                          Laskin et al.  (1980)
                          Kremneva (1960)
                          Grigorowa et al.  (1977)
                          Kremneva (1960)
                          Kremneva (1960)
                          Kremneva (1960)
                          Kremneva (1960)
                          Grigorowa et al.  (1977)
                          Kremneva (1960)
                          Kremneva (1960)
                          Freuder and Leake (1941)
                          Freuder and Leake (1941)
                          Lawrence et al. (1972)

-------
CO

Route
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Oral
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Subcutaneous

Species
Rat
Rat
Rat
Rat
Rat
Rat
Mouse
Mouse
Mouse
Mouse
Rat
Mouse
Guinea Pig
Rabbit
Rat
TABLE 6-1.
Dose
65 mg/kg
125 mg/kg
248 mg/kg
250 mg/kg
260 mg/kg
325, 500 mg/kg
236 mg/kg
250 mg/kg
271 mg/kg
325, 350, and 590 mg/kg
118 mg/kg
165 mg/kg
118 mg/kg
165 mg/kg
65 mg/kg
(continued)
Effect
Polyuria, proteinuria,
reduced urinary chloride
Polyuria, proteinuria,
reduced urinary chloride
increased creatinine
LD5d
Polyuria, proteinuria,
increased urinary
chlorides and creatinine
LD50
L0100
LD50
Not lethal
Not lethal
LD100
LDso
LD50
LD50
LD50
Polyuria, proteinuria,

Reference
Shumskaya and Karamzina
(1966)
Shumskaya and Karamzina
, (1966)
Smyth et al. (1962)
Shumskaya and Karamzina
(1966)
Lawrence et al. (1972)
Kremneva (1960)
Lawrence et al. (1972)
Kremneva (1960)
Freuder and Leake (1941)
Kremneva (1960); Freuder
and Leake (1941)
Lawrence et al. (1972)
Lawrence et al. (1972)
Lawrence et al. (1972)
Lawrence et al. (1972)
Shumskaya and Karamzina
                                                                       reduced urinary  chlorides

-------
                                               TABLE 6-1.   (continued)
Route
Subcutaneous
Subcutaneous
Species
Rat
Rat
Oose
100 Dig/kg
125 ing/kg
Effect
Oliguria
LD5o, anuria, oliguria,
Reference
Pallade et al. (1967)
Pal lade et al. (1967)
Subcutaneous
Subcutaneous
Subcutaneous
Rat
Rat
Rat
150-180 mg/kg
250 mg/kg
500 mg/kg
Intravenous
Intravenous
Intravenous
Intravenous
Cat
Cat
Dog
Oog
9.3 mg/kg
93 mg/kg
9.3 mg/kg
93 mg/kg
  serum protein and sodium
  reduced, serum potassium
  increased

66% mortality, anuria,
  oliguria, carbonic hydrase
  reduced, blood catalase
  reduced, lung and kidney
  changes

Polyuria, proteinuria,
  :urinary.chlorides, blood
  and urine creatinine
  i ncreased

Increased free aromatic
  amines, decreased his-
  taminase activity

Blood pressure decreased

Minimum lethal concentration

Blood pressure decreased

Minimum lethal concentration
Pallade et al.  (1967)
Rotaru and Pallade (1966)
Shumskaya and Karamzina
  (1966)
Shumskaya and Karamzina
  (1966)
                                                                                         Freuder and Leake (1941)

                                                                                         Freuder and Leake (1941)

                                                                                         Freuder and Leake (1941)

                                                                                         Freuder and Leake (1941)

-------
                                                   TABLE 6-1.   (continued)
      Route
Species
Dose
                                                               Effect
                                                                           Reference
0>
I
Dermal
  (single
  application)

Dermal
  (single
  application)

Dermal
  (single
  application)

Dermal
  (tail
  immersion,
  60 min)

Dermal
  (tail
  immersion,
  2 or 3 x,
  20-30 min)

Dermal

Dermal
    Dermal
    Dermal
                        Rat
                        Rat
                        Rat
                        Mouse
                0.5 ml /kg
              Not lethal
                        Mouse
Guinea Pig

Rabbit


Rabbit


Rabbit
                1,180 mg/kg for 1 h   20% dead
                2,360 mg/kg for 1 h   90% dead
                                      70% dead
                                      100% dead
                                        4,420 mg/kg for 1 h   Not lethal

                                        11.8 mg               Mild irritation
                118, 236 mg for 2 h   Lesion size, duration,
                                        intensity less than 0.5 ml
                590 mg for 24 h
              Edema, necrosis
                                                                     Freuder and Leake (1941)
                                             Freuder and Leake (1941)
                                             Freuder and Leake (1941)
                                             Pallade et al. (1967)
                                             Kremneva and Tolgskaya
                                               (1961)
                                             Freuder and Leake (1941)

                                             Smyth et al. (1962)
                                             Weil et al. (1963)

                                             Pallade et al.  (1967)
                                                                                         Pallade et al. (1967)

-------
                                                     TABLE 6-1.   (continued)
        Route
Species
        Dose
                           Effect
      Reference
at
Ch
      Dermal

      Dermal


      Dermal
      Intradermal
      Corneal



      Cornea!


      Corneal


      Corneal


      Corneal


      Corneal
Rabbit

Rabbit


Rabbit
Rabbit
Rabbit



Rabbit


Rabbit


Rabbit


Rabbit


Rabbit
                      No irritation to marked
                        irritation, dose-related
                      Blepharospasm, constriction
755 mg/kg for 24 h    LD56

1,180 mg for 24 h     LD5d
0.2 ml of 0.3125%,    No irritation to marked
  0.625%, 1.25%, 2.5%,  irritation, dose
  5.0% in cottonseed    related
  oil

0.2 ml of 0.002%,
  0.008%, 0.031%,
  0.125%, 0.5%
  in cottonseed oil

1 drop, undiluted
  corneal clouding,
  swelling

1 drop, undiluted
  corneal injury

0.1 ml, 40% in
  cottonseed oil

0.1 ml, 20% in
  cottonseed oil

0.1 ml, 10% in
  cottonseed oil

0.1 ml, 5% in
  cottonseed oil
Lawrence et al. (1972)

Smyth et al. (1962)
Weil et al. (1963)

Lawrence et al. (1972)
Lawrence et al. (1972)
Kremneva and Tolgskaya
  (1961)
                      Grade 4, moderately severe     Smyth et al. (1962)
                      Iritis, palpebral irritation,
                        edema

                      Conjunctiva! and palpebral
                        irritation, edema

                      Dubious irritation
                      No irritation
Lawrence et al. (1972)
Lawrence et al. (1972)
Lawrence et al. (1972)
Lawrence et al. (1972)

-------
     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 LC56 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
dome convulsions.   The animals exposed to  epichlorohydrin  at 2,370 ppm
showed no symptoms of toxicity other  than nose and eye irritation.
                                  6-7

-------
     Kremneva (I960)  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
mg/nr
899-1,199
2,242-3,587
4,036-5,381
5,381-6,278
6,726-8,071
Concentration
ppm
237-316
590-944
1,062-1,416
1,416-1,652
1,770-2,124
No. of
Rats
15
18
10
10
10
Test 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/ms) and the
LC50 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
stveral 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 mucosa 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 LT56 (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 LT5(j  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
mg/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<> 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 LD50 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  LD5& 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 Pal lade (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
proximal 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  LDSo
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  LD5o  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

                                 LD5fl                   95% Confidence Limit
Species                       (mg/Kg)                         (mg/kg)
Mouse                            170                             153-188
Rat                              113                            94.6-134
Guinea pig                       118                            29.5-472
Rabbit                           160                            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--Epichlorohydrin 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
LD5o 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 LDSO 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 1-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  ulcerlike  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 LDSO
value  in  male  New Zealand albino rabbits was 0.64 ml/kg  (755 rag/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-
cMorohydrin (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 studies.   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 coptex 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

-------
                                                         TABLE 6-5.   SUBCHRONIC EFFECTS OF EPICHLOROHYDR1N
Ot
Species Route
Rat Inhalation

Rat Inhalation
Rat Inhalation
Rat Inhalation
Dose
9 ppm
6 h/d X 5 d/wk
18 exposures
17 ppm
6 h/d X 5 d/wk
19 exposures
27 ppm
6 h/d X 5 d/wk
18 exposures
56 ppm
6 h/d X 5 d/wk
18 exposures
Effect
No effects except two pulmonary
infections

Normal weight gain, inferior
condition, no abnormal pathology
Nasal irritation, constant body
weight, no weight gain, no abnormal
pathology except for one consoli-
dated lung
Weight loss, nasal irritation and
discharge, respiratory distress.
No effect on hemoglobin and dif-
ferential cell counts or urinary
protein, no abnormal pathology
Reference
Gage (1959)




                       Rat
Inhalation
                       Rat
Inhalation
120 ppm
6 h/d X 5 d/wk
11 exposures
20-60 mg/ms
3 h/d X 5 d/wk
X 6.5 mo
Weight loss, nasal irritation and
discharge, lethargy, deteriorated
condition, leukocytosis, increased
urinary protein.  Lungs were con-
gested, consolidated, edematus,
with inflammation and abscess for-
mation; kidneys showed atropic
tubules, leukocytic infiltration.
Liver congested.

Weight gain 5-10% less than con-
trol, slightly elevated blood
pressure, pathology same as 170-
250 mg/nr except not as severe.
Kremneva and
Tolgskaya (1961)

-------
                                                                        TABLE  6-5.   (continued)
                         Species     Route
                          Dose
                                   Effect
                                             Reference
                         Rat
Inhalation
en
 i
                         Rat
                         Rat
Inhalation
Inhalation
                         Rat
Inhalation
170-250 ing/m3
3 h/d X 5 d/wk
X 5 mo
0.2 mo/Hi3
continuous
exposure, 98
days

2 ng/n3
continuous
exposure,
98 days
20 Big/B3
continuous
exposure, 98
days
General deterioration in condition,
subcutaneous hemorrhage, respiratory
distress and dyspnea, two deaths
after first month, remaining animals
died at the start of fifth month
of exposure.  Elevated blood pressure
after 1-2 mo exposure.  Bronchitis,
necrosis of bronchial mucosa,
thickened and alveolar septa, pul-
monary edema.  Deterioration of
kidneys, necrosis of convoluted
tubules.  Hepatic cells showed
fatty degeneration and vacuoliza-
tion,  Myocardial tissue stained
irregularly, fragmentation of
myocardial fibers.

No effects observed
Decreased blood nucleic acids,
no effects on erythrocytes,
leukocytes, and hemoglobin
levels.  Increase in fluores-
cent dye-fixing leukocytes.
No abnormal pathology.

Decreased blood nucleic acids
No effects on erythrocytes,
leukocytes, and hemoglobin levels.
Increase in fluorescent dye-fixing
leukocytes.  Emphysema, desquama-
tive interstitial pneumonia, edema
and deterioration of vascular con-
nective tissue of lung.  Intermus-
cular, micro-focal hemorrhages and
venous plethora in heart.  Necrosis
of the convoluted tubules of kidney.
Damaged neurons in medulla oblongata,
cerebellum and hippocampus.
Kremneva and
Tolgskaya (1961)
Fomin (1966)
Foain (1966)

-------
                                                                      TABLE  6-5.   (continued)
                       Species      Route
                       Rat
Inhalation
                       Rat
Inhalation
en
ro
                       Rat
Inhalation
                          Dose
                                   Effect
25 ppn 6 h/d X
5 d/wk 61-62
exposures
50 ppra 6 h/d X
5 d/wk 61-62
exposures
                                             Reference
Rat
Inhalation
5 ppn 6 h/d X No effects
5 d/wk 61-62
exposures
Quast et
al.
1979a
100 ppn 6 h/d X
5 d/wk 9 expo-
sure in 12 days
Decreased activity, local
eye irritation.  Inflam-
matory, degenerative changes
in respiratory and olfactory
epithelium.  One kidney tunor
unrelated to exposure.

Decreased activity, local
eye irritation.  Decreased
weight gain.  Inflammatory
and degenerative changes in
respiratory and olfactory
epitheliun of the nasal
turbinates.  Increased kidney
weights.  Tubular necrosis and
edema of renal cortex.  Decreased
hepatocellular glycogen content.
Increased vacuolization of cyto-
plasm in zona fasciculata of
adrenal cortex.

Nasal discharge, respiratory irri-
tation decreased food intake, de-
creased body weight.   Degeneration,
inflammation, hyperplasia, squamous
metaplasia of respiratory and
olfactory epitheliun.   Other changes
same as 50 ppm exposure, except
more severe.
Quast et al.  1979b

-------
                                                                      TABLE 6-5.   (continued)
Species Route
Mouse Inhalation
Mouse Inhalation

Mouse Inhalation
Mouse Inhalation
Dose
2,370 ppm
1 h/d until
all animals
dead
5 ppm 6 h/d X
5 d/wk 61-62
exposures
25 ppm 6 h/d
5 d/wk 61-62
exposures
50 ppm 6 h/d X
5 d/wk 61-62
exposures
Effect
All animals died after 16 exposures;
half died after 7 exposures.
No effects.

Slightly decreased .weight gain.
Inflammation and degenerative
changes in nasal turbi nates. No
abnormal changes in other organs.
Decreased weight gain. Decreased
food intake. Inflammation and
degeneration changes 1n nasal
Reference
Freuder and Leake
(1941)
Quast et al. 1979a



en

INJ
ro
Mouse
Inhalation
                        Rabbit      Inhalation
                        Rat
            Oral
                        Rat         Oral


                        Rat         Oral
100 ppm 6 h/d X
5 d/wk 9 expo-
sure in 12 days
                    9, 16, 30 ppm
                    6 h/d for 20
                    days

                    94 mg/kg
                                190 mg/kg


                                271 mg/kg
turbanates.  Focal subacute
pneumonitis.  No abnormal changes
in other organs.

Nasal irritation, decreased food
intake, decreased body weight.
Inflammation and degenerative
changes in nasal turbinates.
Decreased hepatocellular glycogen
content.  Decreased hepatocyte size.
Focal subacute pneumonitis.  Slight
atrophy of thyraus.

Nasal irritation, normal body
weight increases, no abnormal
pathology.

Administered daily until all
animals died.  First death after
2 doses.  100X mortality after
21 doses.

First death after first dose.
100X mortality after 8 doses.

First death after first dose.
100% mortality after 4 doses.
Quast et al. 1979b
                                                                                             Gage (1959)
                                                             Freuder and Leake
                                                               (1941)

-------
                                                                     TABLE 6-5.  (continued)
Species Route
Rat Intraperltoneal
Dose
0.00955 and
0.01910 ml /kg
in cottonseed
oil daily for
30 days
Effect
Decreased weight gain, increased,
kidney- to- body weight ratios,
normal hema to logic parameters.
normal sodium sulfobromophthalein
disappearance. Increased
incidence of pulmonary lesions
over controls. Other tissues
normal.
Reference
Lawrence et al.
(1973, 1974)
                       Rat
            Intraperltoneal
ivs
oo
Rat
Rat
                                  Intraperltoneal
                                  Intraperitoneal
0.04774 ml/kg
in cottonseed
oil 2 d/wk
X 12 wk
0.0190 ml/kg
In cottonseed
oil 3 d/wk X
12 wk
0.0095 ml/kg
in cottonseed
oil 3 d/wk X
12 wk
Decreased food consumption, de-
creased body weight gain, decreased
hemoglobin values and erythrocyte
counts.  Increased segmented neu-
trophils and decreased lymphocytes.
Increased heart, kidney and liver-
to-body weight ratio.  No abnormal
tissue pathology reported.

