UNITED NATIONS ENVIRONMENT PROGRAMME
INTERNATIONAL LABOUR ORGANISATION
WORLD HEALTH ORGANIZATION
INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 215
Vinyl Chloride
This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of the United Nations Environment Programme, the International
Labour Organisation, or the World Health Organization.
Published under the joint sponsorship of the United Nations
Environment Programme, the International Labour Organisation, and the
World Health Organization, and produced within the framework of the
Inter-Organization Programme for the Sound Management of Chemicals.
World Health Organization
Geneva, 1999
The International Programme on Chemical Safety (IPCS),
established in 1980, is a joint venture of the United Nations
Environment Programme (UNEP), the International Labour Organisation
(ILO), and the World Health Organization (WHO). The overall objectives
of the IPCS are to establish the scientific basis for assessment of
the risk to human health and the environment from exposure to
chemicals, through international peer-review processes, as a
prerequisite for the promotion of chemical safety, and to provide
technical assistance in strengthening national capacities for the
sound management of chemicals.
The Inter-Organization Programme for the Sound Management of
Chemicals (IOMC) was established in 1995 by UNEP, ILO, the Food and
Agriculture Organization of the United Nations, WHO, the United
Nations Industrial Development Organization and the Organisation for
Economic Co-operation and Development (Participating Organizations),
following recommendations made by the 1992 UN Conference on
Environment and Development to strengthen cooperation and increase
coordination in the field of chemical safety. The purpose of the IOMC
is to promote coordination of the policies and activities pursued by
the Participating Organizations, jointly or separately, to achieve the
sound management of chemicals in relation to human health and the
environment.
WHO Library Cataloguing in Publication Data
Vinyl chloride.
(Environmental health criteria ; 215)
1.Vinyl chloride - analysis 2.Vinyl chloride - toxicity
3.Vinyl chloride - adverse effects 4.Environmental exposure
5.Occupational exposure I.International Programme on
Chemical Safety II.Series
ISBN 92 4 157215 9 (NLM Classification: QV 633)
ISSN 0250-863X
The World Health Organization welcomes requests for permission to
reproduce or translate its publications, in part or in full.
Applications and enquiries should be addressed to the Office of
Publications, World Health Organization, Geneva, Switzerland, which
will be glad to provide the latest information on any changes made to
the text, plans for new editions, and reprints and translations
already available.
(c) World Health Organization 1999
Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. All rights reserved.
The designations employed and the presentation of the material in
this publication do not imply the expression of any opinion whatsoever
on the part of the Secretariat of the World Health Organization
concerning the legal status of any country, territory, city or area or
of its authorities, or concerning the delimitation of its frontiers or
boundaries.
The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar nature
that are not mentioned. Errors and omissions excepted, the names of
proprietary products are distinguished by initial capital letters.
CONTENTS
ENVIRONMENTAL HEALTH CRITERIA FOR VINYL CHLORIDE
1. SUMMARY
1.1. Identity, physical and chemical properties, and
analytical methods
1.2. Sources of human and environmental exposure
1.3. Environmental transport, distribution and transformation
1.4. Environmental levels and human exposure
1.5. Kinetics and metabolism in laboratory animals and
humans
1.6. Effects on laboratory mammals and in vitro test
systems
1.7. Effects on humans
1.8. Effects on other organisms in the laboratory and
field
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, AND ANALYTICAL
METHODS
2.1. Identity
2.2. Physical and chemical properties
2.3. Conversion factors
2.4. Analytical methods
2.4.1. General analytical methods and
detection
2.4.2. Sample preparation, extraction and analysis
for different matrices
2.4.2.1 Air
2.4.2.2 Water
2.4.2.3 PVC resins and PVC products
2.4.2.4 Food, liquid drug and cosmetic
products
2.4.2.5 Biological samples
2.4.2.6 Human monitoring
2.4.2.7 Workplace air monitoring
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
3.1. Natural occurrence
3.2. Anthropogenic sources
3.2.1. Production levels and processes
3.2.1.1 Production of VC
3.2.1.2 Production of PVC from VC
3.2.1.3 PVC products
3.2.2. Emissions from VC/PVC plants
3.2.2.1 Sources of emission during the
production of VC
3.2.2.2 Emission of VC and dioxins from VC/PVC
plants during production
3.2.3. Accidental releases of VC
3.2.3.1 PVC plant and transport accidents
3.2.3.2 Leakage and discharge from VC/PVC
plants
3.2.4. VC residues in virgin PVC resin and products
3.2.4.1 VC residues in different PVC samples
3.2.4.2 VC residues in PVC products
3.2.4.3 VC formation as a result of heating PVC
3.2.5. Other sources of VC
3.2.5.1 VC as a degradation product of
chlorinated hydrocarbons
3.2.5.2 VC formation from tobacco
3.3. Uses
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION AND TRANSFORMATION
4.1. Transport and distribution between media
4.1.1. Air
4.1.2. Water and sediments
4.1.3. Soil and sewage sludge
4.1.4. Biota
4.2. Transformation
4.2.1. Microbial degradation
4.2.2. Abiotic degradation
4.2.2.1 Photodegradation
4.2.2.2 Hydrolysis
4.2.3. Other interactions
4.3. Bioaccumulation
4.4. Ultimate fate following use
4.4.1. Waste disposal
4.4.2. Fate of VC processed to PVC
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
5.1. Environmental levels
5.1.1. Air
5.1.1.1 Outdoor air
5.1.1.2 Indoor air
5.1.2. Water and sediment
5.1.3. Soil and sewage sludge
5.1.3.1 Soil
5.1.3.2 Sewage sludge
5.1.4. Food, feed and other products
5.1.5. Terrestrial and aquatic organisms
5.2. General population exposure
5.2.1. Estimations
5.2.2. Monitoring data of human tissues
or fluids
5.3. Occupational exposure
6. KINETICS AND METABOLISM IN LABORATORY ANIMALS AND HUMANS
6.1. Absorption
6.1.1. Oral exposure
6.1.2. Inhalation exposure
6.1.3. Dermal exposure
6.2. Distribution and retention
6.2.1. Oral exposure
6.2.2. Inhalation exposure
6.2.3. Partition coefficients in vitro
6.3. Metabolic transformation
6.4. Elimination and excretion
6.4.1. Oral exposure
6.4.2. Inhalation exposure
6.5. Reaction with body components
6.5.1. Formation of DNA adducts
6.5.2. Alkylation of proteins
6.6. Modelling of pharmacokinetic data for
vinyl chloride
7. EFFECTS ON LABORATORY MAMMALS AND IN VITRO TEST SYSTEMS
7.1. Acute toxicity
7.2. Short-term toxicity
7.2.1. Oral exposure
7.2.2. Inhalation exposure
7.2.3. Dermal exposure
7.3. Long-term toxicity - effects other than tumours
7.3.1. Oral exposure
7.3.2. Inhalation exposure
7.4. Skin and eye irritation; sensitization
7.5. Reproductive toxicity, embryotoxicity and
teratogenicity
7.5.1. Male reproductive toxicity
7.5.2. Embryotoxicity and teratogenicity
7.6. Special studies
7.6.1. Neurotoxicity
7.6.2. Immunotoxicity
7.6.3. Cardiovascular effects
7.6.4. Hepatotoxicity
7.7. Carcinogenicity
7.7.1. Oral exposure
7.7.2. Inhalation exposure
7.7.2.1 Short-term exposure
7.7.2.2 Long-term exposure
7.7.3. The effect of age on susceptibility to
tumour induction
7.7.4. The effect of gender on susceptibility to
tumour induction
7.7.5. Carcinogenicity of metabolites
7.8. Genotoxicity
7.8.1. In vitro studies
7.8.2. In vivo studies
7.8.3. Genotoxicity of VC metabolites
7.8.4. Other toxic effects of VC metabolites
7.8.5. Mutagenic and promutagenic properties of DNA
adducts formed by VC metabolites
7.8.6. Mutations in VC-induced tumours
7.9. Factors modifying toxicity
7.10. Mechanisms of toxicity - mode of action
7.10.1. Mechanisms of VC disease
7.10.2. Mechanism of carcinogenesis
8. EFFECTS ON HUMANS
8.1. General population
8.2. Controlled human studies
8.3. Occupational exposure
8.3.1. Overview
8.3.2. Non-neoplastic effects
8.3.2.1 Acute toxicity
8.3.2.2 Effects of short- and long-term
exposure
8.3.2.3 Organ effects
8.3.3. Neoplastic effects
8.3.3.1 Liver and biliary tract cancers
8.3.3.2 Brain and central nervous
system (CNS)
8.3.3.3 Respiratory tract
8.3.3.4 Lymphatic and haematopoietic
cancers
8.3.3.5 Malignant melanoma
8.3.3.6 Breast cancer
8.3.3.7 Other cancer sites
8.4. Genotoxicity studies
8.4.1. Cytogenetic studies of VC-exposed
workers
8.4.2. Mutations at the hypoxanthine guanine
phosphoribosyltransferase (hprt) locus
8.4.3. Mutations in ASL from VC-exposed
workers
8.4.3.1 p53 gene
8.4.3.2 ras genes
8.5. Studies on biological markers
8.5.1. Excretion of metabolites
8.5.2. Genetic assays
8.5.3. Enzyme studies
8.5.4. von Willebrand factor
8.5.5. p53 and ras proteins
8.6. Susceptible subpopulations
8.6.1. Age susceptibility
8.6.2. Immunological susceptibility
8.6.3. Polymorphic genes in VC metabolism
9. EFFECTS ON OTHER ORGANISMS IN THE LABORATORY AND FIELD
9.1. Laboratory experiments
9.1.1. Microorganisms
9.1.1.1 Water
9.1.1.2 Soil
9.1.2. Aquatic organisms
9.1.2.1 Invertebrates
9.1.2.2 Vertebrates
9.2. Field observations
9.2.1. Aquatic organisms
10. EVALUATION OF HUMAN HEALTH RISKS AND EFFECTS ON THE
ENVIRONMENT
10.1. Evaluation of human health effects
10.1.1. Hazard identification
10.1.1.1 Non-neoplastic effects
10.1.1.2 Neoplastic effects
10.1.2. Dose-response analysis
10.1.2.1 Non-neoplastic effects
10.1.2.2 Neoplastic effects
10.1.3. Human exposure
10.1.3.1 General population
10.1.3.2 Occupational exposure
10.1.4. Risk characterization
10.2. Evaluation of effects on the environment
11. RECOMMENDATIONS FOR PROTECTION OF HUMAN HEALTH
11.1. Public health
11.2. Occupational health
12. FURTHER RESEARCH
13. PREVIOUS EVALUATIONS BY INTERNATIONAL BODIES
REFERENCES
ANNEX 1. REGULATIONS CONCERNING VINYL CHLORIDE
ANNEX 2. PHYSIOLOGICAL MODELLING AND RECENT RISK ASSESSMENTS
ANNEX 3. EXECUTIVE SUMMARY OF VINYL CHLORIDE PANEL REPORT
RESUME
RESUMEN
NOTE TO READERS OF THE CRITERIA MONOGRAPHS
Every effort has been made to present information in the criteria
monographs as accurately as possible without unduly delaying their
publication. In the interest of all users of the Environmental Health
Criteria monographs, readers are requested to communicate any errors
that may have occurred to the Director of the International Programme
on Chemical Safety, World Health Organization, Geneva, Switzerland, in
order that they may be included in corrigenda.