Decreased food consumption, de-
crease hemoglobin values and
erythrocyte counts.   Decreased
lymphocytes.  No abnormal tissue
pathology reported.
Decreased hemoglobin values.
other findings.
                                                                                  No

-------
0.17  to  0.25 mg/1  (170 to 250 mg/ms).   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  rag/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/nr1 (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                          No. of Survivors
1-2
3-5
6
7
8
9-15
16
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 conjunctiva! 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 mice  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-Oawley 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 (SGOT),
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 cytoplasm*c 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 arable.   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
ofgins 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 LT50
(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 Tolgskaya 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/m3)
                                  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  Pal lade (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/ni3 (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  th^ 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 mg/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  conjunctiva! and palpebral
irritation with  edema;  a  40 percent  concentration  produced iritis and
palpebral irritation with edema; and an 80 percent epichlorohydrin solution
produced corneal injury.
                                  6-37

-------
     Smyth and Carpenter  (1948) described corneal  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 corneal injury.
     Kremnera and Tolgskaya (1961) instilled a single drop of epichlorohydrin
into the conjunctival sac of a rabbit's  eye.  The  epichlorohydrin produced
blepharospasm, hypermia  of the mucosa,  excessive  lacrimation,  papillary
constriction,  and corneal  clouding.   The corneal 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
CNS depression and death  resulting from respiratory paralysis.   The LC50 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  LC50  values.   These  LC5o
values were similar when  rats and mice were analogously exposed.  Intraper-
itoneal LD50s were  found at 187 mg/kg for rats  and 165 mg/kg for mice; and
oral LD50s 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  LDSO 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.   Chronic irritation of the nasal passages can result in neoplastic
lesions.   Subchronic  exposure  to  epichlorohydrin  has been shown to cause
severe renal  toxicity in  rats via different routes  of administration.
Necrosis of the convoluted  tubules  was  found,  which  was  reversible after
exposure ceased.
                                  6-38

-------
     By both the  Inhalation  and subcutaneous routes, epichlorohydrin has
been shown to cause changes  in  the lungs and bronchi.  Subchronic exposure
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 TERATOGENIC EFFECTS

7.1  CARCINOGENICITY
7.1.1  Introduction
       "^      _mmfffffm^^n^                              ^
     The purpose of  this  section is to provide  an evaluation of the likeli-
hood 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  heavily  on animal  bioassays and
epidemiologic evidence.  However,  information on mutagenicity and metabolism
reviewed in  other sections  of  this document, particularly in  relation to
chemical interaction with  DNA and pharmacokinetic behavior, have an important
bearing  on  qualitative  and quantitative  assessments of  carcinogenicity.
This section  presents  an  evaluation of  the  animal   bioassays,  the  human
epidemiologic evidence, the  quantitative aspects of assessment, and finally,
a summary and  conclusions  dealing with relevant aspects of the  carcinogeni-
city of epichlorohydrin.
7.1.2  Animal Studies
     In separate  studies,  epichlorohydrin produced carcinogenic responses at
the site  of exposure:   nasal cancer in an inhalation experiment, sarcomas
at the  site  of subcutaneous injection, and papillomas and carcinomas  in  the
forestomach in a drinking water study.
7.1.2.1   Inhalation Exposure:   Rat--Rats exposed  to epichlorohydrin  vapor
(Aldrich Chemical  Company, ^99  percent pure  by gas chromatography)  have
shown a  statistically  significant increase (P<0.05)  in  nasal cancer (Laskin
et al.  1980).  Exposures were done in 128-liter or 1.3-m3  inhalation chambers.
Ambient  epichlorohydrin  levels  were monitored spectrophotometrically  during
exposure.
     Test animals  were  noninbred male Sprague-Dawley rats initially 8 weeks
old.  Body weights were  recorded monthly.  Rats were allowed to live until
natural  death  or were killed jn  extremis.   Necropsies were  performed, and
tumors  and  lesions  and major  organs were  examined histopathologically.
Heads were fixed,  decalcified,  and sectioned  for examination of the entire
nasal cavity.
                                    7-1

-------
     Based  on  the results of  preliminary  LCrQ studies, 100 ppm  epichloro-
hydrin  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.  Results
of  these  studies  were  reported  for  the two  groups  combined, except for
nasal cavity carcinomas which were reported separately.
    Early mortality, attributed  to respiratory disease, was  higher in sham
(exposed  to air only)  and  untreated  control  groups compared to treated
animals (Figure 7-1).  Body weight gain was  similar  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.    Severe  inflammation in the  nasal  cavity was
noted in  90 percent of the control animals.   Edema,  congestion,  and pneumonia
were observed  in  the  lungs  of exposed rats.   Renal  damage,  including dila-
tation  of cortical  and medullary  tubules,  was found in 63  percent of the
rats.  Control   rats had  congestion,  edema, bronchiectasis,  and  pneumonia in
the lungs as well  as kidney changes commonly found in aging rats.
     Squamous  cell  carcinomas were  found  in  the  nasal tract of treated
animals as  described  in  Tables  7-1  and 7-2.  Many  of these carcinomas
infiltrated the bones  of  the skull;   however,  metastasis of  these tumors  was
not  found.   Additionally,  three  other rats  were  diagnosed with nasal  or
bronchial  papillomas.   Squamous  metaplasia was evident in 10  percent  of the
treated rats.    The  incidence of  other tumor  types  in nonrespiratory organs
was similar between treated and control groups.

Table 7-1.  Squamous Cell Carcinomas of the Nasal Cavity Following Thirty
         6-Hour Exposures to 100 ppm Epichlorohydrin (Laskin et al. 1980)
Experiment*
1
2
No. of
Tumor- bearing
No. of Animals
Animals (Percent)
40 4 (10)
100 11 (11)
Tumor Observation Time
(davs)
Mean Range
540 462-610
623 330-933
*Both experiments represent animals exposed to 100 ppm epichlorohydrin.
                                    7-2

-------
t
          I    I   I    I   I    I   I    I   I    I   I
                     I   I    I   1    I   I    I   I    I   I    I   I    I   I    I
             16      32     48     64     80     96

                         WEEKS AFTER FIRST EXPOSURE
112
128
144
     Figure 7-1. Mortality of rats following exposure to 100 ppm of epichloro-
     hydrin (6 hr/day for 30 days). The curves represent air-treated controls
     (  • ), untreated controls (  A ), and a group of 140 animals exposed to
     epichlorohydrin ( •  ).

     Source:  Laskin et al. (1980).
                                  7-3

-------
       TABLE 7-2.  DOSE-RESPONSE FOR INDUCTION OF SQUAMOUS CELL CARCINOMAS
                 IN THE NASAL CAVITY OF MALE WISTAR RATS EXPOSED
                            TO EPICHLOROHYDRIN VAPOR
                       (adapted from Laskin et al., 1980)
Concentration
ppm
100 (combined studies)
30
10
Air (sham) control
Number of
Exposures
30
290*
250*
—
Dose
(ppm days)
3,000
8,700
2,500
—
(No. with Cancer)
(No. Exposed)
15/140t
1/100
0/100
0/100
   for life
   Untreated
   control
0/50
   tP < 0.00001 vs.  combined controls.
   *Lifetime exposures were based on median survival time.

     A second  study was  done  in which 100  rats  per treatment group were
exposed to 10  or  30 ppm epichlorohydrin 6  hours  per day,  5 days per week,
for their lifetimes.   Treated  animals  were compared to  concurrent  sham and
untreated control  groups.
     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.
     Respiratory tract tumors were not found with  the  exception  of a nasal
papilloma and  a  squamous  cell  carcinoma  in  two rats  exposed  to 30 ppm.
Severe inflammation  in the  nasal  cavity was noted in  90  percent  of the
control animals.   Exposure  to  30 and 10 ppm  epichlorohydrin produced 4 and
2 percent  incidences, respectively, of squamous  cells  metaplasia in  the
nasal cavity.  Renal  damage occurred in 65,  37,  24, and 17 percent of the
30 ppm,  sham,  and  untreated groups, respectively.  Severity of renal damage,
diagnosed as mainly tubular degenerative changes, was related to dose.
                                    7-4

-------
   100 —
£80-
O

III

I
3

3
O
I-
iu
U
e
LU
0.
         I   i    I   I   I   I   1   I   I   I   I   I   i   i   i   i   i   I
                           i   i   i   i   i   i   I   I   i   i   i   i
20 —
                           48      64      80     96     112

                           WEEKS AFTER FIRST EXPOSURE
       Figure 7-2. Mortality of rats following lifetime exposure (6 hr/day, 5
       days/wk) to epichlorohydrin. The curves represent air-treated controls
       (  a ), untreated controls ( A ), exposure to 30 ppm (  O ), and exposure
       to 10 ppm ( • ).

       Source: Laskin et al. (1980).
    220
              I    I   I   I    I   I   I   I   I   i   I   I    I   I
    100
                           48     64     80     96     112

                           WEEKS AFTER FIRST EXPOSURE
                                                        128
144
       Rgure 7-3. Growth of rats following chronic exposure to epichloro-
       hydrin. The curves represent air-treated controls ( D ), untreated
       controls (  A ). and animals exposed to 30 ppm ( O ) or 10 ppm ( • )
       epichlorohydrin. Exposures were done for 6 hr/day. 5 days/wk for the
       lifetimes of the animals.

       Source: Laskin et al. (1980).
                                   7-5

-------
     Results of the  study by Laskin et al.  (1980)  provide evidence for the
carcinogem"city of epichlorohydrin.   Epichlorohydrin produced a significant
increase  in  nasal  cavity  carcinomas  with a high dose  given early in the
study, but did  not do so when given  as  a lifetime  treatment with one-third
and  one-tenth  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  would  have been  shorter  with  the more intense,
though shorter, exposure  and that  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 whether nasal
tumors would have  been observed in  the absence of nasal  inflammation.   It
is the authors' impression  that inflammation in control  rats used in their
laboratory for  lifetime  inhalation carcinogem'city  studies  is  not apparent
before 1  year  (personal  communication with  R.E.  Albert); therefore,  since
exposure  to  the carcinogenic  100  ppm  exposure  level  of epichlorohydrin
occurred during 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  considered  low.   There is presently no  evidence avail-
able to suggest a  promoting action of  nasal  inflammation on the  induction
of nasal  carcinomas  by exposure to  epichlorohydrin  (personal communication
with R.E.  Albert).
7.1.2.2   Oral  Administration:   Rat—Kom'shi   et  al.  (1980)   and  Kawabata
(1981) described a carcinogem"city bioassay  on epichlorohydrin given  orally
to rats.  The  epichlorohydrin  (Ham*  Kagaku,  Kyoto)  was  99.96 percent pure;
impurities,  if  known,  were  not  reported.   Seventy-two male  outbred Wistar
rats, 6 weeks  old  and weighing 160  g,  were  divided into  four groups  of 18
rats each.   Six animals  were housed In each cage.   Animals  were given fresh
solutions of epichlorohydrin in drinking water each day.   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 epi-
chlorohydrin.  Water  intake was  estimated daily, and  rats were weighed once
                                    7-6

-------
each week.   Survivors were fasted for 17 hours before sacrifice after 81 weeks
of treatment.  All animals were  necropsied, 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.
     Although  survival among  all groups was  similar (Table  7-3),  treatment
with epichlorohydrin was  discontinued  for  short periods after  60 weeks  due to
debilitation of the  rats (Figure 7-4).  The  cause of death in animals that
died was concluded to be pulmonary infection.

        TABLE 7-3.  KIDNEY WEIGHTS AND KIDNEY/BODY WEIGHT RATIOS IN MALE
       WISTAR RATS GIVEN EPICHLOROHYDRIN IN DRINKING WATER FOR 81 WEEKS
                       (adapted  from Kawabata 1981)
Dose
(ppm)


0

375

750

1500


Number

Initial
18

18

18

18


of rats

Effectivet
10

9

10

12

Body
weight*
(fl ± S.D.)

Initial
157±10

159± 6

157± 8

160± 7


Final
595175

494+45

415±46§

295±46§

Organ weights (g ± S.D.)*
(% of body weight)
Kidney
Left Right
1.710.2 1.6 10.2
(0.3110.07) (0.3110.07)
2.2l0.6§ 2.1 10. 4§
(0.44i0.11)§ (0.4310.08)§
2.1i0.2§ 2.210.2§
(0.52i0.07)§ (0.5310.05)§
1.9l0.2§ 1.910. 2§
(0.66i0.09)§ (0.6510.09)§
*S.D. = standard deviation.
 Based on rats sacrificed at 81 weeks.
§P < 0.05.
     Patterns  of epichlorohydrin  intake  during the study are  presented  in
Figure  7-5.   Total epichlorohydrin  consumption per  rat during  the  total
experimental period was  estimated as 0, 5.0, 8.9, and 15.1 g in  the control,
375,  750,  and 1,500 ppm  groups,  respectively.   Food consumption data were
not  reported.   Water  intake  was  stated  to be  similar  among all groups,
without presentation of data.
                                    7-7

-------
                            EXPERIMENTAL PERIOD, weeks
       GROUP 0    20              40               60                81
         NO.  \fi  I	I	I	I
         1
         2    r                                  inr
                                                     vmmsmm

                    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).
     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,  due to comparable organ weights among control and  treatment groups
and decreased body  weights  in treated animals,  were common in the groups
given 750 or 5,500 ppm epichlorohydrin.  A significant  (P  < 0.05)  increase
in pancreas/ body  weight  ratios was  also  evident in the  375 ppm  group.
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  observed in rats exposed to  epichlorohydrin  in
the careinogenicity  study by Laskin et al.  (1980).

                                   7-8

-------
    100
        GROUP 1
        GROUP 2
        GROUP 3
        GROUP 4
                                 _L
I
(9
a

Q
s
               10        20        30       40       SO

                                 EXPERIMENTAL PERIOD, weeks
         60
70
80
       Figure 7-5. Intake of epichlorohydrin in drinking water by male Wistar rats. The Ones
       represent the 375 ppm ( •  ). 750 ppm ( A ). and 1500 ppm I • ) dose groups.

       Source: Kawabata (1981).
                                                            GROUP 1
                                                            GROUP 2
                                                            GROUP 3
                                                            GROUP 4
                                 30       40       SO

                                 EXPERIMENTAL PERIOD, wsoks
                          80
     Figure 7-6. Effect of epichlorohydrin treatment of body-weight in male Wistar rats. The
     curves represent untreated controls ( o I and dose groups given 375 ppm (  • }, 750 ppm
     ( A ), and 1500 ppm ( •  )

     Source: Kawabata (1981).
                                        7-9

-------
     Results of  blood  analyses were normal except for significant (P < 0.05)
increases in cholesterol and neutral lipid levels in each treatment group com-
pared to  controls.   The relationship between  these blood  analysis  results and
epichlorohydrin  treatment  is  presently not understood, and a stronger exami-
nation 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.
These tumors  were histologically diagnosed  as hyperplasia, papilloma, and
squamous  cell  carcinoma.   Both localized  and  diffuse  hyperplasia were
observed.    Proliferation  of squamous epithelium  and  multistage stratifica-
tion of  basal  cells were  discussed in hyperplastic  regions.   Papillomas
consisted of squamous epithelium projecting into  the lumen.  Marked keratini-
zation with little  nuclear division was apparent in the  papillomas.  Carci-
nomas were  characterized  as highly differentiated,  keratinized  squamous
epithelium  which 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 in two rats
given 1,500 ppm  epichlorohydrin and interstitial cell tumors  of the  testes
in two or three rats in each group.
     Results of  the study  by  Konishi   et  al.  (1980)  and Kawabata (1981)
provide evidence for carcinogenic  activity  by epichlorohydrin  in drinking
water in  the  forestomach  of male Wistar  rats. A stronger indication of
carcinogen!city  could have been possibly obtained if  the study  protocol had
included  larger  numbers of animals and lifetime  treatment and observation.
Furthermore,  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 con-
sidered unreliable by the  authors.   The dose levels 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.   Nonetheless, induction
of forestomach neoplasia by a direct action  of epichlorohydrin  is supported

-------
                       TABLE 7-4.   NUMBER OF FORESTOMACH TUMORS AND TUMOR INCIDENCE IN MALE WISTAR RATS
                                     GIVEN EPICHLOROHYDRIN IN DRINKING WATER FOR 81 WEEKS

Number
of Rats
Number
of Tumors in Forestomach Per Rat
Histologlcal Findings on
Squamous Epithelium (X)
Tumor size (mm)
Dose
(PP-)
0
375
750
1500

Initial
18
18
18
18

Effectivet
10
9
10
12

Total
0
5.6 ± 8.4
9.9 ± 12.8
32.8 t 24.0

>5
0
0
0.4 i 0.5
0.8 t 1.0

2-5
0
0.1 t 0.3
1.2 t 1.2
4.4 t 3.6

<2
0
5.4 ± 8.2
8.3 1 12.0
27.6 t 21.3

Hyperplasia
0 (0)
7 (77.8)
9 (90.0)
12 (100.0)

Papilloma
0 (0)
0 (0)
1 (10.0)
7 (58.3)§

Carcinoma
0 (0)
0 (0)
1 (10.0)
2 (16.7)
tEffectlve groups Include teralnally sacrificed animals only.

iP = 0.005 vs.  control.   For papillomas and carcinomas combined P < 0.001 by one-tailed Fisher's Exact Test using terminal
 sacrifice only.

Source:  Konishi  et al.  1980.

-------
by  other  studies  discussed herein that indicate  a  direct  tumorigenic  action
of  epichlorohydrin  at other sites and by  the  chemical  nature of epichloro-
hydHn as an alkylating agent.
     Epichlorohydrin's  solubility in water  is 6.48 percent  at 20°C.   The
authors  stated the  epichlorohydrin  was dissolved  in the  drinking water
solution  given to the  animals,  and comparison  of epichlorohydrin  intake
(Figure 7-5)  with the dose-related decrease in  body  weight in each treat-
ment group  (Figure  7-6) is evidence that  the  nominal  doses were achieved.
7.1.2.3   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 chromatog-
raphy; the  epichlorohydrin sample was 99.8 percent pure  (personal  communi-
cation 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.   The study lasted  for  580 days, and the
median survival time was 506 days.  No skin tumors were observed.
     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 pre-
viously.   The  mice  used  by Weil et al. (1963) were initially 90 days old,
which did not  allow an  evaluation of cardnogenicity  during  early  growth of
the animals.   A "brushful"  does not give  any  indication  of the actual dose
applied.
7.1.2.4   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
applications of 2.5  pg  phorbol myristate acetate  in  0.1  ml acetone for the
                                    7-12

-------
duration of the experiment  (median  survial  >  385 days).   Nine 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.
     7.1.2.5    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 adenocar-
cinoma  (P SO.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  produced local  sarcomas  at the site  of  subcutaneous
injection but did not  produce  distant tumors after  intraperitoneal  injec-
tions.
     Kotin and Falk  (1963)  administered single subcutaneous  injections of 5
MM (462 Mi)  °f epichlorohydrin in  0.1  ml ethyl  laurate  or tricaprylin to 30
C3H-strain  mice,  which were  observed along  with solvent-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 carcino-
genicity testing  compared to  a stronger challenge of  repeated  treatment
over a  lifetime at doses as high as those maximally  tolerated.
                                    7-13

-------
7.1.3  Epidemlo1ogic Studies
     A  retrospective  cohort mortality  study of epichlorohydrin workers  is
being  conducted for Shell  Oil  Company  by  Dr.  Phillip  Enterline  of  the
University of Pittsburgh  (Enterline  1978, 1981).  The  cohort  of 864  comprised
workers from  the  Shell  plants at Norco, Louisiana,  and  at  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  a  least one quarter  before January  1,  1966.  The  EPA
Carcinogen Assessment  Group (CAG) previously reported  on  the 1977 update
(Carcinogen Assessment  Group 1980).  Those data (see Table  7-5 plus  footnote)
showed less observed mortality  than  expected, 54 versus 97.3, respectively,
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  of  leukemia deaths,  11 occurred in workers 15 years or
more after first  exposure  (Table  7-5).   Even though these increases were not
statistically significant,  the  trend  provided  reason  for concern that
increasing observation  time  would produce  more of these cancers,  leading to
positive  conclusions  about  the  human careinogenicity of  epichlorohydrin.
     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 7-5 and  7-6.   In this latest 2-year  followup period,  1978-79, 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
produced decreases in  both  the overall   SMRs  and especially in those for  the
group with greater than 15 years  since  first exposure.   None  of these  SMRs is
statistically significant.
     In this most  recent  update,  Enterline  (1981) also  presented  a smoking
history of 12 of the cancer deaths.  He found that for the 10  lung cancer deaths
as diagnosed on death certificates,  7 individuals were known  smokers,  1 was a
nonsmoker, and  2  had  unknown  smoking  histories.   This confounding factor
makes a positive  causal relationship between epichlorohydrin and  human lung
cancer even more difficult.

                                    7-ld

-------
   TABLE 7-5.   COMPARISON OF MORTALITY  IN ENTERLINE'S  EPICHLOROHYDRIN  STUDY
                          UPDATES BY CAUSE AND  BY LATENCY
                                 (1978  versus 1981)*
         Time Since
      First Exposure
All Cases

Observed/
Expected   SMR
Respiratory
   Cancer

Observed/
Expected   SMR
Leukemia

Observed/
Expected   SMR
Enter! i ne
(1978)


(1981)


Overal 1
< 15 years
> 15 years
Overall
< 15 years
> 15 years
54/97.
19/45.
35/51.
65/115
19/46.
46/69.
3
8
6
.7
0
8
55.
41.
67.
56.
41.
65.
5
5
9
2
3
9
10/6.
1/2.
9/4.
10/8.
1/2.
9/6.
8
2
7
7
2
5
146.
45.
193.
114.
45.
137.
2 2/0.
9 0/0.
1 2/0.
1 2/1.
0 0/0.
8 2/0.
9
5
4
0
5
5
224.7
0
500.0
194.2
0
377.4
    *Inc1uded 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 number
above referring to the 1978 report include this correction.
                                    7-15

-------
                TABLE 7-6.   OBSERVED AND EXPECTED DEATHS AND SMRS AMONG 863 MALES EXPOSED FOR MORE
               THAN THREE MONTHS IN  THE  MANUFACTURE OF EPICHLOROHYORIN, BY TIME SINCE FIRST EXPOSURE
                                  NORCO, LOUISANA AND DEER PARK,  TEXAS 1948-1979
                                                 (Enterline 1981)*
Causes of Death

All Causes
All Cancers
(140-205)
Respiratory
Cancers (160-164)
Leukemias (204)
All Other Cancers
All Other Causes
Total No. of Men
Man-years
Total
OBS EXP
65 115.
16 21.
10 8.
2 1.
4 11.
49 93.
863
19,909.
Time Since First Exposure
> 15 years < 15 years
SMR
72 56.2
73 73.6
74 114.4
03 194.2
96 33.4
99 52.1

9
OBS EXP SMR
46 69.75 65.9
15 14.94 100.4
9 6.53 137.8
2 0.53 337.4
4 7.88 50.7
31 54.81 56.6
824
7406.8
9JS EXP
19 45.97
1 6.79
1 2.22
0 0.49
0 4.08
18 39. 18
863
12,503.1
SMR
41.3
14.7
45.0
0
0
45.9


     *A reexamination of both the cohort and the death certificates by Enterline has
the nuabers of the previous update.   Specifically, even though only nine lung cancer
in the previous report,  a reexamination showed that actually ten had occurred.   None
update 1978-1979.
led to slight changes in
deaths were presented
occurred during the

-------
     Additionally,  Enterline  considered  the  severity  of epichlorohydrin
exposure.   Regarding the  group  with at least 15 years since first exposure,
he  stratifies  "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  by two  separate  studies by  Enterline that  share  some of the
same cohort  (Table 7-7).   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  with those 350 men
from Deer Park who were known  not  to have  worked in the  unit.*  The results
show that the  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  epichlorohydrin alone  (SMR  = 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 7-7 shows  the mortality 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 in  respiratory cancer deaths.  Also,  the epichlorohydrin
group alone  had  approximately the  same  respiratory/cancer  mortality as the
IPA group alone.
     The conclusion  made by  CAG is  that the   most  recent  update of  the
Enterline data  has provided  less  clear  evidence on the  human  carcinogeni-
city of  epichlorohydrin.   The  evidence for carcinogenicity of epichloro-
hydrin  includes  increased respiratory cancer with  increasing  latent period
and  the  higher respiratory  SMR in cancer  in   the combined IPA plus epi-
chlorohydrin group versus  the IPA group  alone.  Further,  there is elevated
respiratory  cancer in  both  epichlorohydrin production  plants  (Carcinogen
*The exposures to IPA and epichlorohydrin were considered by CAG in its
 most recent report (1980) and concluded that the confounding effects
 between the exposures detracted from the significance of the findings.
                                    7-17

-------
     TABLE 7-7.  COMPARISON OF MORTALITY IN EPICHLOROHYDRIN (ECH) ALONE,
           AND COMBINED WITH ISOPROPYL ALCOHOL (IPA) EXPOSURE GROUPS
                              IN DEER PARK, TEXAS
                            (Enterline 1980 and 1981)
Exposure Group  Number
        All Cases
        Observed/
        Expected   SMR
 All Cancer
Observed/
Expected   SMR
Respiratory Cancer
Observed/
Expected   SMR
Enterline

(1981)
ECH (alone)

IPA and ECH*
308     38/60.89   62.4  11/11.79   93.3   3/4.61   65.1

124     16/23.24   68.9    5/4.68  107.0   4/1.86  214.8
(1980)

IPA (alone)       350

IPA and ECH*      125
        24/44.30   54.2

        16/22.64   70.7
 5/8.21   60.9   2/3.16   63.3

 5/4.37  114.4   4/1.71  233.9
*These represent the same cohort (except for one unidentified man) with 1
 year's additional  values.
                                    7-18

-------
Assessment Group 1980).   Contrary to this evidence,  there are no statistically
significant increases and actually a decreased SMR in the latest 2-year update
compared to the earlier  update.  Also, the increase  in respiratory cancer SMR
in the combined IPA  plus epichlorohydrin exposure group, compared with either
the epichlorohydrin  group  alone or the  IPA group  alone,  suggests that the
interaction between IPA and epichlorohydrin exposure leads to increased respir-
atory cancer.   Significantly in the Deer Park, Texas  epichlorohydrin alone
subgroup of 350,  there  is  no increase in respiratory cancer versus controls.
Considering, in addition,  the confounding factor of smoking,  CAG  is of the
opinion that these studies provide only limited evidence on the human carcino-
gem'city of epichlorohydrin.
     Shellenberger et al.  (1979) conducted a retrospective  cohort mortality
study of  533 white  male full-time Dow Chemical Company  employees  who had
potential epichlorohydrin  exposure  in a production area for at least 1 month
between October 1957 (the  date that 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 carcinogenic!ty  assessment.   First,  only  2 percent
(12/553) of  the cohort  died during the  11-year followup  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
followup period was  only 39 years, and  61.8  percent of the cohort was less
than age  40 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 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  since  first  exposure to the  end of  the study.  This  epidemiologic
                                    7-19

-------
study on  epichlorohydrin, while  negative,  is inadequate for the evaluation of
carcinogenicity  due  to low exposure, short exposure  duration,  short latent
period, and very young age of the cohort.
7.1.4  Quantitative Estimation
This  section  deals with the  unit  risk for epichlorohydrin in  air  and the
potency of  epichlorohydrin relative to  other  carcinogens  that the CAG has
evaluated.  The  unit  risk  estimate for an air  pollutant  is  defined as the
lifetime cancer risk occurring in a hypothetical population in which all indi-
viduals are exposed  continuously from birth throughout their  lifetimes to a
concentration of 1 ug/m  of the  agent in the air they breathe.  This calcula-
tion 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,  which might be associated  with air or water exposure
to these agents,  if the actual exposures are 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)  lifetime
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 the extrapolation model.
     There is no solid scientific basis for any mathematical  extrapolation
model that relates carcinogen exposure to cancer risks at the  extremely low
concentrations that must be dealt with in evaluating environmental  hazards.
For  practical reasons  such  low  levels of  risk cannot be measured  directly
either by animal experiments  or by epidemiologic studies.  We must, there-
fore, depend  on  our  current  understanding of  the mechanisms of carcino-
genesis for  guidance as to which  risk model  to use.  At  the  present  time
the  dominant  view of  the  carcinogenic  process involves the concept  that
most 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
quanta! 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
                                    7-20

-------
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  non-threshold dose-response  relation-
ship  is  also consistent  with  the relatively few  epidemiologic  studies  of
cancer responses to  specific  agents  that contain  enough  information to make
the  evaluation possible  (e.g.,  radiation-induced leukemia,  breast and
thyroid cancer,  skin cancer  induced  by  arsenic  in drinking water, liver
cancer induced by  aflatoxin  in the diet). There  is also  some evidence from
animal experiments that  is consistent  with  the linear non-threshold model
(e.g., liver tumors induced in mice  by 2-acetylaminofluorene in the large
scale EDg,  study  at the National  Center  for  Toxicological  Research and the
initiation  stage  of the  two-stage carcinogenesis  model  in rat  liver and
mouse skin).
     Because  it  has the  best, albeit limited,  scientific  basis  of any of
the  current mathematical  extrapolation  models,  the  linear  non-threshold
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  non-threshold
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.

7.1.4.2  Description of the Low-Dose Extrapolation  Model
      Let  P(d)  represent the lifetime risk  (probability)  of cancer  at  dose
d. The multistage model has the form

      P(d) =  1 - exp [-(qQ + Qjd + q2d2 +  ... + 
-------
     q. > 0, i = 0, 1, 2, ....

Eqirlvalently,

     A(d) = 1 - exp [(qid + q2

where

                  - P(o?
            1 - P(o)

is the extra risk over background rate at dose d.
     The point  estimate  of the coefficients q. ,  i  = 0, 1, 2,  ...,  k,  and
consequently the  extra risk function,  Pt(d)  at any given dose  d,  is cal-
culated 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 GLOBAL 79
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,  q?, on parameter q,.  Whenever  q, > 0, at low doses  the
extra risk Pt(d)  has  approximately the form Pt(d) = q, 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  LQ be  the  maximum value of  the  log- likelihood function.   The  upper-
limit,  q, , is  calculated  by increasing  q, 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  L, satisfies
the equation

                              2 (LQ - Lx) = 2.70554

where 2.70554  is  the  cumulative 90 percent point  of the chi-square distri-
bution with one degree of freedom, which  corresponds  to a 95 percent upper-
limit  (one-sided).  This  approach  of  computing the upper  confidence limit
for  the  extra  risk,  Pt(d)  is an  improvement of the  Crump  et al.  (1977)
model.   The upper confidence limit for the extra risk calculated at low

                                    7-22

-------
doses is  always  linear.  This  is conceptually consistent with  the  linear
non-threshold concept  discussed earlier.  The  slope,  q£,  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 poly-
nomial 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 refitted 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

                            h
                                (X  - NP>
                                N.P. (1-P.)
                                 11     i'
                           i =
is calculated where  N.  is the number  of  animals in the i   dose group, X.
                                   th                                      ^
is the  number of animals in the  i   dose group  with a  tumor response, P. is
                                        th                              ^
the probability  of  a response  in the i   dose  group  estimated by fitting
the multistage model to the data, and h  is  the number of remaining groups.
                                                     2
The fit is determined  to be unacceptable whenever X   is larger than the
cumulative  99%   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  coefficients.
7.1.4.3  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
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 where  the  incidence  is  statistically significantly higher
than the control for at least one test dose  level, or where the tumor
                                    7-23

-------
incidence rate shows  a  statistically significant trend with respect to dose
level, or both.   The  data set which gives the highest estimate of the life-
time carcinogenic risk,  q?,  is selected in most cases.   However,  efforts are
made  to  exclude  data sets  which 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 which  has  larger sample size is selected  for calculating
the carcinogenic  potency*.
     2.  If  there are two or  more data sets of comparable  size  which  are
identical  with respect  to species,  strain, sex,  and tumor  sites, the geo-
metric mean of q?,  estimated from each of these data sets,  is used for risk
assessment.   The  geometric  mean of  numbers  A,,  A0, ..., Am is defined  as
                                              L    c.         m
                            (A-, x A0  x  ...  x A_)
                              L    £.           m
     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.
7.1.4.4  Calculation of  Human Equivalent Dosages from Animal Data
      Following the suggestion of Mantel  and Schneiderman  (1975), we assume
that mg/surface area/day  is  an equivalent dose between  species.   Since,  to
a close approximation,  the  surface  area is proportional  to  the 2/3rds power
of  the weight  as  would  be  the case  for  a  perfect  sphere,  the
exposure in mg/day  per  2/3rds  power of the weight  is also considered to be
equivalent exposure.   In an  animal experiment this equivalent  dose is
computed in  the following manner.