* * *
A detailed data profile and a legal file can be obtained from the
International Register of Potentially Toxic Chemicals, Case postale
356, 1219 Chātelaine, Geneva, Switzerland (telephone no. + 41 22 -
9799111, fax no. + 41 22 - 7973460, E-mail irptc@unep.ch).
Environmental Health Criteria
PREAMBLE
Objectives
In 1973 the WHO Environmental Health Criteria Programme was
initiated with the following objectives:
(i) to assess information on the relationship between exposure to
environmental pollutants and human health, and to provide
guidelines for setting exposure limits;
(ii) to identify new or potential pollutants;
(iii) to identify gaps in knowledge concerning the health effects of
pollutants;
(iv) to promote the harmonization of toxicological and
epidemiological methods in order to have internationally
comparable results.
The first Environmental Health Criteria (EHC) monograph, on
mercury, was published in 1976 and since that time an ever-increasing
number of assessments of chemicals and of physical effects have been
produced. In addition, many EHC monographs have been devoted to
evaluating toxicological methodology, e.g. for genetic, neurotoxic,
teratogenic and nephrotoxic effects. Other publications have been
concerned with epidemiological guidelines, evaluation of short-term
tests for carcinogens, biomarkers, effects on the elderly and so
forth.
Since its inauguration the EHC Programme has widened its scope,
and the importance of environmental effects, in addition to health
effects, has been increasingly emphasized in the total evaluation of
chemicals.
The original impetus for the Programme came from World Health
Assembly resolutions and the recommendations of the 1972 UN Conference
on the Human Environment. Subsequently the work became an integral
part of the International Programme on Chemical Safety (IPCS), a
cooperative programme of UNEP, ILO and WHO. In this manner, with the
strong support of the new partners, the importance of occupational
health and environmental effects was fully recognized. The EHC
monographs have become widely established, used and recognized
throughout the world.
The recommendations of the 1992 UN Conference on Environment and
Development and the subsequent establishment of the Intergovernmental
Forum on Chemical Safety with the priorities for action in the six
programme areas of Chapter 19, Agenda 21, all lend further weight to
the need for EHC assessments of the risks of chemicals.
Scope
The criteria monographs are intended to provide critical reviews
on the effect on human health and the environment of chemicals and of
combinations of chemicals and physical and biological agents. As such,
they include and review studies that are of direct relevance for the
evaluation. However, they do not describe every study carried out.
Worldwide data are used and are quoted from original studies, not from
abstracts or reviews. Both published and unpublished reports are
considered and it is incumbent on the authors to assess all the
articles cited in the references. Preference is always given to
published data. Unpublished data are used only when relevant published
data are absent or when they are pivotal to the risk assessment. A
detailed policy statement is available that describes the procedures
used for unpublished proprietary data so that this information can be
used in the evaluation without compromising its confidential nature
(WHO (1990) Revised Guidelines for the Preparation of Environmental
Health Criteria Monographs. PCS/90.69, Geneva, World Health
Organization).
In the evaluation of human health risks, sound human data,
whenever available, are preferred to animal data. Animal and in vitro
studies provide support and are used mainly to supply evidence
missing from human studies. It is mandatory that research on human
subjects is conducted in full accord with ethical principles,
including the provisions of the Helsinki Declaration.
The EHC monographs are intended to assist national and
international authorities in making risk assessments and subsequent
risk management decisions. They represent a thorough evaluation of
risks and are not, in any sense, recommendations for regulation or
standard setting. These latter are the exclusive purview of national
and regional governments.
Content
The layout of EHC monographs for chemicals is outlined below.
* Summary - a review of the salient facts and the risk evaluation
of the chemical
* Identity - physical and chemical properties, analytical methods
* Sources of exposure
* Environmental transport, distribution and transformation
* Environmental levels and human exposure
* Kinetics and metabolism in laboratory animals and humans
* Effects on laboratory mammals and in vitro test systems
* Effects on humans
* Effects on other organisms in the laboratory and field
* Evaluation of human health risks and effects on the environment
* Conclusions and recommendations for protection of human health
and the environment
* Further research
* Previous evaluations by international bodies, e.g. IARC, JECFA,
JMPR
Selection of chemicals
Since the inception of the EHC Programme, the IPCS has organized
meetings of scientists to establish lists of priority chemicals for
subsequent evaluation. Such meetings have been held in Ispra, Italy,
1980; Oxford, United Kingdom, 1984; Berlin, Germany, 1987; and North
Carolina, USA, 1995. The selection of chemicals has been based on the
following criteria: the existence of scientific evidence that the
substance presents a hazard to human health and/or the environment;
the possible use, persistence, accumulation or degradation of the
substance shows that there may be significant human or environmental
exposure; the size and nature of populations at risk (both human and
other species) and risks for environment; international concern, i.e.
the substance is of major interest to several countries; adequate data
on the hazards are available.
If an EHC monograph is proposed for a chemical not on the
priority list, the IPCS Secretariat consults with the Cooperating
Organizations and all the Participating Institutions before embarking
on the preparation of the monograph.
Procedures
The order of procedures that result in the publication of an EHC
monograph is shown in the flow chart on p. xv. A designated staff
member of IPCS, responsible for the scientific quality of the
document, serves as Responsible Officer (RO). The IPCS Editor is
responsible for layout and language. The first draft, prepared by
consultants or, more usually, staff from an IPCS Participating
Institution, is based initially on data provided from the
International Register of Potentially Toxic Chemicals, and reference
data bases such as Medline and Toxline.
The draft document, when received by the RO, may require an
initial review by a small panel of experts to determine its scientific
quality and objectivity. Once the RO finds the document acceptable as
a first draft, it is distributed, in its unedited form, to well over
150 EHC contact points throughout the world who are asked to comment
on its completeness and accuracy and, where necessary, provide
additional material. The contact points, usually designated by
governments, may be Participating Institutions, IPCS Focal Points, or
individual scientists known for their particular expertise. Generally
some four months are allowed before the comments are considered by the
RO and author(s). A second draft incorporating comments received and
approved by the Director, IPCS, is then distributed to Task Group
members, who carry out the peer review, at least six weeks before
their meeting.
The Task Group members serve as individual scientists, not as
representatives of any organization, government or industry. Their
function is to evaluate the accuracy, significance and relevance of
the information in the document and to assess the health and
environmental risks from exposure to the chemical. A summary and
recommendations for further research and improved safety aspects are
also required. The composition of the Task Group is dictated by the
range of expertise required for the subject of the meeting and by the
need for a balanced geographical distribution.
The three cooperating organizations of the IPCS recognize the
important role played by nongovernmental organizations.
Representatives from relevant national and international associations
may be invited to join the Task Group as observers. Although observers
may provide a valuable contribution to the process, they can only
speak at the invitation of the Chairperson. Observers do not
participate in the final evaluation of the chemical; this is the sole
responsibility of the Task Group members. When the Task Group
considers it to be appropriate, it may meet in camera.