                                      1  x m
Let
        L  = duration of experiment
        1  = duration of exposure
        m  = average dose per day in mg during administration of the agent
             (i.e.,  during 1 ), and
        W  = average weight of the experimental animal


                                    7-24

-------
Then, the lifetime average exposure is
     Often exposures are  not  given in units of ing/day and it becomes neces-
sary to  convert the given  exposures into  mg/day.   For  example,  in most
feeding studies exposure  is  in terms of ppm in  the diet.  In this case the
exposure in mg/day is

                                m = ppm x F x r

where ppm  is  parts per million of the carcinogenic agent, F is the weight of
the  food 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 2/3rds  power
of the weight, so that

                                      ..2/3
                            m « ppm x w    x r
   or
     As  a result, ppm  in the  diet is often assumed  to  be an equivalent
exposure  between  species.   However,  this  is not  justified because  the
calories/kg of  food  is very different in the diet of man compared to  labora-
tory animals primarily due to moisture content differences.  Instead an empiri-
cally-derived food factor, f = F/W, is used, which is the fraction of  a species
body weight that is  consumed per day as  food.  The  following rates were  used:

                                                   Fraction  of Body
                                                   Weight Consumed as
Species
Man
Rats
Mice
W
70
0.35
0.03
ffood
0.028
0.05
0.13
water
0.029
0.078
0.17
Thus  when  the exposure is given as  a certain dietary concentration in ppm,
the exposure in mg/w    is
                                    7-25

-------
             m         ppm x F   ppm x f x W             , ,~
                                            = PP»" f * «
7.1.4.5  Inhalation—When exposure is given in terms of mg/kg/day = m/Wr = s,
the conversion is simply
     When exposure  is  via inhalation,  the calculation of  dose can be con-
sidered 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  inspired and  2) where the carcinogen is a  poorly water-soluble
gas which reaches  an equilibrium between the air breathed and the body compart-
ments.  After  equilibrium is  reached,  the rate of absorption of these agents
is expected to be  proportional to the metabolic rate, which in turn is propor-
tional 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  S0~,  can reasonably be  expected to be
absorbed proportional  to  the  breathing rate.   In this case  the exposure in
mg/day may be expressed as

                                 m = I  x v x r

                                       3          3
where I =  inhalation  rate per day in m , v = mg/m   of the agent in  air, and
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
m /day as follows:
                                    7-26

-------
                  For mice, I = 0.0345 (W/0.025) 2/3 m3/day
                                               O/O  -1
                  For rats, I = 0.105 (W/0.113) '  m /day

For humans,  the value  of  20 m /day*  is adopted as  a  standard breathing
rate  by  the  International  Commission  on  Radiological  Protection (ICRP
1977).
     The equivalent  exposure in  mg/w    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 previous stated relationships are tabulated as follows:
Species
Man
Rats
Mice
W
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/w    is
                             T       -ui
                       m     Ivr     iW   vr     , ,-
                 d =  ni- =  o/o  = —5-T5— = iW   vr
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 species.
     Case 2—The  dose  in mg/day of partially  soluble vapors is proportional
                                                            2/3
to the  02 consumption,  which  in  turn is proportional to W    and is also
proportional to the solubility of the gas  in body  fluids, which can be
expressed as an absorption coefficient, r, for the gas.  Therefore, expressing
   *From  "Recommendation  of  the  International   Commission  on  Radiological
    Protection,"  page  9.  The  average breathing  rate is-107 cm3 per 8-hour
    workday and 2 x 107 cm3 in 24 hours.
                                    7-27

-------
consumption as 0« =  k W   ,  where k is a  constant independent of species,
it follows that

                                     .  .2/3
                                   =  k  w
                                     .  .
                                 m = k w    xvxr
   or
                                      m
                                 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
                             3
concentration in ppm or  ug/m  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  pharmaco kinetic
evidence to  the contrary,  that absorption is  equal  by  either exposure
route,
7.1.4.6  Calculation of the Unit Risk from Animal  Studies—The  risk associa-
                2/3
ted with d mg/kg   /day  is  obtained from GLOBAL  79 and, for most cases of inter-
est to risk assessment,  can be adequately approximated by P(d) = 1  - exp(-q?d).
A "unit risk" in  units  X is simply the risk corresponding to  an exposure of
                                                    2/3
X = 1.  To estimate  this value the number  of mg/kg   /day corresponding to
one unit of X is determined and substituted into the above relationship.  Thus,
                                      3
for example, if X  is  in units of pg/m  in the  air, then for case 1, d = 0.29
    1/3     -3      2/3                                       3
x 70    x 10   mg/kg  /day,  and for case 2, d = 1, when ug/m  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 ing/in  is

                         1  ppm = 1.2 x molecular weight (gas  mg/m
                                       molecular weght (air
                                    7-28

-------
Note that an equivalent method of calculating unit risk would be to use mg/kg
for the animal  exposures  and then increase the jth polynomial  coefficient by
an amount

                         (Wh/Wa)j/3     j = 1,2, .... K

and use mg/kg equivalents for the unit risk values.
7.1.4.7      Adjustment for  Less  than Natural  Lifetime Experiment—If the
duration of experiment  (Lg)  is less than  the  natural  lifespan of the test
animal (L), the slope q*, or more  generally the exponent g(d), is increased
                               3
by multiplying  a  factor (L/Le) .   We assume that  if the average dose d,  is
continued, the  age-specific  rate of cancer will continue  to increase as a
constant  function of  the background  rate.   The  age-specific rates for
humans increase at  least by the  2nd power of  the  age and  often  by  a con-
siderably higher  power as  demonstrated by Doll  (1971).   Thus,  we  would
expect the cumulative tumor rate to increase  by at least the 3rd power of
age.  Using this  fact,  we assume that  the slope  q*, or more  generally the
exponent g(d),  would  also increase by  at  least the 3rd power of age.  As a
result, if the  slope  q? [or g(d)]  is  calculated  at age L  , we would expect
that if the  experiment had been  continued for the full  lifespan, L, at the
given average exposure,  the slope q? [or  g(d)] would  have  been increased by
at least (L/Le)3.
     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 probability  of  cancer by age t  and at dose d  is
given by

                       P(d,t) = 1 - exp [-f(t) x g(d)]

7.1.4.8   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 con-
centrations.    First,  there  are  important species  differences in uptake,
matabolism,  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
                                    7-29

-------
doses for  humans  compared  to animals on a mg/surface  area basis is virtu-
ally  without  experimental  verification regarding  carcinogenic response.
Finally, human  populations are  variable  with respect  to genetic consti-
tution  and diet,  living environment, activity patterns,  and  other cultural
factors.
     The unit  risk estimate  can give a rough  indication of the relative
potency of  a given agent 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 by inhalation.
     The quantitative  aspect  of the carcinogen risk assessment is included
here  because  it may  be  of use in the  regulatory decision-making  process,
e.g., setting regulatory priorities,  evaluating  the adequacy of technology-
based controls, etc.   However,  it should  be recognized  that  the estimation
of cancer  risks to humans  at low levels of exposure is  uncertain.  At best,
the  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 very well be
considerably lower.   The risk  estimates  presented  in  subsequent  sections
should  not be  regarded  as  an accurate representation  of the true cancer
risks even  when the exposures are accurately defined.   The estimates pre-
sented  may  be  factored  into  regulatory decisions  to  the extent that the
concept of upper risk limits is found to be useful.
7.1-4.9  AlternativeMethodologicalApproaches—The methods  used by the CAG
for  quantitative  assessment are  consistently conservative,  i.e.,  tending
toward  high estimates  of risk.   The most important part of the methodology
contributing to this conservatism in this respect is the linear non-threshold
extrapolation model.   There are  a  variety of  other  extrapolation models
that  could be  used, all of which would give lower risk estimates.   These
alternative models  have  not been used by the CAG  in  the following analysis
but  are included   for  comparison in the Appendix E.   The models presented
there are  the one-hit,  probit and Weibull.  With the  limited data available
from these  animal  bioassays,  especially at the high-dose levels required for
testing, almost nothing  is  known about the true shape of the dose-response
curve at low environmental  levels.  The position is taken by  the CAG that the
                                    7-30

-------
risk estimates obtained  by  use of the linear non-threshold model are upper-
limits and the true risk could be lower.
     Another alternative  method involves the choice  of animal  bioassay as
the basis  for extrapolation.   The  present approach  is to use  the most
sensitive  responder.   Alternatively,  the  average  responses of  all  of  the
adequately tested bioassay animals could be used.
     Extrapolations from  animals  to  humans could 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;  it is not clear
which of  the two approaches are  more  appropriate  for carcinogens.   In the
absence of information  on this point, it  seems  appropriate to  use the most
generally obtained  method,  which also is more conservative.  In the case  of
epichlorohydrin  drinking water studies,  the  use  of extrapolation based on
surface area rather than weights increases  the  unit risk estimates by a
factor of 5.8.
7.1.4.10   Estimation of Unit Risk Based on Human Data—If  human epidemic-
logic studies  and  sufficiently valid exposure information  are available for
the compound,  they are always  used in some way.   If they 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  equivalent to the factor
Bu.   If  they show  no carcinogenic effect when  positive animal   evidence is
 n
available, then  it  is  assumed that a risk does  exist, but it  is  smaller
than could have  been observed  in the epidemiologic study,  and an upper-limit
to  the cancer incidence  is  calculated assuming hypothetically that the true
incidence  is just  below  the level of  detection  in the  cohort studied, which
is  determined largely  by the  cohort  size.   Whenever possible,  human data
are used in  preference to animal bioassay data.
     Very  little information  exists  that  can be  utilized to extrapolate
from  high exposure occupational  studies  to  low  environmental levels.
However,  if  a 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 compared  to the control group.
The mathematical model  employed assumes that for  low exposures  the lifetime
                                    7-31

-------
probability of  death  from  lung  cancer  (or any cancer), PQ, 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 some  units,  say
ppm.    The factor  BH  is the  increased probability  of cancer associated
with each unit increase of the agent in air.
     If we make  the assumption  that R, the  relative risk of lung cancer for
exposed workers, compared  to  the general  population,  is  independent  of  the
length or age of exposure but depends only upon the average lifetime exposure,
it follows that
                            P0   A + BH
   or
                            RPQ = A + BH (Xj + x2)

where x,  = lifetime average  daily exposure to the  agent for the general
population, x« =  lifetime  average daily exposure to the  agent in the occu-
pational setting, and  PQ = lifetime probability of  dying of cancer with no
or negligible epichlorohydrin exposure.
     Substituting Pfl =  A +  BH x, and rearranging gives

                            BH = PQ (R - l)/x2

To use  this model,  estimates of R and x~ must be obtained from the epidemi-
ologic studies.  The value PQ is derived from the  age-cause-specific death
rates for  combined  males  found in 1976  U.S.  Vital  Statistics tables using
the life table methodology.   For lung cancer the estimate  of P. is  0.036.
This methodology is  used in the section on unit risk based on human studies.

7.1.5  Interpretation of Quantitative Estimates
7-1.5.1  Unit Risk Estimate Based on Human Studies—In making a risk  estimate
from the Shell epichlorohydrin workers, the confounding  effect of epichlor-

                                    7-32

-------
ohydrin and IPA  exposures  cannot be ignored.   Neither can the fact that the
SMR increase was  not  statistically significant.   Because this study was in-
conclusive and not positive, only an upper-bound  to the risk can be calculated
based on the sufficient evidence of carcinogenicity in animals.
     The basis for calculating a risk estimate is the one-sided upper 95% con-
fidence limit of  the  SMR for  respiratory cancer  (International Classification
of Diseases  160-164).  Enterline's corrected cause  of  death classification
(Enterline 1981,  Table  7)  which attributes two of the ten lung cancer deaths
to other cancers  was  also  used.   In addition, all eight of the remaining 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 obser-
ved vs. 6.53 expected deaths).  The associated 95 percent confidence 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
Enterline1s report  gave the duration of exposure  prior to January 1,  1966
and the date  Df death for  the  eight  who  died from respiratory cancer.   For
these eight the  average length from beginning of exposure  to death  is  about
19.7 years.   If  the  known  time of  non-exposure between  the  beginning of expo-
sure 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 separating
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 1976, 1978).  Exposure must be  more  than 1
ppm since recent plant improvements in epichlorohydrin manufacturing facilities
have 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
*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.

                                    7-33

-------
reasoned that these  workers  were probably exposed  to  more than the 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
                   c        8 K     24° ^      13.4
        Exposure = 5 ppm x — hrs x - days x - yrs
                           24       365        50
        Exposure = 0.29 ppm

     The probability  of  dying from  respiratory  cancer  from a  lifetime
exposure to 1 ppm epichlorohydrin/m  air is given by
                         0
                        Pp(R-
                    'H =     X
                            x2

where P  ,  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, X,  is exposure at 1  ppm, and X« is the exposure expe-
rienced by the workers.  Substituting the appropriate numbers, we get

                        R  - 0.036 x (2.21 - 1) x 1 ppm _
                        DIJ """                            """ U * JLD
                                      0.29 ppm

Thus, the  upper  95% limit of the SMR  for  lung cancer based on the observed
8 deaths and the expected  6.53  deaths  yields a unit risk of 0.15.   To convert
           3
ppm to ug/m ,  the formula is

                           3                 IP'3
                       M9/m     1.2 (m.w.  chemical)/(m.w. air)
                           3
                       Mg/m  _    1.2 (92.5)7(28.8)
                            3                            -4
                      1 jjg/m  epichlorohydrin = 2.59 x 10   ppm
                                    7-34

-------
Thus, the upper limit of risk of death from lung cancer from breathing 1 ug/m3
epichlorohydrin is

                  2.59 x 10"4 x 0.15 = 3.9 x 10"5 (ug/m3)"1

     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 evalu-
ation and  the quantitative extrapolation process.   The  plausibility of the
upper bound is enhanced when there is clear evidence of mutagenicity, which is
the case for epichlorohydrin.
7.1.5.2  Unit Risk  Based on  Animal  Studies—Because  of the  limitations  dis-
cussed previously the bioassay of epichlorohydrin in the  drinking  water of
male Wistar rats (Konishi  et al. 1980,  Kawabata  1981)  may be considered  a
pilot study.   The results  of this study  nevertheless are chosen as  presenting
adequate evidence  of carcinogenic!*ty  for calculating a unit risk  by  the
drinking water  route.   The pertinent cancer data, shown  in  Table 7-4, present
a dose-response  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 signi-
ficant at  the P < 0.001 level  for  high  dose vs. control  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 unit of
forestomach area.  Unfortunately, the relative surface areas of the  two  species
cannot be  compared.   Furthermore, dose to the target organ  is also  dependent
on comparative  residence  time, on which  we have no  information.   Therefore,
the  estimates of unit risk based on this study are  subject to the caveats of
the  uncertain exposure.
     The method  chosen  to  estimate exposure is  to  determine the equivalent
concentration  of epichlorohydrin a  human must ingest in drinking  water in
order to adjust for the  difference in the water  ingestion to body weight
ratios in the two species.    In formula terms this is
                                    7-35

-------
      water, rat  x  water cone,  (rat) =  water,  human x water cone, (human)

This is equivalent to setting doses equal on a dose/body weight basis as oppo-
                          2/3
sed to  dose/(body weight)    as discussed in the  methodology section.   The
water to body  weight ratios were given earlier as  water = 0.078 for the rat
and  water = 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 concentrations for
the three dose  groups were 375, 750, and 1500 ppm.  For amount ingested, the
authors state that  total  dose was 5.0  gm, 8.9 gm  and  15.1 gm, respectively.
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 1398 ppm, respectively.  Adjusting the treat-
ment levels in  the  bioassay, 375 ppm in the rat would  be equivalent to 1160
ppm in the human.   The other  equivalent human doses are 1974  ppm and 3775 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 gm. 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 fit  using the  linearized  multistage model with  the  equivalent human
dosages.  The upper-limit maximum  likelihood estimate of the  linear component
is

                             q*  = 4.7 x I0"3(mg/kg/day)"1

     Because the  experiment  was  conducted for only 81  weeks,  the adjustment
factor  for the  less-than-natural-lifetime experiment is (104/81)   =  2.1 as
discussed in  a preceding  section.   Thus, the  final  value  of the linear
component is

               q*  = 4.7 x 10"3 x 2.1 = 9.9 x 10"3  (mg/kg/day)"1.

     In order to  estimate a  unit risk for 1 ug/1 of water  it  was assumed that
the average 70  kg human drinks 2  liters  of  water per  day.   Since  2  liters
                                    7-36

-------
weigh approximately 2 kg,  it was estimated that 1 ug/1 water corresponds to 2
ug/day.   Dividing by 70  kg gives 2.9 x  10~2  ug/kg/day or 2.9 x lo"5 mg/kg/
(b.w.)/day.  The upper-limit unit risk corresponding to 1 ug/1 epichlorohydrin
concentration in water is then

             P = 1 - exp (-9.9 x 10"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/m  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

                       3         33         -3
                 1 ug/m  = 1 ug/m  x 20 m /day x 10   mg/ug x 1/70 kg

                         = 2.86 x 10"4 mg/kg/day
                                                              3
     The upper-limit estimate  of the unit risk, P,  of 1 pg/m  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 10"6
     This  is  a  factor  of 13  greater than that  of  the animal inhalation
study 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.  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.
These results are  summarized in Table 7-2.  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
                                    7-37

-------
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 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
exposure.   The control  groups of 100 (sham) for  life  and 50 untreated were com-
bined;  no squamous carcinomas were seen.   Since  epichlorohydrin was administered
as a partially soluble vapor the concentration in ppm  in experimental  animals
is considered  equivalent  to the same concentrations in humans.  Thus,  no cor-
rections   are  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 is

                       q* = 8.5 x  10"4 (ppm)"1
                                      3
        For unit risk in terms of  ug/m ,  we make the  transformation

                            -3                              -3
                          10   ppm                        10   ppm
                  1.2 (m.w.  chemical)/(m.w.  air)       1.2 (92.5)/(28.8)
                = 2.59 x 10"  ppm
                         3
   thus, in terms of ug/m

        q*  _  8.5xlO~4 (ppm)"1 x 2.59 x 10"4 ppm _  2.2xlO~7 (ug/m3)'1
         n                            1 ug/m3

7.1.5.3  Summary of UnitRisks—Three  upper-limit unit  risk estimates were
calculated for epichlorohydrin.   All  three have more  uncertainty than  those
of other  suspect carcinogens that the CAG has evaluated.   Epichlorohydrin
is, however, among the weakest of these.
                                    7-38

-------
     Quantitative unit risks were calculated for epichlorohydrin via both the
drinking water and  inhalation  routes.   For drinking water exposure the study
of male Wistar rats (Konishi  et al. 1980) was used to estimate a unit risk,
2.8 x  10   ,  for  a lifetime exposure to  drinking  water containing 1 ug/1 of
epichlorohydrin.  This estimate  has  the uncertainty of estimated exposure to
the target organ.
     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  unit risk estimates are  not close: the upper-limit
                           -7      3 -i
for animal data  is 2.2 x 10   (ug/m )  ; the upper-limit for human data is 3.9
x 10    (ug/m )   .   In units of  risks  per  ppm, the  upper-limit  estimates are
        ~4      —1                — 1
8.5 x  10    (ppm)   and 0.153 (ppm)   for animals  and  humans,  respectively.  In
view of the weakness of both inhalation data bases, these inhalation unit risk
estimates  must be  taken  with caution.   Animal exposures 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 a lifetime;
however, only this  high  concentration group developed a significant increase
in cancers.   Because it  was short-term, however, and  not consistent with the
result  of  the lower dose groups only,  the  two lifetime exposures were  used  to
provide an upper-limit estimate of  risk.   Thus, while we  are  producing a  unit
risk estimate for epichlorohydrin because of nasal carcinoma response in the
animal  short-term  high exposure group, we are not using that high dose group
to estimate the  unit risk.  In essence, we are calculating  unit risk estimates
for air based on a  qualitative  assessment  of  sufficient evidence  for carcino-
genicity  in animals.   Likewise,  the unit risk estimate based  on human  studies
is  also an upper-bound on  nonstatistically  significant  increases in cancer
mortality.
7.1.5.4  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 expressed
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 CAG.   The  actual
data summarized  by  the histogram are  presented  in  Table  7-8.   When positive
                                    7-39

-------
4th 3rd 2nd 1st
QUARTILE QUARTILE . QUARTILE QUARTILE


I I
1x10*' 4x10" 2x10°


7


1
n



12






6


i i i
17
^•M








6
2
i n i
nl I o n
I I I
•202468
                   LOG OF POTENCY INDEX

Figure 7-7. Histogram representing frequency distribution of the
potency indices of 53 suspect carcinogens evaluated by the Car-
cinogen Assessment Group.
                           7-40

-------
TABLE 7-8.  RELATIVE CARCINOGENIC POTENCIES AMONG 53 CHEMICALS EVALUATED BY
     THE CARCINOGEN ASSESSMENT GROUP AS SUSPECT HUMAN CARCINOGENS1'2'3
Compounds
Acrylonitrile
Aflatoxin B,
Aldrin
Allyl Chloride
Arsenic
B[a]P
Benzene
Benzidine
Beryllium
Cadmi urn
Carbon Tetrachloride
Chlordane
Chlorinated Ethanes
1 , 2-di chl oroethane
Hexachl oroethane
1,1,2, 2-tetrachl oroethane
1,1, 1- tri chl oroethane
1,1,2-trichloroethane
Chloroform
Chromium
DDT
Dichlorobenzidine
1,1-dichloroethylene
Dieldrin
Dinitrotoluene
Slope ,
(mg/kg/day)"1
0.24(W)
2924
11.4
1.19x!0"2
15(H)
11.5
5.2xlO~2(W)
234(W)
4.86
6.65(W)
1. 30x10" l
1.61
6.90xlO"2
1.42x10 *
0.20 ,
1.6x10 %
5.73x10"^
7xlO~2
41
8.42
1.69
1.47xlo""1(I)
30.4
0.31
Molecular
Weight
53.1
312.3
369.4
76.5
149.8
252.3
78
184.2
9
112.4
153.8
409.8
98.9
236.7
167.9
133.4
133.4
119.4
104
354.5
253.1
97
380.9
182
Potency
Index
lxlO+1
9X10*5
4X10*3
9X10"1
2xlO+3
3xlO+3
4x10°
4xlO*4
4xlO+1
7xlO+2
2xlOn
7xlO+2
7xloo
3x10^
3x10^:
2xloo
8xlOu
8x10°
4xlO+3
3xlO*3
4X10"*"2
lxlO+1
IxlO*4
6xlO+1
Order of
Magnitude
(log,0
Indei0
+1
6
+4
0
+3
+3
+1
+5
+2
+3
+1
+3
+1
0
+1
-1
+1
+1
+4
+3
+3
+1
+4
+1
                                  7-41

-------
TABLE 7-8.   (continued)
Compounds
Tetrachlorodioxin
Diphenylhydrazine
Epichlorohydrin
Ethyl ene Di bromide (EDB)
Ethylene Dichloride (HOC)
Ethyl ene Oxide
Formaldehyde
Heptachlor
Hexach 1 orobutadi ene
Hexachlorocyclohexane
technical grade
alpha isomer
beta isomer
gamma isomer
Nickel
Nitrosamines
Dimethyl ni trosami ne
Di ethyl ni trosami ne
Dibutylni trosami ne
N-nitrosopyrrolidine
N-ni troso-N-ethyl urea
N-nitroso-N-methylurea
N-ni troso-diphenyl ami ne
PCBs
Tetrachloroethylene
Toxaphene
Tri chl oroethyl ene
Slope 1
(mg/kg/day)"x
4. 25xl05
0.77
7.8xlO~3
8.51
5.84xlfl"2
0.63 (I)
2.14xlO~2(I)
3.37
7.75x!0"2

4.75
11.12
1.84
1.33
1.15(W)

25.9(not by qf)
43.5(not by qf)
5.43
2.13
32.9
302.6 ,
4.92x10 J
4.34
5. 31xlO"2
1.13
1.26x!0"2
Molecular
Weight
322
180
92.5
187.9
99.0
44.0
30
373.3
261

290.9
290.9
290.9
290.9
58.7

74.1
102.1
158.2
100.2
117.1
103.1
198
324
165.8
414
131.4
Potency
Index
IxlO"1"8
IxlO*2
7X10"1
2xlO+3
6x10°
3X10+1
exio"1
IxlO*3
2xlO+1
+ 0
lxlO*r
3x10*;
5x10*5
4x10 £
7X10+1
+ •3
2xlO*r
4x10*,
9x10*,
2xlO+|
4x104
3xlO*4
1x10°
IxlO*3
9x10°
5xlO*2
2x10°
Order of
Magnitude
Oog10
IndelO
+8
+2
0
+3
0
+1
-1
3
+1

+3
+3
+2
+2
+1

+3
+3
+2
+2
+3
+4
0
+3
0
+2
0
       7-42

-------
                              TABLE  7-8.  (continued)
Compounds
Vinyl Chloride
Vinylidene Chloride
Remarks:
1. Slopes (q$) in mg/
Slope ,
(mg/ kg/day)"
1.75xlO"2(I)
0.13(1)

'kg/day * are calcul
Molecular
Weight
62.5
97

ated based on
Potency
Index
1x10°
1x10+1

animal oral studi
Order of
Magnitude
(1ogJQ
0
+1

es, except
     for those indicated by I (animal inhalation), W (human occupational  exposure),
     and H (human drinking water exposure).

2.    The potency index is a rounded-off slope in (mMol/kg/day) l and is calculated by
     multiplying the slopes in (mg/kg/day) l by the molecular weight of the compound.

3.    Not all the carcinogenic potencies presented in this table are final.   Some are
     subject to change as the CAG is getting the individual risk assessment documents
     approved.
                                    7-43

-------
human data  are  available  for  a  compound, they  have been  used to calculate the
index.   When  no human  data are available, animal oral  studies  and animal
inhalation studies have been used in that order.   In this 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
as follows:   the upper-limit  slope estimate from the drinking water study is
        -3            -1
9.9 x 10   (mg/kg/day)  .   Multiplying by the molecular weight of 92.5 gives a
potency  index of 9.2 x 10  .   Rounding off to the nearest order of magnitude
gives a  value of  10  which is  the  scale presented on the horizontal  axis of
Figure 7-4.  The index  of 0.92 lies in the fourth quartile of the 53 suspect
carcinogens which the CAG has evaluated; it is  among  the weakest of these
carcinogens.
7.1.6  Summary
7.1.6.1  Qualitative Assessment—The  carcinogenicity of epichlorohydrin has
been demonstrated in  rats and mice.  Epichlorohydrin vapor produced squamous
cell  carcinomas  in  the nasal  tract of male Sprague-Dawley  rats  initially
given 30 daily exposures, 6 hours each day, followed by  lifetime observation.
Consumption of epichlorohydrin  in  drinking water elicited neoplastic lesions
in the forestomach  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 evalua-
tions of decedents  were not reported due to postmortem  changes, the  81-week
duration of  the study was  less than  the lifetime of the animals,  and the
number of animals in each dosage group was small.
     Two studies  involving dermal  application of epichlorohydrin  on the
skin of  mice for a lifetime 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  weekly  to  the  skin of C3H  mice.   Hence,  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
                                    7-44

-------
significant increase in  local  sarcomas.   However,  intraperitoneal  injections
once weekly in females of this strain was ineffective.
     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  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.
     Two epidemiologic  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,  a 1979 update  of  an ongoing  study of workers at Shell
Oil Company,  showed  increased deaths from  respiratory  cancer.  Leukemia was
also present  in an otherwise  healthy cohort.   This  increase,  however, was
not  statistically  significant,  and  the trend  in  the most  recent 2-year
follow-up  period  actually weakened  the evidence that epichlorohydrin is  a
human  carcinogen.  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.
7.1.6.2   Quantitative Assessment—Unit  risk  estimates  for  exposure to
epichlorohydrin are  calculated from both the animal  and human studies.  For
animal studies,  unit risk estimates  are calculated  from both drinking water
and  inhalation  studies.  The  drinking water study of male Wistar rats exposed
to epichlorohydrin  in  drinking water showed epichlorohydrin  to cause tumors
of  the  forestomach.   Based on  this  study the  upper-limit lifetime  risk of
2.3  x  10~7 for a lifetime exposure  to  drinking water containing  1  ug/1 of
epichlorohydrin was  estimated.   For 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, however, the  use of the 10%  nasal carcinoma response was eliminated at
the  100  ppm exposure level because  it is a  high-dose,  short-term exposure  and
does not  multistage  model  on  the two lower  doses to  provide  a 95%  upper-limit
on risk.   The linearized multistage  model was used for low-dose extrapolation
in order  to give an  upper-bound estimate of lifetime  cancer  risk,  recognizing

                                    7-45

-------
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 for 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/m~  is 2.2 x 10  .
     The study of employees from the Shell Oil Company was also used to provide
a unit risk estimate for exposure to epichlorohydrin based on a human epidemi-
ologic study.  This study showed an increase in respiratory cancer deaths in
an otherwise healthy  population.   While this increase  was not statistically
significant, the evidence for the careinogenicity of epichlorohydrin in animals
has allowed  calculatation of  a plausible upper-limit risk estimate based on
this study.  Using these  data, the 95%  upper-limit cancer  risk resulting from
                                                                  3           ~5
continuous exposure to air with an epichlorohydrin level of 1 ug/m  is 3.9 xlO  .
7.1.7  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 Cancer Research (IARC) classifica-
tion scheme, this level of evidence would be considered sufficient for conclud-
ing that epichlorohydrin  is carcinogenic in experimental animals.   The human
epidemiologic evidence to epichlorohydrin alone is negative.   However, sequential
exposure to the IPA process and epichlorohydrin produced possible evidence for
the carcinogenicity of epichlorohydrin,  which,  however, had only marginal
statistical significance  (P < 0.1).  Using the IARC  system for describing the
overall carcinogenicity evidence,  epichlorohydrin would be classed as a 2B
chemical.
     As described in the mutagenicity section, epichlorohydrin has been demon-
strated to be mutagenic in both prokaryotic and eukaryotic systems.   Epichloro-
hydrin is  a  direct  acting alkylating agent and,  therefore, does not require
metabolic activation to attack biological macromolecules.
     Quantitative estimates of potency  were made for both drinking water and
inhalation.  Based on  forestomach tumors in male Wistar rats exposed to epichloro-
hydrin via drinking water, a lifetime exposure to 1 ug/liter epichlorohydrin in
drinking water was estimated  to  present an upper-limit risk  of 2.8 x 10  .
     Quantitative estimates of potency  via  inhalation  were made from both
animal data on nasal  carcinomas and human data on a nonstatistically significant

                                    7-46

-------
 increase  in respiratory  cancers.   These two  unit risk estimates  are  not
 close;  the  upper-limit estimate from nonsignificant human data is  3.9 x 10"5
 (MS/IB  )   ;  the  upper-limit estimate from positive animal  data is 2.2 x 10"7
 (ug/m  )   .  The estimate  based  on animal data does not use the nasal  carcinoma
 response  based on short-term high exposure, because the CAG feels that such an
 exposure  does not reflect environmental experience, and is not consistent with
 the  long-term lower dose  response.
     The  carcinogenic potency of epichlorohydrin  lies in the  fourth  quartile
 among  53  suspect carcinogens evaluated by the CAG.  It is among the weakest of
 the  substances that the CAG has evaluated as suspect carcinogens.
 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  hisG46,
TA100,   TA1950,  TA1951,  TA1952, TA1534,  TA1537,   and  TA1538  were  used.
 Epichlorohydrin (purity not given)  concentrations of 1 percent (0.05 umole),
                                    7-47

-------
5  percent  (0.27 umole),  10 percent  (0.54  pinole), and  100 percent (1.10
(jmole) were  employed.   Epichlorohydrin was  applied in 50 ul 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).
     In the suspension assay,  only strains hisG46  and TA100 were used.  The
cells of the  strain  hisG46  were treated with epichlorohydrin  at concentra-
tions 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,
         -2           -1               -1
5.40 x 10  ,  1.08 x 10  ,  and 5.40 x 10   M for  60 minutes without a metabolic
activation system  (S-9) and assayed  for revertant  colonies.  The concentra-
tion of  5.40  x 10   M was  toxic  and produced 100 percent cell  killing.
                        Q
Numbers of revertants/10  survivors were  6, 4,  9,  18,  15,  1.68 x 103, 3.18
x  107  for the above  concentrations,  respectively.   In  TA100, epichloro-
                                    -2            -3           -2
hydrin concentrations  of  1.08  x  10  , 5.40  x  10  , 1.08  x 10   ,  5.40 x
  -2                 -1
10  , and  1.08  x  10    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 1010,
9.64 x 101,  2.85 x 10s, 3.44 x  10s, 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, 2.0t  and  2.5 umoles/plate  (in 100  ul 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
                                    7-48

-------
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  umoles/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.
                  1200
                a.
                ac
                tu
                a.  900
                U)
                uj

                3
                o
                u
                £  600
Ul
ff

O
6
                   300
             0.5      1.0      1.5

                   DOSE PER PLATE, p mole
                                                     2.0
2.5
                     Figure 7-8. Mutagenicity of aromatic epoxy resins and
                     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 Mine (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).
                                     7-49

-------
   3500
UJ

5
Q.
z
o
o
o
I-
z
<
oc
III
   1000
    500
    100
      10
100               1000

CONCENTRATION.
5000
       Figure 7*9. Mutagenicity of epichlorohydrin ( • ), at
       various concentrations (nmol/ml of soft agar) in
       5. rvp/r/mw/c/mTAIOQ8.


       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).
                           7-50

-------
   1500
UI


s
Q.

K
ui
EL
WITH S 9


WITHOUT S-9
   1000
O
U

I-





£   500
           TA-100
                   I
            I
I
I
                 200         400         600         800



                      COMPOUND IN TOP AGAR. \i mole




          Figure 7-10. Dose-response curves for epichlorohydrin.




          Source:  Stolzenberg and Mine (1979).
                                             1000
                               7-51

-------
     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
                                            -2
an epichlorohydrin concentration of 1.1 x 10   umoles/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 ug/1 air to be
mutagenic when the plates were incubated in sealed airtight containers.
                                    7-52

-------
                                 • WITH S-9 MIX

                                 O WITHOUT S-9 MIX
                10                20

          M MOLES/INCUBATION VOLUME (2 ml)
Figure 7-11. Mutagenicity of epichlorohydrin with (  • )
and without ( O ) S-9 mix.

Source:  Eder et al. (1980).
                    7-53

-------
     Laumbach et  al.  (1977) reported epichlorohydrin to  be  mutagenic  for  S.
typhimurium  strain TA100.   Epichlorohydrin  at  a concentration  of 4.746
uuwles/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  TA1003  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 pneumorn'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.   Conse-
quently, 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 (50% LD5Q)  and  100  (100%  LD5Q)  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  epichloroh-
ydrin is mutagenic in the host-mediated assay  employing Salmonella strains
G46,  TA100, and TA1950.
                                    7-54

-------
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
Pol A  plate assay of  Rosenkranz  and  the Rec-assay of Kada (Bridges  1981:
Elmore 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  0/106 survivors.   The positive mutagenic  effect  of  epichlorohydrin  in
N. crassa was  also confirmed by Westergard  (1957).
                                     7-55

-------
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
07.  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/10  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/10  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 (Del 1 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 small,  were  induced by epichlorohydrin.   The
large  mutant colonies followed a  linear dose-response  relationship indica-
ting  a  typical one  hit point  mutations!  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.
                                    7-56

-------
                                20
40
60
                       CONC. OF EPICHLOROHYDRIN (Mg/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-LinkedRecessive Lethal  TestIn 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 umoles of  epichlorohydrin and individually mated to 3 Base females per

                                    7-57

-------
                                           TABLE 7-9.  INDUCTION OF SEX-LINKED RECESSIVE LETHALS IN OROSOPHILA BY EPICHLOROHYDRIN
 i
L71
CO


Experiment
No.
1
2
3
4
5
6b
CD

Concentration
(mM)
2.6
5.1
25.5
25.5
2.6
5.1

Method
of Admin-
istration
Injection
Injection
Injection
Injection
24-h feeding
24-h feeding

Brood
No. of
Chromosomes
526-
958
448
922
628
773
5216
A
%
Lethal
0
0.52
0.67
1.74
0
0.13
0.17
Brood B
No. of
Chromosomes
537
1057
528
836
772
701
5027

*
Lethal
0.93
0.47
1.33
0.96
0
0
0.22
Brood C
No. of
Chromosomes
521
957
412
801
580
702
5247

%
Lethal
0.19
0.31
0.49
0.37
0
0.28
0.17
Brood 0
No. of
Chromosomes
525
1018
450
844
704
603
2050

%
Lethal
0
0.10
0.22
0.47
0.14
0
0.34
Brood E
No. of
Chromosomes
465
1108
252
817
—
811
1735

X
Lethal
0
0.45
0.79
0.12
--
0
0.17
    C = Accumulated laboratory control.