All individuals who as authors, consultants or advisers
participate in the preparation of the EHC monograph must, in addition
to serving in their personal capacity as scientists, inform the RO if
at any time a conflict of interest, whether actual or potential, could
be perceived in their work. They are required to sign a conflict of
interest statement. Such a procedure ensures the transparency and
probity of the process.
When the Task Group has completed its review and the RO is
satisfied as to the scientific correctness and completeness of the
document, it then goes for language editing, reference checking and
preparation of camera-ready copy. After approval by the Director,
IPCS, the monograph is submitted to the WHO Office of Publications for
printing. At this time a copy of the final draft is sent to the
Chairperson and Rapporteur of the Task Group to check for any errors.
It is accepted that the following criteria should initiate the
updating of an EHC monograph: new data are available that would
substantially change the evaluation; there is public concern for
health or environmental effects of the agent because of greater
exposure; an appreciable time period has elapsed since the last
evaluation.
All Participating Institutions are informed, through the EHC
progress report, of the authors and institutions proposed for the
drafting of the documents. A comprehensive file of all comments
received on drafts of each EHC monograph is maintained and is
available on request. The Chairpersons of Task Groups are briefed
before each meeting on their role and responsibility in ensuring that
these rules are followed.
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR VINYL CHLORIDE
Members
Dr A. Barbin, International Agency for Research on Cancer, Lyon,
France
Professor V.J. Feron, TNO Nutrition and Food Research Institute, HE
Zeist, Netherlands
Ms P. Heikkilä, Uusimaa Regional Institute of Occupational Health,
Helsinki, Finland
Dr J. Kielhorn, Chemical Risk Assessment, Fraunhofer Institute for
Toxicology and Aerosol Research, Hanover, Germany (Co-Rapporteur)
Professor M. Kogevinas, Respiratory and Environmental Health Research
Unit, Municipal Institute of Medical Investigation (IMIM), Barcelona,
Spain
Mr H. Malcolm, Institute of Terrestrial Ecology, Monks Wood, Abbots
Ripton, Huntingdon, Cambridgeshire, United Kingdom (Co-Rapporteur)
Dr W. Pepelko, National Center for Environmental Assessment, Office of
Research and Development, US EPA, Washington DC, USA
Dr A. Pintér, National Institute of Environmental Health, Budapest,
Hungary (Vice-Chairman)
Dr L. Simonato, Department of Oncology, University of Padua, Venetian
Tumours Registry, Padua, Italy
Professor H. Vainio, Division of Health Risk Assessment, National
Institute of Environmental Medicine, Karolinska Institute, Stockholm,
Sweden (Chairman)
Dr E.M. Ward, Division of Surveillance Hazard, Evaluation and Field
Studies, National Institute for Occupational Safety and Health
(NIOSH), Robert Taft Laboratory, Cincinnati, Ohio, USA (Contact
address: Environmental Cancer Epidemiology, International Agency for
Research on Cancer, Lyon, France
Dr J.M. Zielinski, Biostatistics and Research Coordination Division,
Ottawa, Ontario, Canada
Secretariat
Dr A. Aitio, International Programme on Chemical Safety, World Health
Organization, Geneva, Switzerland (Secretary)
Dr I. Mangelsdorf, Chemical Risk Assessment, Fraunhofer Institute for
Toxicology and Aerosol Research, Hanover, Germany
Dr C. Melber, Chemical Risk Assessment, Fraunhofer Institute for
Toxicology and Aerosol Research, Hanover, Germany
Dr U. Wahnschaffe, Chemical Risk Assessment, Fraunhofer Institute for
Toxicology and Aerosol Research, Hanover, Germany
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR VINYL CHLORIDE
A WHO Task Group on Environmental Health Criteria for Vinyl
Chloride met at the Fraunhofer Institute for Toxicology and Aerosol
Research, Hanover, Germany, from 25 to 29 January 1999. Professor H.
Muhle welcomed the participants on behalf of the Institute and its
Director, Professor U. Heinrich. Dr A. Aitio, IPCS, welcomed the
participants on behalf of the Director, IPCS, and the three IPCS
co-operating organisations (UNEP, ILO, and WHO). The Group reviewed
and revised the draft and made an evaluation of the risks for human
health and the environment from exposure to vinyl chloride.
The first and second drafts of this monograph were prepared,
under the co-ordination of Dr I. Mangelsdorf, by the authors
Dr J. Kielhorn, Dr C. Melber and Dr U. Wahnschaffe. In the preparation
of the second draft, the comments received from the IPCS contact
points were carefully considered.
Dr A. Aitio of the IPCS Central Unit was responsible for the
scientific aspects of the monograph, and Dr P.G. Jenkins for the
technical editing.
The efforts of all who helped in the preparation and finalisation
of the monograph are gratefully acknowledged.
* * *
The Federal Ministry for the Environment, Nature Conservation and
Nuclear Safety, Germany, contributed financially to the preparation of
this Environmental Health Criteria monograph, and the meeting was
organised by the Fraunhofer Institute for Toxicology and Aerosol
Research.
ABBREVIATIONS
Epsilon A 1, N 6-ethenoadenine
Epsilon C 3, N 4-ethenocytosine
Epsilon dA 1, N 6-etheno-2'-deoxyadenosine
Epsilon dC 3, N 4-etheno-2'-deoxycytidine
Epsilon G ethenoguanine
7-OEG 7-(2'-oxoethyl)guanine
ALAT alanine aminotransferase
ASAT aspartate aminotransferase
ASL angiosarcoma of the liver
BCF bioconcentration factor
CA chromosomal aberration
CAA chloroacetaldehyde
CEO chloroethylene oxide
CI confidence interval (95% unless otherwise stated)
CNS central nervous system
CYP2E1 cytochrome P-450 isozyme 2e1
ECD electron capture detection
EDC 1,2-dichloroethane
FDA Food and Drug Administration (USA)
FID flame ionization detector
GC gas chromatography
GST glutathione S-transferase
HCC hepatocellular carcinoma
HLA human-leukocyte-associated antigen
HPLC high performance liquid chromatography
HWD hazardous waste dump
IR infrared
LOAEL lowest-observed-adverse-effect level
MN micronuclei
MOR morbidity odds ratio
MS mass spectrometry
MSW municipal solid waste
NER non-extractable residue
NOAEC no-observed-adverse-effect concentration
NOAEL no-observed-adverse-effect level
NOEL no-observed-effect level
PCDD polychlorinated dibenzodioxin
PCDF polychlorinated dibenzofuran
PCE tetrachloroethene (perchloroethene)
PID photoionization detector
PVC polyvinyl chloride
SCE sister chromatid exchange
SIR standardized incidence ratio
SLRL sex-linked recessive lethal
SMR standardized mortality ratio
TCE trichloroethene
TEQ toxic equivalent quantity
UV ultraviolet
VC vinyl chloride
VOC volatile organic compound
1. SUMMARY
This monograph deals with vinyl chloride (VC) monomer itself and
is not an evaluation of polyvinyl chloride (PVC), the polymer of VC.
Exposures to VC in mixtures are not addressed.
1.1 Identity, physical and chemical properties, and analytical
methods
Under ambient conditions, VC is a colourless, flammable gas with
a slightly sweet odour. It has a high vapour pressure, a high value
for Henry's Law constant and a relatively low water solubility. It is
heavier than air and is soluble in almost all organic solvents. It is
transported in liquid form under pressure.
At ambient temperatures in the absence of air, dry purified VC is
highly stable and non-corrosive but above 450°C, or in the presence of
sodium or potassium hydroxide, partial decomposition can occur.
Combustion of VC in air produces carbon dioxide and hydrogen chloride.
With air and oxygen, very explosive peroxides can be formed,
necessitating a continuous monitoring and limitation of the oxygen
content, particularly in VC recovery plants. In the presence of water,
hydrochloric acid is formed.
Polymerization reactions to PVC are technically the most
important reactions from an industrial view, but addition reactions
with other halogens at the double bond, e.g., to yield
1,1,2-trichloroethane or 1,1-dichloroethane, are also important.
The concentration of VC in air can be monitored by trapping it on
adsorbents and, after liquid or thermal desorption, analysis by gas
chromatography. In ambient air measurements, several adsorbents in
series or refrigerated traps may be needed to increase the efficiency
of trapping. Peak concentrations at workplaces can be measured with
direct-reading instruments based on, for instance, FID or PID. In
continuous monitoring, IR and GC/FID analysers combined with data
logging and processing have been used. In analysis of VC in liquids
and solids, direct injection, extraction and more increasingly head
space or purge-and-trap techniques are applied. Also in these samples,
VC is analysed by GC fitted to, for instance, FID or MS detectors.
1.2 Sources of human and environmental exposure
VC is not known to occur naturally although it has been found in
landfill gas and groundwater as a degradation product of chlorinated
hydrocarbons deposited as solvent wastes in landfills or in the
environment of workplaces using such solvents. VC is also present in
cigarette smoke.