    NOTE:   Four-day-old Oregon-K males were treated and individually mated to three Base females (virgins) per brood; 0.7% NaCl or 5% sucrose was added in injection
    and feeding solutions, respectively.   After injection of 25.5 mM, 75 percent of the flies were fertile through brood E; after feeding of 2.6 mM, this, was the same;
    during feeding of 5.1 mM, however, 70 percent of the flies died within 24 hours, while 25.5 mM was 100 percent lethal; here a higher stickiness of the solution may
    have played a role.   DMSO was used (except in experiment 6) as an auxiliary solvent at final concentrations of 0.5% or lower, although it was not necessary to get
    perfect mixing of epichlorohydrin in water.
    Source:   Knaap et al.  (1982).

-------
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 organs 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
                                    7-59

-------
(Czechoslovak Chemical  Industry) was  added for the  last  24  hours of culti-
                               _C    -C    —7    -O    -Q    -Ifl    -11
vat ion at concentrations  of  10  ,  10  ,  10  , 10   , 10   , 10   , 10    , and
  -14                                                                    -4
11    M.  In  the  second series of experiments, cells  were exposed to  10
and 10   M  concentrations of epichlorohydrin in three ways:  (1) for 1 hour
before the  beginning  of cultivation  (G^);  (2) for  1 hour between the  24th
and 25th hour of  cultivation (Gj); 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 (Gn and  G,)  and control groups.
                                              u      i            -4
However, in  the  24-hour  treatment group,  the concentration of 10    M was
too toxic;  only 10   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 (Gg)
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
                                    7-60

-------
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 (LC50) was 40 ug/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
mg/m3)  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
                                                     2
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
                                    7-61

-------
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-
                                    2
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  mutagem'c  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--C1astogem'c   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).
                                    7-62

-------
     Sram et al.  (1976)  studied the in vivo clastogenic effects of epichlor-
ohydrin  in  mouse bone marrow cells.   Epichlorohydrin  (LD5Q 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.
                                    7-63

-------
7.2.8.5  Micronucleus Assay—Epichlorohydrin  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%
LD50),  0.045 mg/kg (25%  LD5()),  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  LD50), 0.04  (25% of LD50)t 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

                                    7-64

-------
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
                                              -3         -4         -4
follows:  (1) cultures  were exposed to 1  x  10   , 4 x 10 ,  2  x 10  ,  1 x
                                    7-65

-------
  -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
                             -3        -5        -4            -4
three concentrations, 1  x  10  , 8 x 10  , 4 x 10   , and 2 x 10   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
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 were significant at concentrations  of 1 x  10    M  (p<0.001) and 1 x
  _o
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-
ot ic 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
                                    7-66

-------
killing  of repair-deficient  bacteria.   Chromosomal  effects  induced by
epichlorohydrin were  detected both in vivo and  in  vitro mammalian assays.
The tnicronucleus 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 12 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
                                    7-67

-------
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
                      No.  of Days
Dai ly
Dose            Average Weekly Litter Size
Compound
alpha-Chlorohydrin

Epichlorohydrin


of Exposure
5
1
5
5
1
(mg/kg)
20
10
20
50
100
Weeks:!
0
0
0
0
0
2
0
0
0
0
4
3 4
7 9
0 0
0 11
0 0
3 4
5
3
0
11
0
4
6

0

0
2
7

0

0
2
8

0

0
4
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
                                    7-68

-------
-g
CTi
                        TABLE  7-11.   THE EFFECTS OF INHALED EPICHLOROHYORIN ON THE SEHEN OF RABBITS AND ON  THE  FERTILITY OF  HALE  AND  FEMALE  RATS
       Compound:      Epichlorohydrin
       Species:       New Zealand white rabbits (males) and Sprague-Oawley rats (males and females)
       Exposure:      Inhalation, 6 hours/day,  5 days/week, 20 weeks
       Level:         0,  5,  25, and 50 ppm
       Group Sizes:   10  male rabbits, 30 male and 30 female rats per level of exposure
                                                                                   -EXPOSURE	POSTEXPOSURE-
       RABBITS                           Weeks:        -2   -1   0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19   20
       Semen evaluation                               XXXXXXXXXXXXXXX         X         X         X         X
       Hating                                                                                           X
       RATS
       Fertility:  Hales (matings)                              X       X       X           X           X         X                                       X
                   Females (matings)                                                                    x
S     Toxicology Research Laboratory, Dow Chemical Company.
       Source:  John et al. (1979).

-------
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  unexpcsed  female  rats.   After  the
females were mated  with the exposed  males,  there were significantly fewer
                                    7-70

-------
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   Hale Clinical-Epidemiologic Investigations--Mi1by  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  Norco,  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.
                                    7-71

-------
     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-Oawley) and  rabbits  (New Zealand).  For the
tolerance study,  five or six pregnant  rats  and  five pregnant rabbits were
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,
                                    7-72

-------
Texas.   Concurrent  controls were exposed to  filtered  air.   Exposure to 50
and 100 ppm epichlorohydrin  in the tolerance  study produced signs of maternal
toxlcity  (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,
                                    7-73

-------
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 concept!.
     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);
                                    7-74

-------
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.   Epichlorohydrin's  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,
                                    7-75

-------
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.
                                    7-76

-------
          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
rag/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-
cardiograph!" c 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 (ppm)
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 IT
Bromosulphalein
Retention after
36.6/36.7
36.9/36.9
426/414
410/398
0.83/0.86
0.1/6.5
36.2/36.3
37.1/37.2
331/348
318/350
0.87/0.83
1.4/4.2
35.5/36.1
36.7/36.9
275/264
281/278
0.87/0.84
2.5/3.5
33.4/33.5
36.8/36.6
235/251
244/247
1.02/1.01
8.6/18.9
  exposure*"
Urine
  24-h volume, ml
Total protein, g/100 ml
  After exposure
          2.6/4.3
4.9/2.6
4.0/3.0
          7.08/6.81    6.78/7.10   6.90/7.44
4.4/4.6

7.80/8.01
aSource:  Shumskaya et al. (1971).
 One day after exposure.
cAfter 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
       o
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 heat  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 LC5fl 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 al., 1971).  Another study of rats  indicated that heat (35°C) enhanced the
toxicity of epichlorohydrin (Grigorowa et al. 1977).

   TABLE 8-2.  THE EFFECT OF HEAT STRESS ON THE LD50 OF EPICHLOROHYDRIN
                           IN THE RAT AND MOUSE3
Species
Rat
Mouse
Condition
No heat
With heat
No heat
With heat
LC50
(mg/1)
2.40
2.20
3.00
4.00
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
aeruglnosa (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
bioconcentration.   Bioconcentration  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 0).   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

-------
                                      TABLE 9-2.   EPICHLOROHYORIN TOXICITY TO FOUR FISH MID ONE AQUATIC INVERTEBRATE
       Species
                          Temperature
           Toxic Level
No-Effect Level
Test"
                           Comments
                                       Reference
FISH
  FRESHWATER
    Bluegill                  23
    (Lepoais nacrochlrus)


    Goldfish                  20 l 1
    (Carassius auratus)
    Ide                       20 t 1
    (Leyciscus idus
     nelanotus)

  SALTWATER
    Tidewater silverslde      20
    (Henidia beryllina)

INVERTEBRATE
    Waterflea                 20
    (Daphnia «aona)
 S,U       96- h LCSO = 35
 S,H       24-h TL  = 23b
 S,U       48-h LCSO = 24
 S,U       96- h LC50 = 18
 S,U       24-h LC50 = 30
       10
       23
       12
       10
       20
Test conducted in well
water:  pH 7.6-7.9, hard-
ness 55 mg/1 as (CaC03)

Only study with measured
epichlorohydrin levels;
chemically defined tapwater.

Test conducted in tapwater:
pH 7-8 hardness 268 t 54 mg/1
"Instant Ocean" sea salt nix;
sp gr = 1.018
Test conducted in chlorine-
free tapwater at pH 7.6
Dawson et al.  (1977)



Bridie* et al.  (1979b)



Juhnke and LUdemann (1978)




Dawson et al.  (1977)



Bringmann and KUhn (1977)
aS = Static; U = Unmeasured Concentrations; H = Measured Concentrations.
Median Tolerance Liait.

-------
              Table 9-3.  The Acute Toxicity of Epichlorohydrig to
                        Bluegill and Tidewater Silverside Fish
Initial
Concentratioi
Species (mg/1)
Bluegill
(Lepomis
macrochirus)



Tidewater
silverside
(Menidia
beryl lina)

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 r°


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 LC5fl  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  LC5fl for Daphnia magna was deter-
mined  to be 30  mg/1 (Bringmann  and Kuhn  1977).  In this static test,  no
0. 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 beryl!Jan.
an estuarine fish.   Oawson  et al. (1977) reported the 96-hour IC™ 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
LCgQ values  were determined as  for the  bluegill (Section  9.3.1).   The
death rate of control  fish during the experiment was  acceptably  low at 3.0
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  biconcentration
suggest that epichlorohydrin would not accumulate substantially in food
chains.
     Limited toxicity data for five aquatic animals  indicated that exposure
to epichlorohydrin concentrations of  less than 10 mg/1 for 1-4 days would
not be harmful.   In  only one of the tests, however,  was the actual epichlo-
rohydrin 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
     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
MAC3
MAC
MAC
MAC
MAC
MAC
TWAe
Maximum
b-
Source:
Sources:
Source:
Country
Netherlands
U.S.S.R.
Czechoslovakia
Federal Republic of Germany
German Democratic Republic
Rumania
U.S.
Allowable Concentration
IRPTC (1979).
Winell (1975), Sram et al. (1980).
Wexler (1971).
Level (ppm)
2.0br
0.26^
0.26C
3'6r
l'°*
2.6 *
5.0T




Time-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:   In  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 HW,  Yousef  SH,  Tadros FE, Soliman NZ.   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.

     k P, Peterka V.  1971.  The determination of epichlorohydrin in aqueous
     solutions in  the presence  of  glycerin, monochlorohydrin  and glycidol.
     Analyst 96:807-809.           a jr      ,            y         a *

Addy JK,  Parker  RE.  1965.   The mechanism of  epoxide reactions.  Part  VII.
     The reactions of 1,2-epoxybutane, 3,4-epoxybut-l-ene, 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.

ACGIH.   1979.  American Conference of Governmental  Industrial Hygienists.
     Documentation  of  threshold limit values  for chemical  substances in
     workroom    air.  4th ed, pp.  16-17, 34-37.

AIHA.  1961.   Epichlorohydrin  (l-chloro-2,3-epoxypropane).   Am. Ind.  Hyg.
     Assoc. J.   22(6):511-513.

APHA.  1976.  American  Public Health Association.  Standard Methods for the
     Examination of Water and Wastewater, 14th ed. Washington, DC:   American
     Public Health Association, p. 1076.

Mines BN,  McCann  J, Yamasaki E.   1975.   Methods  for detecting carcinogens
     and    mutagens with the Salmonel1a/mammalian-microsome  mutagenicity
     tests. Mutat. Res.  31:347r3Pk

Andersen M, Kiel P, Larsen H, Maxild J.   1978.  Mutagenic action of aromatic
     epoxy  resins.  Nature (London) 276:391-392.

Anderson GE, Liu CS, Holman  HY, Killus JP.  1980.  Human exposure to atmos-
     pheric concentrations of selected chemicals.  Appendix A-13 - A Summary
     of  Data on  Epichlorohydrin.   Research Triangle  Park, NC: Office of Air
     Quality Planning and Standards, 20 pp.

nndersson K, Levin JO,  Lindahl  R,  Nilsson CA.  1981.  Sampling of epichlor-
     ohydrin  and ethylene chlorohydrin  in workroom  air  using  Amberlite
     XAD-7  resin. Chemosphere 10(2):143-146.

p-,-ndel G.   1971.  Methods for  inducing mutations in Eucalyptus.   Cicenc.
     Cult.  23(3):241-246.  (In  Portuguese; abstract  in English)

Bartsch  H,  Malaveille  C,  Camus AM, 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
     various metabolizing systems.  Mutat. Res.   76:1-50.
                                  R-l

-------
Beckwith ALJ.  1972.  Phosphorous and hypophosphorous acid derived radicals
     and their  reactions  -  an  E.P.R.  study.   Aust.  J.  Chem.   25:1887-1905.

Blackfofd JL   1978.   Marketing research report on  epichlorohydrin.   In:
     Chemical Economics  Handbook.   Menlo Park,  CA:   SRI,  pp. 642.3021A-
     642.3022V.

Bridges  BA.   1978.   On  the detection of volatile  liquid  mutagens with
     bacteria:   Experiments with dichlorvos and epichlorohydrin.  Mutat.
     Res.  54:367-371.

Bridie AL,  Wolff CJM, Winter  M.   1979a.   BOD  and  COD  of some petro-
     chemicals.   Water Res.  13:627-630.

Bridie AL, Wolff CJM, Winter M.  1979b.  The  acute  toxicity  of  some  petro-
     chemicals to goldfish.   Water Res.  13:623-626.

Briggs GG.   1973.   A simple relationship  between adsorption of organic
     chemicals and their octanol/water partition coefficients.  Proceedings
     of the  7th  British  Insecticide and Fungicide  Conference,  pp. 83-85.

Bringmann G, Ktihn 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,  Kiihn  R.   1977.   Findings  on the damaging  action  of water
     pollutants  on Daphnia magna. Z. Wasser Abwasser Forsch.  10(5):161-166.
     (In German; English abstract)

Bringmann G,  K'uhn R.  1976.  Comparative results of  the  harmful effects of
     water pollutants  on bacteria (Pseudomonas putida) and blue-green algae
     (Microcystis aeruginosa).   Gas-Wasserfach, Wasser-Abwasser 117:410-413.
     (In German; English trans.)

Bronsted JN, 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 JG,  Pitts JN Jr.   1966.  Photochemistry.   New York, NY:   John Wiley
     & Sons, Inc.,  pp.  522-528.

Carbone P,  Barbata  G, Margiotta G, Tomasino A, Granata G.   1981.  Low
     epichlorohydrin concentrations induce  sister chromatid  exchanges  in
     human lymphocytes iji vitro.  Caryologia 34(3):261-266.

Carcinogen Assessment Group, U.S.  Environmental  Protection Agency.   1980a.
     Preliminary risk assessment of epichlorohydrin.  Type I  - air program.
     Prepared for the Office of Air Quality Planning and Standards.
                                  R-2

-------
Carpenter CP,  Smyth  HF  Jr. ,  Pozzani  UC.  1949.  The assay of acute vapor
     toxicity, and the grading and interpretation of results on 96 chemical
     compounds.  J. Ind. Hyg. Tox.  31(6):343-346.

Castro CE,  Bartnicki  EW.   1968.   Biodehalogenation.  Epoxidation of halo-
     hydnns,  epoxide opening, and  transhalogentation  by a Flavobacterium
     sp.  Biochemistry  7(9):3213-3218.                       	

Chemical  Week.  1978.  Derailment spills epi,  p.  22.

Chiou CT, Freed VH, Schmedding DW, Kohnert RL.   1977.   Partition coefficient
     and bioaccumulation  of selected  organic  chemicals.  Environ. Sci.
     Techno!.    11(5)-.475-478.

Cooper ER,  Jones  AR, Jackson H.   1974.  Effects of alpha-chlorohydrin and
     related compounds on the reproductive organs and fertility of the male
     rat.  J.  Reprod. Fert. 39(2):379-386.

Cox CR.  1972.   Regression model and  life  tables.  J.  Roy.  Stat. Soc. B
     34:187-220.

Crump  KS.   1979.   Dose-response  problems  in  carcinogenesis.  Biometrics
     35:157-167.

Crump  DS,  and Watson WW.   1979.   GLOBAL 79.   A  Fortran  program  to extra-
     polate  dichotomous animal careinogenicity  data to  low dose.  Natl.
     Inst.  of  Environ. Health Science Contract No. l-ES-2123.

Crump  DS,  Guess HA and Deal LL.  1977.  Confidence intervals  and test of
     hypotheses concerning dose-response relations  inferred from animal
     careinogenicity data.  Biometrics 33:437-451.