VC is produced industrially by two main reactions: a) the
hydrochlorination of acetylene; and b) thermal cracking (at about
500°C) of 1,2-dichloroethane (EDC) produced by direct chlorination
(ethylene and chlorine) or oxychlorination (ethylene, HCl and air/O2)
of ethylene in the "balanced process". The latter process is the most
usual nowadays.
The world production of PVC (and therefore VC) in 1998 was about
27 million tonnes. PVC accounts for 20% of plastics material usage and
is used in most industrial sectors. About 95% of the world production
of VC is used for the production of PVC. The remainder goes into the
production of chlorinated solvents, primarily 1,1,1-trichloroethane
(10 000 tonnes/year).
Three main processes are used for the commercial production of
PVC: suspension (providing 80% of world production), emulsion (12%)
and mass or bulk (8%). Most of the case studies describing adverse
effects of VC concern plants using the suspension (also called
dispersion) process.
There have been reports of VC release through accidents in PVC
plants or during transportation. VC recovery has been introduced in
many countries to recover residual non-converted VC from
polymerization and other sources of the process such as in off-gas and
water effluents. Where special precautions are not taken, VC can be
detected in PVC resins and products.
The level of residual VC in PVC has been regulated since the late
1970s in many countries. Since then, release of VC from the thermal
degradation of PVC is either not detectable or is at very low levels.
Dioxins can be formed as contaminants in VC production. The
levels of dioxins emitted into the environment are controversial.
1.3 Environmental transport, distribution and transformation
Owing to its high vapour pressure, VC released to the atmosphere
is expected to exist almost entirely in the vapour phase. There are
indications for wet deposition.
VC has a relatively low solubility in water and has a low
adsorption capacity to particulate matter and sediment. Volatilization
of VC is the most rapid process for removal of VC introduced into
surface waters. Half-lives reported for volatilization from surface
waters range from about 1 to 40 h.
Volatilization half-lives from soil were calculated to be 0.2-0.5
days. Estimated losses of VC (after one year under a 1 m soil cover)
ranged from 0.1-45%, depending on soil type. Soil sorption
coefficients estimated from physicochemical data indicate a low
sorption potential and therefore a high mobility in soil. Another
important distribution route is leaching through the soil into
groundwater where VC may persist for years.
Laboratory experiments with aquatic organisms showed some
bioaccumulation, but no biomagnification within the foodchain.
With few exceptions, VC is not easily degraded by unadapted
microbial consortia under environmental conditions. Maximum
unacclimated biodegradation half-lives of VC were estimated to be in
the order of several months or years. However, special enrichment or
pure (e.g., Mycobacterium sp.) cultures are capable of degrading VC
under optimal culture conditions. The main degradation products were
glycolic acid or carbon dioxide after aerobic conversion and ethane,
ethene, methane or chloromethane after anaerobic transformation.
Frequently, the degradation reaction of VC proceeded faster with
aerobes than with anaerobes.
Reaction with photochemically produced OH radicals is the
dominant atmospheric transformation process, resulting in calculated
tropospheric half-lives of 1 to 4 days. Several critical compounds,
such as chloroacetaldehyde, formaldehyde and formyl chloride, are
generated during experimental photolysis reactions.
Photolytic reactions as well as chemical hydrolysis are thought
to be of minor importance in aqueous media. However, the presence of
photosensitizers may enhance the transformation of VC.
There are indications for reactions of VC with chlorine or
chloride used for water disinfection, thus leading to
chloroacetaldehyde and other undesirable compounds. Another
possibility for interaction is with salts, many of which have the
ability to form complexes with VC, perhaps resulting in increased
solubility.
Methods employed (with differing success) for removal of VC from
contaminated waters include stripping, extraction, adsorption and
oxidation. Some in situ bioremediation techniques (for groundwater
or soil) couple evaporative and other methods with microbial
treatment. VC in waste gases can be recycled, incinerated or
microbially degraded. Most of the VC produced industrially is bound in
PVC articles. Their incineration involves a risk of formation of
PCDDs/PCDFs and other unwanted chlorinated organic compounds.
1.4 Environmental levels and human exposure
There is very little exposure of the general population to VC.
Atmospheric concentrations of VC in ambient air are low, usually
less than 3 µg/m3. Exposure of the general population may be higher
in situations where large amounts of VC are accidentally released to
the environment, such as in a spill during transportation. However,
such exposure is likely to be transient. Near VC/PVC industry and
waste disposal sites, much higher concentrations (up to 8000 µg/m3
and 100 µg/m3, respectively) have been recorded.
Indoor air concentrations in houses adjacent to land fills
reached maximal concentrations of 1000 µg/m3.
The main route of occupational exposure is via inhalation and
occurs primarily in VC/PVC plants. Occupational exposures to VC
amounted to several thousands of mg/m3 in the 1940s and 1950s, and
were several hundreds of mg/m3 in the 1960s and early 1970s. After
the recognition of the carcinogenic hazards of VC, occupational
exposure standards were set at approximately 13-26 mg/m3 (5-10 ppm)
in most countries in the 1970s. Compliance with these guidelines has
considerably lowered workplace VC concentrations, but even in the
1990s higher concentrations have been reported and may still be
encountered in some countries.
VC has occasionally been detected in surface waters, sediment and
sewage sludges, with maxima of 570 µg/litre, 580 µg/kg, and
62 000 µg/litre, respectively. Soil samples near an abandoned chemical
cleaning shop contained very high VC concentrations (up to 900 mg/kg).
Maximal VC concentrations in groundwater or leachate from areas
contaminated with chlorinated hydrocarbons amounted to 60 000 µg/litre
(or more). High concentrations (up to 200 mg/litre) were detected in
well water in the vicinity of a PVC plant 10 years after leakages.
The few data available show that VC can be present in tissues of
small aquatic invertebrates and fish.
In the majority of drinking-water samples analysed, VC was not
present at detectable concentrations. The maximum VC concentration
reported in finished drinking-water was 10 µg/litre. There is a lack
of recent data on VC concentrations in drinking-water, but these
levels are expected to be below 10 µg/litre. If contaminated water is
used as the source of drinking-water, higher exposures may occur. Some
recent studies have identified VC in PVC-bottled drinking-water at
levels below 1 µg/litre. The frequency of occurrence of VC in such
water is expected to be higher than in tap water.
Packaging with certain PVC materials can result in VC
contamination of foodstuff, pharmaceutical or cosmetic products,
including liquors (up to 20 mg/kg), vegetable oils (up to 18 mg/kg),
vinegars (up to 9.8 mg/kg) and mouthwashes (up to 7.9 mg/kg). Owing to
the legislative action of many countries, a significant reduction in
VC levels and/or in the number of positive samples has been achieved
since the early 1970s.
Dietary exposure to VC from PVC packages used for food has been
calculated by several agencies and, based upon estimated average
intakes in the United Kingdom and USA, an exposure of < 0.0004 µg/kg
per day was estimated for the late 1970s and early 1980s. An early
study identified VC in tobacco smoke at the ng/cigarette range.
1.5 Kinetics and metabolism in laboratory animals and humans
VC is rapidly and well absorbed after inhalation or oral
exposure. The primary route of exposure to VC is inhalation. In animal
and human studies, under steady-state conditions, approximately 40% of
inspired VC is absorbed after exposure by inhalation. Animal studies
showed an absorption of more than 95% after oral exposure. Dermal
absorption of VC in the gaseous state is not significant.
Data from oral and inhalation studies on rats indicate rapid and
widespread distribution of VC. Rapid metabolism and excretion limits
accumulation of VC in the body. Placental transfer of VC occurs
rapidly in rats. No studies on distribution after dermal exposure have
been reported.
The main route of metabolism of VC after inhalation or oral
uptake involves oxidation by cytochrome P-450 (CYP2E1) to form
chloroethylene oxide (CEO), a highly reactive, short-lived epoxide
which rapidly rearranges to form chloroacetaldehyde (CAA). The primary
detoxification reaction of these two reactive metabolites as well as
chloroacetic acid, the dehydrogenation product of CAA, is conjugation
with glutathione catalysed by glutathione S-transferase. The
conjugation products are further modified to substituted cysteine
derivatives (S-(2-hydroxyethyl)-cysteine, N-acetyl- S-
(2-hydroxyethyl)cysteine, S-carboxymethyl cysteine and
thiodiglycolic acid) and are excreted via urine. The metabolite
carbon dioxide is exhaled in air.
CYP2E1 and glutathione S-transferase isoenzymes are known to
have large inter-species and inter-individual variation in activity.
After inhalative or oral exposure to low doses, VC is
metabolically eliminated and non-volatile metabolites are excreted
mainly in the urine. Comparative investigations of VC uptake via
inhalation revealed a lower velocity of metabolic elimination in
humans than in laboratory animals, on a body weight basis. However,
when corrected on a body surface area basis, the metabolic clearance
of VC in humans becomes comparable to that of other mammalian species.
With increasing oral or inhalative exposure, the major route of
excretion in animals is exhalation of unchanged VC, indicating
saturation of metabolic pathways. Independently of applied dose, the
excretion of metabolites via faeces is only a minor route. No studies
were located that specifically investigated excretion via the bile.
CEO is thought to be the most important metabolite in vivo,
concerning the mutagenic and carcinogenic effects of VC. CEO reacts
with DNA to produce the major adduct 7-(2'-oxoethyl)guanine (7-OEG),
and, at lower levels, the exocyclic etheno adducts,
1, N6-ethenoadenine (Epsilon A), 3, N4-ethenocytosine (Epsilon C)
and N2,3-ethenoguanine (Epsilon G). The etheno DNA adducts exhibit
pro-mutagenic properties, in contrast to the major adduct 7-OEG.
7-OEG, Epsilon A, Epsilon C and Epsilon G have been measured in
various tissues from rodents exposed to VC. Physiologically based
toxicokinetic (PBTK) models have been developed to describe the
relationship between target tissue dose and toxic end-points for VC.
1.6 Effects on laboratory mammals and in vitro test systems
VC appears to be of low acute toxicity when administered to
various species by inhalation. The 2-h LC50 for rat, mouse,
guinea-pig and rabbit were reported to be 390 000, 293 000, 595 000
and 295 000 mg/m3, respectively. No data are available on acute
toxicity after oral or dermal application. VC has a narcotic effect
after acute inhalation administration. In rats, mice and hamsters,
death was preceded by increased motor activity, ataxia and
convulsions, followed by respiratory failure. In dogs, severe cardiac
arrythmias occurred under narcosis after inhalative exposure to
260 000 mg/m3. After acute inhalation exposure to VC in rats,
pathological findings included congestion of the internal organs,
particularly lung, liver and kidney, as well as pulmonary oedema.
No studies or relevant data are available for assessing effects
of dermal exposure, skin irritation or sensitizing property of VC.
Short-term oral exposure to VC for 13 weeks in rats resulted in a
no-observed-effect level (NOEL), based on increase in liver weight, of
30 mg/kg.
In various species, the main target organ for short-term (up to
6 months) inhalation exposure to VC was the liver. Increases in
relative liver weights and hepatocellular changes were noted in rats
at 26 mg/m3 (the lowest dose level tested); at higher levels
(> 260 mg/m3) more pronounced liver changes occurred in a
dose-related manner. Other target organs were the kidney, lung and
testis. Rats, mice and rabbits seem to be more sensitive than
guinea-pigs and dogs.
Long-term exposure to VC by inhalation resulted in statistically
significant increases in mortality in some strains of rats at a dose
of as low as 260 mg/m3, in mice at 130 mg/m3 and in hamsters at
520mg/m3 for various lengths of exposure. Rats exposed to
130 mg/m3 showed reduced body weight and increased relative spleen
weight, hepatocellular degeneration and proliferation of cells lining
the liver sinusoids. Exposure to higher levels produced degenerative
alteration in the testis, tubular nephrosis and focal degeneration of
the myocardium in rats. For rats and mice exposed via inhalation, the
no-observed-adverse-effect level (NOAEL) concerning non-neoplastic
effects is below 130 mg/m3.
Chronic feeding studies showed increased mortality, increased
liver weights and morphological alteration of the liver.
After oral exposure, liver cell polymorphism (variation in size
and shape of hepatocytes and their nuclei) could be seen in rats at
levels as low as 1.3 mg/kg body weight. The NOAEL was 0.13 mg/kg body
weight.
Long-term feeding studies in rats with VC in PVC granules yielded
significantly increased tumour incidences of liver angiosarcoma (ASL)
at 5.0 mg/kg body weight per day and neoplastic liver nodules
(females) and hepatocellular carcinoma (HCC) (males) at 1.3 mg/kg body
weight per day.
In inhalation studies with VC in Sprague-Dawley rats, a clear
dose-response relationship was observed for ASL and, at high
concentrations, Zymbal gland carcinomas. No clear dose-dependency for
hepatoma or extrahepatic angiosarcoma, nephroblastomas,
neuroblastomas, or mammary malignant tumours was observed. In mice,
the spectrum of tumours induced by long-term inhalation exposure is
similar to that observed in rats but an increase in lung tumours was
only observed in mice. In hamsters, an increased tumour incidence of
ASL, mammary gland and acoustic duct tumours, melanomas, stomach and
skin epithelial tumours was reported.
The mutagenic and genotoxic effects of VC have been detected in a
number of in vitro test systems, predominantly after metabolic
activation. VC is mutagenic in the Ames test in S. typhimurium
strains TA100, TA1530 and TA1535 but not in TA98, TA1537 and TA1538,
indicating mutations as a result of base-pair substitutions
(transversion and transition) rather than frameshift mutations. This
is in agreement with the finding that etheno-DNA adducts formed by the
reactive metabolites CEO and CAA convert to actual mutations by
base-pair substitutions.
Other gene mutation assays in bacteria, yeast cells and mammalian
cells have revealed positive results exclusively in the presence of
metabolic activation. Mutagenic effects were also reported in a human
cell line containing cloned cytochrome P-450IIE1, which is capable of
metabolizing VC. Gene mutation was also detected in plant
( Tradescantia ) cuttings exposed to VC. In gene conversion assays,
positive results were reported with Saccharomyces cerevisiae in the
presence of a metabolic activation system. VC exposure induced
unscheduled DNA synthesis in rat hepatocytes and increased
sister-chromatid exchange (SCE) in human lymphocytes after addition of
exogenic activation system. No growth inhibition was detected in DNA
repair-deficient bacteria without metabolic activation. Cell
transformation assays revealed positive results both with and without
metabolic activation.
VC exposure induced gene mutation and mitotic recombination in
Drosophila melanogaster but not gene mutation in mammalian germ
cells. VC showed clastogenic effects in rodents, increased SCE in
hamsters and induced DNA breaks in mice. In host-mediated (rat)
assays, VC induced gene conversion and forward mutations in yeast.
CEO and CAA were found to be mutagenic in different test systems.
CEO is a potent mutagen, whereas CAA is highly toxic. CEO and CAA were
found to be carcinogenic in mice, CEO being much more active than CAA.
Mutations of the ras and p53 genes were analysed in liver
tumours induced by VC in Sprague-Dawley rats: base-pair substitutions
were found in the Ha- ras gene in hepatocellular carcinoma (HCC) and
in the p53 gene in ASL. These mutations are in agreement with the
observed formation and persistence of etheno adducts in liver DNA,
following exposure of rats to VC, and with the known pro-mutagenic
properties of etheno adducts.
Studies into the mechanism of carcinogenicity of VC suggest that
the reactive epoxide intermediate CEO interacts with DNA to form
etheno adducts, which result in a base-pair substitution leading to
neoplastic transformation.
1.7 Effects on humans
Concentrations of VC in the region of 2590 mg/m3 (1000 ppm),
which were not unusual prior to 1974, over periods ranging from 1
month to several years, have been reported to cause a specific
pathological syndrome found in VC workers called the "vinyl chloride
illness". Symptoms described were earache and headache, dizziness,
unclear vision, fatigue and lack of appetite, nausea, sleeplessness,
breathlessness, stomachache, pain in the liver/spleen area, pain and
tingling sensation in the arms/legs, cold sensation at the
extremities, loss of libido and weight loss. Clinical findings
included scleroderma-like changes in the fingers with subsequent bony
changes in the tips of the fingers described as acroosteolysis,
peripheral circulatory changes identical with the classical picture of
Raynaud's disease and enlargement of the liver and spleen with a
specific histological appearance, and respiratory manifestations.
Studies in humans have not been adequate to confirm effects on
the reproductive system. A few morbidity studies have reported
elevated incidence of circulatory diseases among vinyl chloride
workers. However, large cohort studies have found lower cardiovascular
disease mortality.
There is strong and consistent evidence from epidemiological
studies that VC exposure causes the rare tumour, angiosarcoma of the
liver. Brain tumours and hepatocellular carcinoma of the liver may
also be associated with VC, although the evidence cannot be considered
definitive. Other cancer sites reported to be in excess, but less
consistently, include lung, lymphatic and haematopoietic tissue, and
skin.
VC is mutagenic and clastogenic in humans. Frequencies of
chromosomal aberrations (CA), micronuclei (MN) and SCE in the
peripheral blood lymphocytes of workers exposed to high levels of VC
have been shown to be raised compared to controls. Although in many
studies the exposure concentrations and duration of exposure were only
estimated, a dose-response relationship and a "normalization" of
genotoxic effects with time after reduction of exposure can be seen.
Point mutations have been detected in p53 and ras genes in
tumours from highly exposed (before 1974) autoclave workers with liver
angiosarcoma (ASL) and another VC worker with hepatocellular
carcinoma.
Biological markers that have been investigated as indicators for
VC exposure or VC-induced effects include a) excretion of VC
metabolites (e.g., thiodiglycolic acid), b) genetic assays (e.g.,
chromosomal abnormalities or micronucleus assay), c) levels of enzymes
(e.g., in liver function tests), d) serum oncoproteins (p21 and p53)
and/or their antibodies as biomarkers of VC-induced effects.
Children living near landfill sites and other point sources may
be at increased risk based on suggested evidence of early life
sensitivity in animal studies. However, there is no direct evidence in
humans.