Dabney BJ,  Johnston  RV, Quast JF, Park CN.  1979.  Epichlorohydrin -  sub-
     chronic  studies.   III.   Cytogenetic evaluation of  bone marrow cells
     from  rats exposed by inhalation  to epichlorohydrin for four weeks.
     ICPEM    (International Commission for Protection Against Environmental
     Mutagens  and   Carcinogens)  Document No.   128,  15pp.

Daniel JW,  Gage JC.   1956.  The  determination of epichlorohydrin in  air.
     Analyst 81:594-598.

Uawson GW,  Jennings  AL, Drozdowski  D, Rider E.  1977.    The  acute  toxicity
     of 47  industrial chemicals to fresh and saltwater fishes.   J. Hazardous
     Materials  1:303-318.

Dean BJ  Hodson-Walker G.   1979.   An jn vitro  chromosome assay using cultured
     rat-liver cells.  Mutat. Res.  64:329-337.

DePetrocellis  L.,  Tortoreto M., Paglialunga S.,  Paesani  R. ,  Airoldi L., Ramos
     Castaneda  E., and Pantarotto C. 1982.   Determination of epichlorodydrin
     in blood  by gas chromatography and selected ion monitoring.   J. Chroma-
     togr., 240, (l):218-23,.
                                  R-3

-------
 Hilling  WL,  Bredeweg CJ, Tefertiller  NB.   1976.   Organic photochemistry.
      Simulated  atmospheric  photodecomposition rates of methylene chloride,
      1,1,1-trichloroethane,  trichloroethylene,  tetrachloroethylene,  and
      other compounds.   Environ.  Sci. Technol.   10(4):351-356.

 Dflling  WL.   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(4):405-409.

 Dobbs AJ, Gilbert BC, Laue HAH,  Norman ROC.   1976.   Electron  spin resonance
      studies.   Part IL.   Radicals derived from  oxirans  in aqueous solution.
      J.   Chem. Soc.  Perkin Trans.  2:1044-1047.

 Dobinson B,  Hofmann  W,  Stark  BP.  1969.   The determination  of epoxide
      groups.  Oxford:   Pergamon Press, pp. 2-61.Dow.   1980.   Dow Chemical
      Company.   Epichlorohydrin  -  properties,  uses, storage,  handling.
      Midland, MI:   Dow Chemical  Company, 37pp.

 Doll  R.  1971.  Weibull distribution of cancer.  Implications  for models of
      carcinogenesis.  J.  Roy. Statistical Soc.  A 13:133-166.

 Dripps RD,  Eckenhoff JE and  Vandam  LD.  1977.   Introduction to anesthesia,
      the principles  of safe practice.  5th  Ed.  W.B.  Saunders  Company,  Phil.
      Pa. 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  jn vitro  mutagenic  properties.
      Biochem. Pharmacol.  29:993-998.

 Elmore JD, Wong JL, Laumbach AD, Streips UN.   1976.   Vinyl chloride muta-
      genicity  via  the  metabolites  chlorooxirane  and chloroacetaldehyde
      monomer  hydrate.  Biochim.  Biophys. Acta 442:405-419.

 Enter!ine, PE.   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, PE.   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
      MO. OTS-1180-0069.

 Enterline, PE.  1981.   Further  update  of mortality among workers exposed to
      epichlorohydrin.   Report  to Shell Oil Co., Jan. 13, 1981.  Submitted
      to  U.S.  Environmental  Protection Agency,  Office of Toxic Substances.
      Docket No. 8EHQ-0878-0230  Supplement.

Enterline,  PE.   1982  Annals of the New York Academy of Sciences.  Vol 381
     pp 344-349.
                                   R-4

-------
Epstein SS, Arnold  E,  Andrea J, Bass W, Bishop  Y.   1972.   Detection  of
     chemical mutagens by the dominant lethal assay in the mouse.  Toxicol.
     Appl. Pharrnacol.  23:288-325.

Fakhouri G, Jones  AR.   1979.  Epichlorohydrin:  Metabolism and toxicity in
     the rat.  Aust. J. Pharm. Sci.  8(1):11-14.

Falco JW, Mulkey LA, Swank RR Jr., Lipcsei  RE, Brown SM.  1982.  A screening
     procedure for  assessing the transport and degradation of solid waste
     constitutents  in  surface and subsurface waters.   Environ.  Toxicol.
     Chem.    1:121-134.

FASEB.  1974.  Biological Data Books.  2nd ed., Vol. III.  Philip L. Altman
     and Dorothy S. Dittmen, eds.  Bethesda, MD:   FASEB.  Library of Congress
     No. 72-87738.

Fjellstedt TA, Allen RH,  Duncan BK, Jakoby WB.   1973.  Enzymatic conjuga-
     tion of  epoxides  with  glutathione.  J.  Biol.  Chem.  248(10):3702-3707.

Fomin  AP.   1966.   Biological effect of  epichlorohydrin and its hygienic
     significance as  an atmospheric pollutant.   Hyg. Sanit.   31:357-363.

Freuder E,  Leake  CD.   1941.  The  toxicity  of epichlorohydrin.   Berkeley,
     CA:   Univ.  Cal.  Pub. Pharm.  2(5):69-78.

Gage JC.   1959.   The toxicity of epichlorohydrin vapour.   Br.  J.  Ind.  Med.
     16:11-14.

Gay BW Jr.,  Bufalini JJ.   1971.   Nitric acid and the nitrogen balance of
     irradiated hydrocarbons  in  the presence of oxides  of nitrogen.  Environ.
     Sci. Technol.  5(5):422-425.

Gillenwater  LE.   1965.  Stream  pollution by chemical  spills from common
     carriers.  J.  Am. Water Works Assoc.  57:201-207.

Gomes,  WR.  1977.   Pharmacological  agents and male  fertility.  The Testis,
     ed. by Johnson, AD and Gomes WR, vol.  IV, pp.  605-628.

Greve  PA.  1971.   Chemical wastes  in the sea:  New  forms of marine pollution.
     Science  173:1021-1022.

Grigorowa  R,  Gohlke R, Rothe R, Weigmann  HJ.   1977.  MAC value for epi-
     chlorohydrin at   normal  and elevated ambient temperatures.  Z. Gesamte
     Hyp. Ihre Grenzgeb.  23(9):620-623.

Gusev  MI, Minayev AA.   1973.  Determination  of latent time of  reflex reactions
     of laboratory  animals  acted upon by toxic agents  in atmospheric  air.
     In-  Ministertro  Zdravoochraneniya RSFSR.   1968.   Moskovskiy Nauchno-
     Issledovatel'skiy Institut  Gigieny Im.  F.F.  Erismana.  "Voprosy gigieny
     atmosfernogo vozdukha  i planirovki naselennykh mest."  Uchenye zapiski,
     Moskva,  pp. 24-26.

Hahn,  JO.   1970.   Posttesticular antifortility effects of epichlorohydrin
    'and 2,3-epoxypropanol.  Nature. (London) 226:87.


                                   R-5

-------
llansch C,  Leo A.   1979.   Substituent  Constants  for  Correlation Analysis  in
     Chemistry  and Biology.   New York, NY:  John Wiley & Sons, p.  321.

Hawley GG,  ed.  1977.  The  Condensed Chemical  Dictionary, 9th ed.   New
     York, NY:  Van Nostrand Reinhold Co., p. 343.

"^yakawa T, Udenfriend S, Yagi H, Jerina DM.  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.  Berlin KL. Med.  1:193-228.

Hine C, Rowe VK, White ER, Darmer KI Jr., Youngblood GT. 1981.  Miscellaneous
     epoxy compounds.   In:   Clayton GO, Clayton FE, eds.   1981.   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:  The MITRE Corporation.  MTR-79W171, p. 117.

International  Commission of  Radiological  Protection (ICRP).   1977.   Recom-
     mendation of  the  International Commission  on Radiological Protection,
     Pub. No.  26, adopted Jan.  17, 1977.  Pergammon Press, Oxford, England.

IRPTC.   1979.   International  Register of Potentially Toxic Chemicals Bulletin.
     Epichlorohydrin.    2(1):11.Ippen  H,  Mathies  V. 1970.   "Protracted
     chemical  burns"  (with special  consideration  of dermatologies!  damage
     due to epoxides and propansultone).  Berufsdermatosen 18:144-165.   (In
     German; summary in English)

Jaraczewska W,  Kaszper W.  1967.  Determination of  epichlorohydrin in air.
     Med. Pr.   18(2)-.168-176.  (Summary in English)

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
     epichlorohydrin on the  semen of rabbits and on the fertility of male
     and female rats.  Unpublished  report,  Toxicology  Research  Laboratory,
     Health and Environmental  Sciences, Dow Chemical U.S.A., Midland, MI.

Jones AR, Davies P, Edwards K,  Jackson H.  1969.  Antifertility effects  and
     metabolism of alpha- and epichlorohydrins  in the rat.  Nature (London)
     224(5214):83.

Jones AR, O'Brien  RW.   1980.  Metabolism of three  active analogues  of the
     male antifertility agent  alpha-chlorohydrin  in the rat.  Xenobiotica
     10(5):356-370.
                                  R-6

-------
.luhnke I,  Ludemann  D.   1978.   Results of the  investigation of 200 chemical
     compounds for  acute  fish toxicity with the Golden Orfe test.  Z. F.
     Wass. Abwass. Forsch.  11:161-164.   (In German; English trans.)

Kalla NR,  Bansal  MP.   1977.    In vivo & in  vitro alkylation of testicular
     cysteine  by alpha-chlorohydrin.  Ind.  J.  Exp.  Biol.   15:232-233.

Kastenbaum MA, Bowman  KO.  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.   J.  Nara Med.  Assoc.
     32:270-280.

Kenaga EE.   1980.  Predicted  bioconcentration  factors and  soil  sorption
     coefficients of  pesticides and other chemicals.  Ecotox. Environ.
     Safety 4:26-38.

Kenaga EE, Goring CAI.   1980.   Relationship between water solubility,  soil
     sorption, octanol-water  partitioning,  and  concentration of chemicals
     in  biota.   In:   Eaton JG,  Parrish  PR, Hendricks AC, eds.   Aquatic
     Toxicology,    ASTM STP  707.   Philadelphia, PA:   American Society for
     Testing and Materials, pp. 78-115.

Kilian DJ, Pull in TG,  Conner  TH, Legator MS, Edwards  HN.  1978.  Mutagen-
     icity 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.   53:72.

Kirkhart  B.   1981.  Micronucleus  test on 21 compounds.  In:   deSerres  FJ,
     Ashby J, eds.  Evaluation of Short-Term Tests  for Carcinogens.  Elsevier/
     North Holland, pp. 698-704.

Knaap AGAC,  Voogd  CE,  Kramers PGN.   1982.   Comparison of  the  mutagenic
     potency of 2-chloroethanol, 2-bromoethanol, 1,2-epoxybutane epichloro-
     hydrin and glycidaldehyde in Klebslella pneumoniae, Drosophila melanogaster
     and   L5178Y mouse lymphoma cells.  Mutat. Res.   101:199-208.

Kolmark G, Giles NH.   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, Higashiguchi
     R. 1980.  Forestomach tumors induced by orally  administered epichloro-
     hydrin in male Wistar rats.  Gann 71:922-923.

Kotin P,  Falk HL.   1963.  Organic peroxides, hydrogen peroxide, epoxides,
     and  neoplasia.  Radiat.  Res. Supp. 3:193-211.

Kremneva  SN, Tolgskaya  MS.  1961.  Toxicology of epichlorohydrin.  Toksikol.
     Nov.  Prom. Khim.  Veschestv.  2:28-41.

Kremneva  SN.   1960.   Investigation  of toxicity  of epichlorohydrin and
     dichlorohydrin.   Pr.  Toksikol.  Moscow  Sb.  123-131.
                                  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 in vitro in comparison
     with TEPA.   Mutat. Res.   34:271-278.

Kucerova M,  Zhurkov  VS,  Polivkova Z, Ivanova JE.  1977.  Mutagenic effect
     of epichlorohydrin.  II.  Analysis of chromosomal aberrations in lympho-
     cytes of persons  occupationally exposed to epichlorohydrin.  Mutat.
     Res.   48:355-360.

Kwart H, Goodman AL.   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 JL.  1978.  Cutaneous
     allergy to epichlorohydrin.   Ann. Dermatol. Venereol.   105(5):521-525.
     (In French;  summary in English)

Laskin S,  Sellakumar AR,  Kuschner M, Nelson N, La Mendola S, Rusch GM, Katz
     GV, Dulak NC, Albert  RE.  1980.  Inhalation carcinogenicity  of epi-
     chlorohydrin in noninbred Sprague-Dawley rats.   J. Natl.  Cancer Inst.
     65(4):751-755.

Laumbach AD,  Lee S,  Wong J, Streips UN.   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(1):155-170.

Lawrence WH, Autian  J.   1972.  Possible toxic  effects from inhalation of
     dental  ingredients by alteration of drug biologic half-life.  J. Dent.
     Res.   51(3):878.

Lawrence WH,  Malik M, Autian J.   1974.  Development of a toxicity evaluation
     program for dental materials and products.  II. Screening  for  systemic
     toxicity.   J.  Biomed,  Mater.  Res.  8:11-34.

Lawrence WH, Malik  M,  Turner JE, Autian  J.   1972.   Toxicity  profile of
     epichlorohydrin.  J.  Pharm.  Sci. 61(11):1712-1717.

le Noble WJ,  Duffy M.  1964.   The effect of pressure on the rates of  acid-catalyzed
     hydrolysis of trimethylene  oxide and of epichlorohydrin.   J.     Phys.
     Chem.  68(3):619-622.

Long FA, Pritchard JG.   1956.  Hydrolysis of substituted ethylene oxides in
     HpO solutions.   J. Am. Chem.  Soc.  78(12):2663-2667.

Lowenheim FA, Moran MK.  1975.  Faith, Keyes, and Clark's Industrial  Chemicals,
     4th ed.   New York, NY:  John Wiley & Sons, pp. 335-338.

Lu P-Y, Metcalf  RL.   1975.   Environmental  fate and  biodegradability of
     benzene derivatives as studied  in a model aquatic  ecosystem.  Environ.
     Health Perspect. 10:269-284.


                                  R-8

-------
Lukaneva AM,  Rodionov  GA.   1978.   Study of the  influence  of polychloro-
     camphene  and epichlorohydrin  on the development  of experimental
     atherosclerosis.  Fiziol.  (Kiev) 24(3):371-377.

Mackey BE, MacGregor JT.  1979.   The micronucleus test:   Statistical design
     and analysis.  Mutation res.   64:195-204.

Mann H.   1976.   Fischtest  mit  Goldorfen  zur vergleichenden Prufung der
     akuten  Toxizitat  von Wasserinhaltsstoffen  und  Abwassern-Praktische
     Erfahrungen aus drei Ringtesten.  Z.  Wasser Abwasser Forsch.  9:103-109.
     (In German, English abstract)

Mantel N, and Schneiderman MA.   1975.  Estimating "safe" levels, a hazardous
     undertaking.  Cancer Res.  35:1379-1386.

Marks TA, Gerling FS,  Staples  RE.   1982.  Teratogenic  evaluation  of epi-
     chlorohydrin  in the mouse  and rat and glycidol in  the mouse.  J.
     Toxicol. Environ.  Health 9:87-96.

Mashford M, Jones AR.  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):II-463-II-467.   (In
     French; summary in English)

Milby TH, Whorton D.  1980.  Epidemiological assessment of occupationally
     related  chemically  induced sperm count suppression.  J.  Occup. Med.
     22(2):77-82.

Milby TH, Whorton MD,  Stubbs HA, Ross CE, Joyner RE, Lipshultz  LI.  1981.
     Testicular  function  among  epichlorohydrin  workers.  Br. J. Ind. Med.
     38(4):372-377.
                                              +/-
Moore-Brown  MM,  Clive D.  1979.  The  L5178Y/TK '  mutagen assay  system:  In
     situ results.   Banbury Report 2:71-88.

NIOSH.   1976a.   National Institute  for Occupational Safety and Health.
     Criteria  for a  recommended standard	Occupational exposure to epi-
     chlorohydrin.   Washington,  DC:   Department of Health, Education, and
     Welfare.     DHEW Pub. NIOSH 76-206.

NIOSH.  1976b.   National  Institute for Occupational Safety and  Health.  A
     recommended standard for  occupational  exposure to epichlorohydrin.
     Washington,  DC:   Department  of Health,  Education, and  Welfare, 7pp.

NIOSH.   1978.   National  Institute  for Occupational  Safety  and Health.
     Current  Intelligence Bulletin 30:  Epichlorohydrin.   DHEW Publication
     No. 79-105.

Neeley WB,  Branson  DR, Blau GE.   1974. Partition coefficient to measure
     bioconcentration  potential  of  organic  chemicals  in fish.  Environ.
     Sci  Technol.  8(13):1113-1115.
                                  R-9

-------
 Norppa  H,  Hemminki  K,  Sorsa M,  Vainio H.   1981.   Effect of monosubstituted
     epoxides on chromosome aberrations and SCE in cultured human lymphocytes.
     Mutat. Res.  91:243-250.