In epidemiological studies, a clear dose-response is only evident
for ASL alone or in combination with other liver tumours. Only one
epidemiological study has sufficient data for quantitative
dose-response estimation.
1.8 Effects on other organisms in the laboratory and field
There is a lack of standard toxicity data on the survival and
reproduction of aquatic organisms exposed to VC. Care must be taken
when interpreting the data that are available, as most of it was
generated from tests where the exposure concentration was not measured
and therefore losses due to volatilization were not taken into
account.
The lowest concentration of VC that caused an effect in
microorganisms was 40 mg/litre. This was an EC50 value based upon
inhibition of respiration in anaerobic microorganisms in a batch assay
over 3.5 days.
The lowest concentration that caused an effect in higher
organisms was 210 mg/litre (48-h LC50 for a freshwater fish); with a
corresponding no-observed-adverse-effect concentration (NOAEC) of 128
mg/litre. Effects due to VC have been reported at lower concentrations
in other species, but the ecological significance of these effects was
not verified.
VC concentrations predicted to be non-hazardous to freshwater
fish were calculated to range from 0.088 to 29 mg/litre.
There is a paucity of data concerning the effects of VC on
terrestrial organisms.
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, AND ANALYTICAL METHODS
This monograph deals with vinyl chloride (VC) monomer itself and
is not an evaluation of polyvinyl chloride (PVC), the polymer of VC.
2.1 Identity
Chemical formula: C2H3Cl
Chemical structure: H2C=CHCl
Relative molecular mass: 62.5
Common names: Vinyl chloride
CAS chemical name: Ethene, chloro-
IUPAC name: Chloroethene
CAS Registry 75-01-4
number:
EC Number: 602-023-007
EINECS Number: 2008310
Synonyms: vinyl chloride monomer,
monochloroethene;
monochloroethylene;
1-chloroethylene, chloroethene,
chloroethylene
Purity 99.9% (by weight); water: max.
120 mg/kg; HCl: max. 1 mg/kg
(BUA, 1989)
Up to the 1960s the purity was not
so high (Lester et al., 1963)
Typical trace 10-100 mg/kg range: chloromethane,
components chloroethane; 1-10 mg/kg range:
ethyne
(acetylene), 1,3-butadiene, butene,
1,2-dichloroethane, ethene,
propadiene
(allene), propene, 1-butyne-3-ene
(vinyl acetylene) (BUA, 1989)
2.2 Physical and chemical properties
Some physical properties of VC are given in Table 1. Under
ambient conditions, vinyl chloride is a colourless, flammable gas with
a slightly sweet odour. It is heavier than air and has relatively low
solubility. There are discrepancies in the literature with regard to
Henry's Law constant (air-water partition coefficient, Hc). Whereas
some authors give a value between 1 and 3 kPa.m3/mol, other sources
quote a value two orders higher. Large uncertainties in the absolute
aqueous solubility in older studies probably contribute most to these
discrepancies (Ashworth et al., 1988). It is an azeotrope with water:
0.1 parts water/100 parts vinyl chloride (Bönnighausen, 1986; Rossberg
et al., 1986). VC is soluble in almost all organic solvents.
Since it is a gas that is heavier than air, VC can spread over
the ground creating an exposure long distances away from the original
source and can form explosive mixtures. The odour threshold value is
very subjective (see Table 1) and is far above the present accepted
occupational safety threshold values (see Annex 1).
VC is transported as a compressed liquid. As it does not tend to
polymerize easily, liquid VC (in the absence of oxygen and water) can
be stored and transported without polymerization inhibitors
(Bönnighausen, 1986).
At ambient temperatures in the absence of air, dry purified VC is
highly stable and non-corrosive. Above 450°C, partial decomposition
occurs yielding acetylene, hydrogen chloride and trace amounts of
2-chloro-1,3-butadiene (chloroprene) (Rossberg et al., 1986). This
reaction also occurs by lower temperatures (at 30°C and under) in the
presence of sodium or potassium hydroxide (Bönnighausen, 1986).
Combustion of VC in air produces carbon dioxide and hydrogen
chloride. Under oxygen deficient conditions, traces of phosgene may be
formed (Rossberg et al., 1986). In chlorine-atom-initiated oxidation
of VC, the vinyl chloride peroxide formed decomposes to formaldehyde,
hydrogen chloride and carbon monoxide (Bauer & Sabel, 1975; Sanhueza
et al., 1976).
With air and oxygen, very explosive peroxides can be formed
(Rossberg et al., 1986). There are reports of explosions in vinyl
chloride plants (Terwiesch, 1982). In VC recovery plants there is a
higher chance of explosion, which necessitates continuous monitoring
and limitation of the oxygen content.
Table 1. Some physical and chemical properties of vinyl chloride
Melting point -153.8 °C Bönnighausen (1986);
Dreher (1986)
Boiling point -13.4 °C Bönnighausen (1986);
(at 101.3 kPa) Dreher (1986)
Flash point -78 °C Bönnighausen (1986);
(open cup) Dreher (1986)
Autoignition 472 °C Bönnighausen (1986);
temperature Dreher (1986)
Critical temperature 156 °C Bönnighausen (1986);
Dreher (1986)
Critical pressure 5600 kPa Bönnighausen (1986);
Dreher (1986)
Explosion limits in air 3.8-29.3 vol% Bönnighausen (1986)
in air (20 °C);
4-22 vol% Dreher (1986)
Decomposition 450 °C Bönnighausen (1986)
temperature
Density (20 °C) 0.910 g/cm3 Bönnighausen (1986)
Vapour pressure at -20 °C 78 kPa Dreher (1986)
0 °C 165 kPa
20 °C 333 kPa
Solubility of VC in 0.95 wt% (9.5 g/litre) DeLassus & Schmidt
water; extrapolated over temperature (1981)
from low pressure range
experiments 1.1 g/litre Euro Chlor (1999)
over range 15-85 °C
at 20 °C
Solubility of water 0.02 ml (-20 °C) Bönnighausen (1986)
in 100 g VC 0.08 ml (+20 °C)
Henry's Law Constant 1.96 at 17.5 °C Gossett (1987)
(Hc) (kPa.m3/mol) 2.0-2.8 at 25 °C Ashworth et al. (1988)
18.8 at 20 °C Euro Chlor (1999)
Table 1. (cont'd)
Solubility in organic soluble in most Dreher (1986)
solvents organic liquids and
solvents;
insoluble in lower Bönnighausen (1986)
polyalcohols
log n-octanol/water 1.58 (measured; BUA (1989)
partition coefficient 22 °C)
(log Kow) 1.36 (calculated) BUA (1989)
1.52 Gossett et al. (1983)
Odour threshold value 26-52 mg/m3 by Hori et al. (1972)
some, but by all at
2600 mg/m3
650 mg/m3 Baretta et al. (1969)
10 700 mg/m3 Patty (1963)
Polymerization reactions to form PVC are the most important
reactions from an industrial view (see section 3.2.1.2).
nH2C = CHCl -> (- H2C - CHCl -)n; Delta HR = -71.2 kJ/mol
The reaction is exothermic. Addition reactions with other halogens at
the double bond, for instance, to yield 1,1,2-trichloroethane or
1,1-dichloroethane, are also important. Catalytic halogen exchange by
hydrogen fluoride gives vinyl fluoride (Rossberg et al., 1986). In the
presence of water, hydrochloric acid is formed which attacks most
metals and alloys. This hydrolysis probably proceeds via a peroxide
intermediate (Lederer, 1959).
Vinyl chlorine reacts with chlorine to form trichloroethane.
1,1-Dichloroethane is formed from the exothermal reaction of VC with
hydrogen chloride in the presence of iron compounds.
2.3 Conversion factors
1 ppm = 2.59 mg/m3 at 20°C and 101.3 kPa
1 mg/m3 = 0.386 ppm
2.4 Analytical methods
2.4.1 General analytical methods and detection
Stringent regulations for the production, use and handling of
carcinogenic VC have been made in several countries (see Annex 1)
necessitating the usage of reliable methods to detect trace amounts of
this compound in air, water and in PVC articles in such human contact
applications as food packing, medical equipment and potable water
transport.
VC in air has been monitored by trapping it on different
adsorbents, e.g., activated charcoal, molecular sieve and carbotrap.
VC can be removed from adsorbents by liquid or thermal desorption
and analysed by GC fitted with FID, PID or MS detection. In ambient
air measurements, several adsorbents in sieves or refrigerated traps
have been used to increase the efficiency of trapping. In continuous
monitoring of workplace and ambient concentration, IR and GC/FID
analysers can be used.
Direct injection, extraction and more increasingly head space or
purge and trap techniques have been applied for analysis of liquids
and solids. VC can be detected by GC fitted to, for instance, FID,
PID, MS or Hall detectors.
A pre-concentration step and chemical derivatization may increase
sensitivity.
An overview of analytical methods for detecting VC in various
matrices is given in Table 2.
2.4.2 Sample preparation, extraction and analysis for different
matrices
2.4.2.1 Air
Most methods are based on that of Hill et al. (1976a), using
adsorption on activated charcoal, desorption with carbon disulfide and
analysis by GC/FID. Kruschel et al. (1994) used a three-stage carbon
molecular sieve adsorbent cartridge to collect a wide range of
selected polar and non-polar VOCs. After purging with helium prior to
analysis, levels of water and other interfering compounds were reduced
sufficiently to allow cryogenic preconcentration and focusing of the
sample onto the head of the analytical column. VC was detected at
levels below the detection limit of former methods.
Landfill gas monitoring has been carried out by trapping VC on a
molecular sieve, and samples have been analysed using, for instance,
GC/MS (Bruckmann & Mülder, 1982) or GC/ECD with prior conversion to
the 1,2-dibromo derivative (Wittsiepe et al., 1996).
Table 2. Analytical methodsa
Matrix Sampling/preparation Separation Detector Detection Comments References
limitb
Air
Expired air collected in 50 ml GC FID 50 ppb Baretta et al. (1969)
pipettes; direct (packed (130 µg/m3)
injection column)
Expired air multistage cryogenic GC FID; MS low ppb low reproducibility; Conkle et al. (1975)
trapping; thermal (packed & cap.) long sampling time
desorption
Expired air 500 ml charcoal tubes GC 0.3 mg/m3 Krajewski & Dobecki
(1978, 1980)
Expired air 1 litre canister; capGC MS n.g. for collecting Pleil & Lindstrom
pressurized with alveolar samples; e.g., (1997)
neutral gas; cryogenic 16 and 25 µg/m3
concentration
Air in car charcoal tube, CS2 GC FID 10 ppb Going (1976); Hedley
interior desorption (26 µg/m3) et al. (1976)
Ambient air activated charcoal/CS2 GC FID 2.6 mg/m3 Hill et al. (1976a)
Ambient air silica gel at -78°C, GC FID 2.6 mg/m3 IARC (1978)
thermal desorption
Ambient air activated charcoal GC n.g. 0.5 ppb Dimmick (1981)
column; 24-h sampling (1.3 µg/m3)
Ambient air sampling (1 to 10 HRGC MS 1 ng VOCs Kruschel et al.
litre) on carbon trap; FID (0.3 µg/m3) (1994)
thermal desorption
Table 2. (cont'd)
Matrix Sampling/preparation Separation Detector Detection Comments References
limitb
Ambient air solid phase sample capGC IMS 2 mg/m3 new method for Simpson et al.
trap preconcentration field monitoring (1996)
Landfill gas (20 litre) carbon capGC ECD 82 ng/m3 Wittsiepe et al.
molecular sieve; CS2 (1996)
desorption;
conversion to
1,2,-dibromo
derivative
Tobacco charcoal tube, CS2 GLC ECD 15 pg per Hoffmann et al.
smoke extraction; injection (1976)
conversion to
1,2,-dibromo
derivative
Workplace CS2 desorption GC (packed FID 0.04 µg working range NIOSH (1994)
air column) (5 litre 0.4 to 40 mg/m3 (based on Hill
sample) et al., 1976a)
Workplace charcoal sorbent GC (packed FID 5 mg/m3 Kollar et al.
air tube; extraction column) (3 dm3 sample) (1988)
with
nitro-methane
Workplace carbon trap, GC FID 2.6 mg/m3 Hung et al. (1996)
air thermal desorption
Workplace activated charcoal, GC FID 0.1 mg/m3 working range HSE (1987);
air CS2 desorption 0.07-25 mg/m3 for ASTM (1993)
30-litre samples
Workplace continuous process GC FID n.g. Pau et al. (1988)
air sampling
Table 2. (cont'd)
Matrix Sampling/preparation Separation Detector Detection Comments References
limitb
Workplace continuous pyrolysis detection 1 mg/m3 Nakano et al.
air sampling of HCl (1996)
Water
Water purge & trap GC MC n.g. in PVC pipes Dressman &
McFarren (1978)
Water purge & trap GC MS 0.05 µg/litre Schlett &
Pfeifer (1993)
Water headspace capGC MS 1 µg/litre Gryder-Boutet &
Kennish (1988)
Water purge & trap capGC PID-ELCD n.g. modification of US Driscoll et al.
EPA Methods 601 (1987)
& 602 for VOC
Water purge & trap capGC PID-ELCD 0.1 µg/litre VOC Ho (1989)
Water purge & trap; capGC ECD 1.6 ng/litre Wittsiepe et al.
CS2 desorption; (1990, 1993)
1,2-dibromo
derivatization
Water GC PID and n.g. VCM loss during Soule et al.
Hall laboratory holding (1996)
detector time
Water purge & trap GC FID n.g. VOC Lopez-Avila et
al. (1987a)
Table 2. (cont'd)
Matrix Sampling/preparation Separation Detector Detection Comments References
limitb
Water CS2 desorption; capGC ECD 1.6 ng/litre Wittsiepe et al.
conversion to (1990, 1996)
1,2,-dibromo
derivative
Bottled headspace with GC MS 10 ng/litre Benfenati et al.
drinking- thermal desorption (1991)
water cold-trap
injector (TCT)
Water solid phase capGC FID n.g. Shirey (1995)
micro-extraction MS
Food, liquids, biological fluids and tissues
Liquid headspace GC FID 0.1 ppb Watson et al.
drugs; (1979)
cosmetics
Food; headspace GLC confirmation 10 ppb Williams (1976a)
liquids with MS
Liquids derivatization to GLC ECD 15 µg/litre Williams (1976b)
1-chloro-1,2- (vinegar);
dibromoethane 50 µg/litre oil
Food direct injection GC FID 2-5 µg/kg detection limit UK MAFF (1978)
depends on medium
Food headspace GC FID 1 µg/kg IARC (1978)
Table 2. (cont'd)
Matrix Sampling/preparation Separation Detector Detection Comments References
limitb
Oil GC FID 5 µg/litre Rösli et al.
(1975) based on
Williams & Miles
(1975)
Food headspace GC n.g. 2-5 µg/kg UK MAFF (1978)
Intravenous headspace capGC FID 1 µg/litre Arbin et al.
solutions (1983)
Blood (rat) headspace GC FID 5 µg/litre Zuccato et al.
ethanol-water (1979)
extraction
Tissues freezing, GC FID 30 µg/kg Zuccato et al.
(rat) homogenization (1979)
then as above
Urine dry; dissolution GC MS 50 µg/litre TDGA Müller et al.
in methanol; (1979)
methylation with
diazomethane
Urine extraction and GC FID 10 mg/litre TDGA; standard: Draminski &
silylation o-phthalic acid Trojanowska
(1981)
Urine conversion to GC MS < 0.5 TDGA; standard: Pettit (1986)
dibutyl ester µmol/litre pimelic acid
PVC
PVC products charcoal tube, GC FID 10 ppb Going (1976)
CS2 desorption (26 µg/m3)
Table 2. (cont'd)
Matrix Sampling/preparation Separation Detector Detection Comments References
limitb
PVC headspace packed column 5 ppb ASTM (1985)
GC
PVC headspace capGC FID update suggestion for Wright et al.
FID-PID ASTM (1985) (1992)
PVC extraction/headspace GC FID 0.1 mg/kg Puschmann (1975);
IARC (1978)
PVC HPLC < 1 ppm for temperatures Kontominas et al.
packaging simulating storage (1985)
of foods conditions (8 to 27 °C)
PVC dynamic headspace GC FID low ppb Poy et al. (1987)
with a sparging and
focusing step
before thermal
desorption
PVC film or GC FID 2.2 ng Gilbert et al.
resin (5 ppb (w/w)) (1975)
Packaging MS 8.7 pg
materials
PVC bags purge/trap GC FID/ECD 0.3 ppb Thomas & Ramstad
(Tenax/charcoal) (1992)
Table 2. (cont'd)
a Abbreviations: capGC = capillary gas chromatography; ECD = electron capture; ELCD = electrolytic conductivity detector;
FID = flame ionization detector; GC = gas chromatography; GLC = gas-liquid chromatography; HRGL = high-resolution gas
chromatography; IMS = ion mobility spectrometry; MC = microcoulometric titration detector; MS = mass spectrometry;
PID = photoionization detector; SPME = solid-phase microextraction; TDGA = determination of the metabolite, thiodiglycolic
acid; VOC = volatile organic chemical (a general method); n.g. = not given
b The % recovery was not given in most cases
2.4.2.2 Water
VC is first purged from the water and then collected for GC
analysis by headspace/purge and trap. VC is highly volatile and has a
low specific retention volume on Tenax-GC, the most commonly used
trapping medium in purge/trap analysis. Combination traps such as
Tenax/silica gel/charcoal (Ho, 1989) or Tenax/OV-1/silica (Lopez-Avila
et al., 1987b) have been used. Another approach is to bypass the trap
altogether by purging directly onto a cryocooled capillary column
(Gryder-Boutet & Kennish, 1988; Pankow & Rosen, 1988; Cochran, 1988;
Cochran & Henson, 1988), but here there are complications due to the
need to remove water when stripping from an aqueous solution. A more
recent adaptation of the headspace method uses solid-phase
microextraction (SPME) in which a stationary phase, usually
poly(dimethylsiloxane), coated on a fused-silica fibre is used to
extract aqueous samples in completely filled sealed vials (Shirey,
1995).