 Pallade  S, Oorobantu  M,  Rotaru G,  Gabrielescu  E.   1967.  Experimental
     studies of intoxication by epichlorohydrin.   Arch. Mai. Prof. 28:505-516.
     (In French)

 Perry P, Wolff  S.   1974.   New Giemsa method for the differential staining
     of sister chromatids.  Nature 251:156-158.

 Petty WL,  Nichols PL 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 MK,  Lederer  TS,  Murray JS, Deacon MS, Hanley TR, Quast JF, and John
     JA.   1979.  Epichlorohydrin-subchronic  studies.  IV.  The effects  of
     maternally inhaled  epichlorohydrin on rat and  rabbit embryonal and
     fetal development.  Unpublished report Toxicology Research Laboratory,
     Health and   Environmental  Sciencies, Dow Chemical  U.S.A., Midland,
     MI.

 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  1ikely
     metabolite, as an alkylating agent.  Chem. Biol.  Interact. 25:157-175.

 Pritchard JG, Long FA.   1956.  Kinetics and mechanisms of the acid-catalyzed
     hydrolysis of substituted ethylene oxides.  J. Am. Chem. Soc. 78:2667-2670.

 Pritchard  JG, Siddiqui  IA.   1973.  Activation parameters  and  mechanism of
     the   acid-catalyzed  hydrolysis  of epoxides.   J. Chem.  Soc. Perkin II.
     (4)452-457.

 Quast JF, Henck JW, Postma BJ, Schuetz DJ, McKenna MJ.   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.  166 pp.

Quast JF,  Lederer  TS,  Postma BJ,  Scheutz  DJ,  John JA, McKenna MJ.  1979b.
     t|"chlorohydrin - 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.  57pp.

 Rapoport IA. 1948.   The effect of ethylene  oxide,  glycide and glycol  on
     genetic mutations.   Doklady  Akademii  Nauk SSR 60;469-472  (English
     translation)

 Riesser GH.   1078.   Fpichlorohydrin.   In:  Kirk-Othmer  Encyclopedia  of
     Chemical Technology.   New York:  John Wiley & Sons  5:858-864.


                                  R-10

-------
Ross WCJ.  1950.   The  reactions of certain epoxides in aqueous solutions.
     J. Chem. Soc. (3):2257-2272.

Rotaru G,  Pal lade S.   1966.   Experimental  studies of histopathological
     features in  acute epichlorohydrin  toxicity.   Morfol. Norm.  Patol.
     11:155-164.

Rutledge JE, Islam MN.   1973.   Canning and pH stability of epichlorohydrin-
     treated parboiled rice.  J. Agric.  Food chem.   21(3):458-460.

SRI.   1981.   Stanford  Research Institute, Menlo Park,  CA.   Directory of
     Chemical Producers.

Sax NI,  ed.  1975.   Dangerous  properties of  industrial materials,  4th ed.
     New York, NY: Van Nostrand Reinhold Co., p. 717.

Schultz VC.  1964.  Fettleber und chronisch-asthmoide Bronchitis nach  Inhalation
     eines  Farbenlosungsmittels (Epichlorohydrin).   Dtsch.  Med.  Wschr.
     89:1342-1344.  (In German)

Shell  Chemical  Company.   1969.  Epichlorohydrin. Technical Bulletin No.  1C
     69-16.  Shell Oil Co.,  Industrial Chemicals Division, 51 pp.

Shellenberger RJ, McClimans CD, Ott MG, Flake RE, Daniel  RL.   January 1979.
     An  evaluation  of  the mortality  experience  of  employees  with potential
     for  exposure to  epichlorohydrin.   Submitted  to U.S. Department of
     Labor,  OSHA Docket No. H-100. Unpublished.

Shell  Oil  Company.   1978.  Testicular function among ECH workers  at  Shell
     Deer  Park and Norco  facilities.  Shell  Oil Co. 22 pp.

Sherman  S, Gannon J, Buchi  G, Howell WR. 1980.  Epoxy resins. In:  Kirk-Othmer
     Encyclopedia  of Chemical  Technology,  3rd ed.   New York:   John Wiley &
     Sons  9:267-290.

Shumskaya  NI, Karamzina NM. 1966.  An evaluation of the functional  state  of
     rat  kidneys  during poisoning  by industrial  substances.  Toksikol. Nov.
     Prom. Khun. Veshchestv.  8:14-27.

Shumskaya  NI, Karamzina NM, Savina MY. 1971.  Evaluating  the sensitivity  of
     integral and specific  parameters in acute epichlorohydrin  intoxication.
     Toksikol. Nov. Prom. Khim. Veshchestv.   12:33-44.

Shvets VF, Aleksanyan  LV.  1973.  Kinetics  of the  hydrolysis of 1-chloro-
     2,3-epoxy  propane in aqueous  carbon dioxide salt solutions.   Zh. Org.
     Khim. 10(10):2027-2030.

Smith  FA,  Langvardt PW, Young JD.   1979.   Pharmacokinetics  of epichloro-
     hydrin  (EPI)  administered to rats by gavage or  inhalation.   Midland,
     MI: Dow Chemical  USA,  Toxicology Research Laboratory, 52pp.

Smvth  HF  Jr.,  Carpenter  CP,  Weil  CS,  Pozzani  UC, Streigel JA.   1962.
     Range-finding toxicity data, List VI.  Ind. Hyg. J. 23:95-96.
                                  R-ll

-------
 Smyth  HF Jr. ,  Carpenter CP.  1948.   Further  experience with the range
     finding test in the  industrial  toxicology laboratory.  J.  Ind.  Hyg.
     Toxicol. 30(1):63-68.

 Sora S,  Zelaschi  D, Ronzio E.  1979.  Activity  of epichlorohydrin on different
     genetic systems  of Saccharomyces cerevisiae.  Atti Assoc.   Genetica
     Italiana 24:274-276.

 Sram RJ, Cerna  M, Kucerova  M.  1976.  The genetic risk  of epichlorohydrin
     as  related to the occupational  exposure.  Biol. Zbl.  95(4):451-462.

 Sram RJ, Tomatis L,  Clemmesen J, Bridges BA.   1981.  An evaluation of the
     genetic toxicity of epichlorohydrin. Mutat. Res.  87:299-319.

 Sram RJ, Zudova Z, Kuleshov  NP.   1980.   Cytogenetic  analysis  of  peripheral
     lymphocytes  in  workers  occupationally  exposed to  epichlorohydrin.
     Mutat.  Res. 70:115-120.

 Stolzenberg  SJ, Hine CH.  1979.  Mutagenicity of halogenated and oxygenated
     three-carbon  compounds.   J.  Toxicol.  Environ.  Health 5:1149-1158.

 Suskov II, Sazonova LA.  1982.  Cytogenetic effects of epoxy, phenolformal-
     dehyde  and polyvinylchloride resins in  man.  Mutat.  Res.  104:137-140.

 Swan JO.   1954.  Determination of  epoxides  with sodium sulfite.  Anal.
     Chem.  26(5):878-880.

 Thorgeirsson A,  Fregert S.   1977.  Allergenicity of epoxy  resins in the
     guinea pig.  Acta Dermatovener (Stockholm) 57:253:256.

 Thurman  GB,  Simms BG,  Goldstein  AL, Kilian DJ.  1978.   The  effects of
     organic compounds  used  in the  manufacture of plastics on the respon-
     sivity  of  murine and human  lymphocytes.   Toxicol.  Appl. Pharmacol.
     44:617-641.

 Tsuchimoto T, Matter BE.   1981.   Activity of coded  compounds  in  the  micro-
     nucleus test.  In:  de Serres FJ, Ashby J, eds.  Evaluation of Short-Term
     Tests for Carcinogens.  Elsevier/North Holland, pp. 705-711.

 U.S. EPA.  1974.   U.S.  Environmental Protection Agency.  Office of  Water
     and Hazardous Materials.  Addendum  to development  document  for pro-
     posed effluent  limitations  guidelines  and new  source performance
     standards  for the  synthetic  resins.  Washington, DC:  U.S.  Environ-
     mental Protection Agency, pp. 17-24.  EPA 440/1-74/036-a.

U.S. EPA.  1975.  U.S.  Environmental  Protection Agency.  Office of Research
     and Development.   Methods for  acute toxicity tests with fish, macro-
     invertebrates, and amphibians.   Corvallis,  OR:   U.S.  Environmental
     Protection Agency, 61 pp.  EPA 660/3-75-009.

U.S. EPA.  1978.  U.S.  Environmental  Protection Agency.  Region III.  Field
     Investigation, Emergency Response Action, and Hydrogeological Assistance
     at Chessie System Derailment and Spill of  Epichlorohydrin, Point Pleasant,
     WV, January 23-39, 1978.  Wheeling  Field Office.


                                  R-12

-------
l.S. EPA.  1980.  U.S. Environmental  Protection Agency.   Office of Research
     and Development.  Fate  of  toxic and hazardous materials  in  the air
     environment.  Research  Triangle  Park:   U.S.  Environmental Protection
     Agency, 29 pp.   EPA 600/53-80-084.

U.S. EPA.   1983.  U.S.  Environmental Protection Agency.   TSCA - Plant and
     Production (TSCAPP).  CICIS - Chemical Information Tape.  EPA/DF-80/005.

Van Duuren  BL,  Katz  C,  Goldschmidt BM.   1972a.  Direct-acting alkylating
     carcinogens --  Chloro  ethers and related compounds.  Toxicol.  Appl.
     Pharmacol.  22:279-280.

Van Duuren  BL,  Katz  C,  Goldschmidt  BM,  Frenkel  K,  Livak A.  1972b.
     Carcinogenicity  of  halo-ethers.  II.   Structure-activity relation-
     ships  of  analogs of bis(chloromethyl)ether.  J.  Natl.  Cancer  Inst.
     48:1431-1439.

Van Duuren  BL,  Goldschmidt  BM,  Katz  C, Seidman  I, Paul JS.   1974.  Carcino-
     genic activity of alkylating agents. J. Natl.  Cancer Inst. 53:695-700.

van Lierop  JBH.   1978.   Simple and rapid determination of epichlorohydrin
     at  the  lower parts  per billion  level  by gas chromatography-mass frag-
     mentography.  J. Chromatogr.  166(2):609-610.

Vashishat RK,  Vasudeva  M,  Kakar  SN.  1980.  Induction of mitotic  crossing
     over, roitotic gene  conversion and reverse mutation by epichlorhydrin
     in  Saccharomyces cerevisiae.   Indian  J.  Exp. Biol.  18:1337-1338.

Veith  GO,  Macek KJ,   Petrocelli  SR,  Carroll  J.   1980.  An evaluation of
     using   partition  coefficients  and  water solubility to estimate bio-
     concentration factors  for  organic  chemicals in  fish.  In:  Eaton JG,
     Parrish  PR, Hendricks  AC,  eds.  Aquatic Toxicology, ASTM STP  707.
     American Society for Testing and Materials, pp.  116-129.

Venable  JR, McClimans CD, Flake RE, Dimick DB.  1980.  A fertility study of
     male  employees  engaged in the  manufacture  of  glycerine.  J. Occup.
     Med.    22(2):87-91.

yerschueren K.   1977.  Handbook of Environmental Data on Organic Chemicals.
     New York,  NY:  Van  Nostrand  Reinhold Co., pp. 293-294.

Vogel  E, Schalet A,  Lee WR, Wurgler F.   1981.   Mutagenicity  of selected
     chemicals  in Drosophila.    In:  de Serres  FJ  (ed.) Comparative
     Mutagenesis.

Voogd  CE.  1973.  Mutagenic  action of epoxy compounds  and several  alcohols.
     Mutat. Res.  21:52-53.

Vooqd  CE  van  der Stet  JJ,  Jacobs  JJJAA.   1981.   The mutagenic action of
     aliphatic  epoxides.  Mutat.  Res. 89:269-282.

Weast  RC»  ed.   1978.   CRC  Handbook of Chemistry and Physics, 59th ed. West
     Palm Beach,  FL:  CRC Press  Inc., p.  C-447.
                                  R-13

-------
Weigle WW, Plotnick HB, Conner WL.  1978. Tissue distribution and excretion
         14
     of   C-epichlorohydrin  in  male and female  rats.   Res.  Commun.  Chem.
     Pathol. Fharmacol.  20(2):275-288.

Weil CS,  Condra N, Hann C, and Streigel JA.  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
     industrial  facility  for the manufacturing  of epoxy resins.  Mater.
     Plant.  (Bucharest) 8:322-323.

White AD.  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
     chemicals in the work environment.  Ambio 4(1):34-36.

Wurgler FE,  Graf U. 1981.   Mutagenic activity of ten coded compounds  in the
     Drosophila  sex-linked  recessive   lethal assay.   In:  de Serres  FJ,
     Ashby J, eds.  Evaluation of Short-Term Tests for Carcinogens.   Elsevier/
     North Holland, pp.  666-672.
                                  R-14

-------
Appendix A.  Evaporation Rate of Epichlorohydrin Calculated According
                    to the Method of Oilling (1977)
Equations:

     1.   H = (16.04)(P)(M)/(T)(S)

     2.   Kx = 221.1/[(1.042/H] + 100](M)0'5

     3.   Half-life (days) = (0.6931/^X3/1,440)
where:
     H    =  Henry's law constant
     P    =  vapor pressure in mmHg at 20° C
     M    =  gram molecular weight of the solute
     T    =  temperature in °K (20° C = 293° K)
     S    =  solubility of the solute in water in mg/1 (ppm) at 20° C
     K..   =  overall liquid exchange constant in cm/min
     d    =  solution depth in cm
Calculations:
     H    =  (16.04)(12)(92.53)/(293)(60,000) = 1.01 x 10"3

          =  221.1/[(1.042/0.00101) + 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 (K  ) and Soil  Organic
                    Matter/Water Partition Coefficient (Q)
Equations
     1-   log KQC = 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               KQC
       4.78              0.26          10.28 x 10*1'23          5.68
       4.82              0.26           9.76 x 10*1'23          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  H-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-constant
value was used together  with its "uncertainty units."  The  n-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 10s Mmole/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 extrapola-
tion doses and  is consistent with the data at all dose levels of the experi-
ment.   It  is a  nonthreshold model holding that the upper-limit of risk pre-
dicted by  a linear extrapolation to  low levels of the dose-response  relation-
ship is the most plausible upper-limit for the risk.
     Other models have also been used for risk extrapolation.  Three non-
threshold models  are  presented here:  the one-hit, the  log-probit,  and the
Weibull.  The one-hit  model is characterized by a continuous downward curva-
ture 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 which are  at  least as large  as those of  the multistage model.
Further, whenever the  data  can be fit adequately by the one-hit model, esti-
mates 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 the general "S"
curvature.  The  low-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 the problems of
biological assay  such  as  the assessment of potency of  toxicants and  drugs  and
has usually been  used to estimate such values  as percentile lethal dose or
percentile effective  dose.   Its development was strictly  empirical, i.e., it
was 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,c) = c +  (1-c) 4» (a+blog.^ D)  a,b > 0 < c < 1

where P is the proportion responding at dose D, c is an  estimate  of  the back-
ground rate, a  is an estimate of  the standardized mean  of individual toler-
ances, and b is an estimate of the log dose-probit response slope.

                                      E-l

-------
     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
Weibull  model.  It is of the form
               P(D;b,k) = l-exp-(bdk)  b.k > 0
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  from the
log-Probit.
     The estimates of added risk for low doses  for the above models are given
below for the  ECH 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 incor-
porate Abbott's correction for independent background rate.
     The results  (Table E-l)  show  that in  order of  descending  risk the
one-hit > multistage  >  Weibull  > log-probit.   The best fit of the data with
the multistage  model  is a cubic with zero linear component, which accounts for
its non-linear  behavior at low doses.
                                      E-2

-------
                                      TABLE E-l.   ESTIMATES OF EPICHLOROHYDRIN LOW-DOSE RISK IN HALE WISTAR RATS

                                                         DERIVED FROM FOUR DIFFERENT MODELS
Dose
H9/1
Males

0.1

1
10
100
1
Maximum Likelihood Estimates of
Additional Risks
Multistage One-Hit Wiebull Log-Probit
Model Model Model Model


M/l

MB/1
MO/1
MB/1
«g/l





1
2
2


0

0
.4X10'17
.6X10'14
.6X10-"

-8
2. 1x10
-7
2.1x10
2. Ixlfl"6
2.1x!0"5
2. Ixio"4


0

0
0
l.lxio"14
1.3xlo"U


0

0
0
0
0
95% Upper Confidence Limit of
Additional Risks
Multistage One-Hit Weibull Log-Probit
Model Model Model Model


2.

2.
2.
2.
2.

-8
8x10
.7
8x10 '
8x!0"6
8xlO~5
8x!0"4

-O
3.4x10 °
.7
3.4x10
3.4xlO~6
3.4x!0"5
3.4xlO~4


0

0
0
2.1X10"13
2.0X10'10


0

0
0
0
0
             Source:  Konishi et al. (1980)
m
 i
CO

-------