It should be noted when measuring VC content in water or
groundwater that samples should be analysed as soon as possible, as
the VC content decreases with holding time (Soule et al., 1996).
2.4.2.3 PVC resins and PVC products
For the quantification of residual VC in PVC, a solid and a
solution approach have been used. The former involves the
equilibration of a solid polymer sample at 90°C in a sealed system,
followed by headspace analysis with single or multiple extraction
(Berens et al., 1975; Kolb, 1982). The solution approach involves the
equilibration of a 10% solution of PVC in dimethylacetamide in a
sealed system, followed by analysis of the headspace gas (Puschmann,
1975). A automatic dynamic headspace method involving a sparging and a
focusing step before desorption into the GC column has been developed
to increase the sensitivity of the solution approach method (Poy et
al., 1987).
2.4.2.4 Food, liquid drug and cosmetic products
For monitoring VC in foods in contact with PVC packaging,
headspace GC is the usual method. Before the levels of VC allowed in
PVC was regulated in the 1970s (see Annex I), many of the VC levels
observed were high enough to be determined by direct injection
methods. Limits of detection are given as 2, 5 and 5 µg/kg for
aqueous, ethanolic, and oleaginous medium respectively using
headspace-GC-FID (UK MAFF, 1978). Williams (1976a,b) reported a
gas-liquid chromatographic method using subsequent GC-MS confirmation,
and a further GC/ECD method requiring derivatization to
1-chloro-1,2-dibromoethane for determination of VC content in liquid
foods.
Methods for VC levels in liquid drug and cosmetic preparations
were described by Watson et al. (1979). A weighed aliquot of the
commercial product in a tightly septum-sealed vial with accurately
known headspace volume is heated to 50°C for 30 min. A portion of the
warm headspace gas is then injected into a GC equipped with FID and a
styrene-divinylbenzene porous polymer column.
2.4.2.5 Biological samples
There are few data on VC analysis in biological tissues. The only
report available was on rat blood and tissues (Zuccato et al., 1979).
2.4.2.6 Human monitoring
Methods for measuring VC concentrations in exhaled air (breath)
have been described (see Table 2) but, although useful for studying
metabolism, they are not well suited for biological monitoring due to
the short half times in the body and the saturable metabolism of VC.
Metabolites of vinyl chloride have been identified in the urine
of rats (Müller et al., 1976; Green & Hathway, 1977) and humans
(Müller et al., 1979) using GC-MS. As there is a strong correlation
between VC exposure in humans and increased excretion of
thiodiglycolic acid (Müller et al. 1978), this metabolite has been
using for monitoring purposes. It is, however, not specific for VC as
certain drugs and other C2 compounds also have thiodiglycolic acid as
a urinary metabolite (Müller et al., 1979). Since thiodiglycolic acid
is also detected in unexposed subjects (Müller et al., 1979; van
Sittert & de Jong, 1985) and even premature babies (Pettit, 1986),
this approach can only be used to demonstrate high levels of exposure.
A discussion of biological markers for VC exposures and VC-induced
liver cancer is presented in section 8.5.
Thiodiglycolic acid has been determined by dissolving the dried
urine residue in methanol, methylating with diazomethane and analysing
with GC-MS (Müller et al., 1979), analysing the metabolite as its
dibutyl ester by GC-MS using selected ion monitoring (Pettit, 1986),
and by using GC/FID (Draminski & Trojanowska, 1981). Care must be
taken with methods which analyse for VC metabolites, as these
metabolites are not specific to VC.
A specific and sensitive new method has been reported for the
quantification of the VC metabolite N-acetyl- S-(2-hydroxyethyl)
cysteine by exchange solid-phase extraction and isotope dilution
HPLC-tandem mass spectrometry (Barr & Ashley, 1998). This method may
prove useful for monitoring occupational VC exposure, as the detection
limit of 0.68 µg/litre is low enough to detect this metabolite even in
people with no overt exposure to VC, ethylene oxide or ethylene
dibromide.
2.4.2.7 Workplace air monitoring
Before the 1960s, when it was established that VC was a
carcinogenic substance, halogen detectors and explosimeters were used,
non-specific for VC, with detection limits of 518-1295 mg/m3 (200-500
ppm). Gradually more sophisticated techniques became available for
detection of low ppm levels of VC, such as IR analysers, FID, PID
(ECETOC, 1988) and more recently mass spectrometry. In order to check
for leaks or for control measurements during cleaning and repair work,
detector tubes or direct-reading instruments with FID or PID can be
used, although they are not specific for VC and regular calibration is
necessary (Depret & Bindelle, 1998).
Continuous analyses based on, for instance, IR, GC/FID or HCl
detection have been developed (IARC, 1978; Pau et al., 1988; Nakano et
al., 1996). Analysers can be equipped for computerized data logging
and processing. The detection limit of an IR analyser depends on, for
instance, path length and is about 1.3 mg/m3 (IARC, 1978). The
analyser detecting HCl from VC pyrolyzed in a quartz tube was reported
to have a detection limit of 1 mg/m3 when the sampling time was 40
seconds (Nakano et al., 1996).
Breathing zone concentrations can be measured by sampling VC with
portable pumps or diffusion on activated charcoal (Nelms et al., 1977;
Heger et al., 1981; ASTM, 1993; NIOSH, 1994; Du et al., 1996). By
using thermosorption tubes (carbotrap 110-400), detection limits can
be decreased and the use of carbon disulfide in desorption avoided
(Hung et al., 1996).
Passive monitors for occupational personal monitoring of exposure
to VC are commercially available.
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
3.1 Natural occurrence
VC is not known to occur naturally.
3.2 Anthropogenic sources
Anthropogenic sources of VC include the intentional manufacture
of the compound for further processing, primarily to PVC, and
unintentional formation of VC in, for instance, sanitary landfills, as
a degradation product of chlorinated hydrocarbons such as those used
as solvents, and the subsequent presence of VC in emitted gases and
groundwater. VC is also found in tobacco smoke.
3.2.1 Production levels and processes
VC was first synthesized by Regnault in 1835. It was not until
the 1930s that techniques were devised to polymerize VC into stable
forms of PVC.
VC production methods were altered in 1974 in many countries
after the confirmation that VC was a human carcinogen. Manufacturers
developed closed production methods to reduce exposure of the
workforce.
Annual total world production of VC, which is approximately equal
to PVC production, was about 17 million tonnes in 1985 and over 26
million tonnes in 1995 (see Table 3). More than half the world's
capacity (64%) in 1985 was concentrated in Western Europe and the USA.
Since that time many new VC/PVC plants have been opened or are under
construction in SE Asia, Eastern Europe, the Indian subcontinent and
developing and oil-producing countries. Thus there has been a
geographical shift of VC/PVC production.
The leading producers of PVC and therefore also of VC are the
USA, Japan, Germany, France and SE Asian countries such as Taiwan and
China (CHEM-FACTS, 1992). The capacity has only increased moderately
in the USA and Western Europe in recent years. Significant increases
in production have been reported for Japan and Taiwan. All the
countries of Eastern Europe have PVC plants and have exported PVC to
Western European countries. The PVC capacity and imports and exports
for each country are given in CHEM-FACTS (1992).
VC is produced in Western Europe by 14 companies. The plants are
in Belgium (3 plants), France (3 plants), Germany (8 plants), Italy (4
plants), the Netherlands (1 plant), Spain (2 plants), Sweden (2
plants), United Kingdom (1 plant) (Euro Chlor, 1999).
Table 3. World PVC (and therefore VC) production/capacity 1980-1998
Region Production/capacity in 1000 tonnes/year
1980a 1985a 1990a 1995b 1998c
World capacity 16 000 17 000 20 700 26 400 approx. 27 000
World production 11 750 14 200 18 300
North America total 3200 3390 4700 6070
Suspension and mass 2810 2990
Vinyl acetate copolymer 200 210
Emulsion 190 190
Western Europe total 3900 4330 4800 5750 approx. 5600
Suspension and mass 33 350 3700
Vinyl acetate copolymer 130 130
Emulsion 420 500
Eastern Europe 925 1100 1200 2700d
Former Soviet Union 370 700 760
China 150 400 790
Japan 1400 1550 2070 8200e
SE Asia 330 600 900
South America 400 540 780
Rest of the world 1075 1590 2300 3680
a Allsopp & Vianello (1992)
b Rehm & Werner (1996)
c Although exact figures are not available, an increase in world production
was seen in 1998 but no great increase in Western Europe (personal
communication, European Council of Vinyl Manufacturers, 1999)
d Including former Soviet Republics
e Total Asia
The manufacture of VC/PVC is one of the largest consumers of
chlorine.
3.2.1.1 Production of VC
VC is produced industrially by two main reactions, the first is
hydrochlorination of acetylene, which proceeds via the following
reaction:
C2H2 + HCl -> CH2=CHCl
This route was used in the past when acetylene, produced via calcium
carbide from coal, was one of the important basic feedstocks for the
chemical industry (Rossberg et al., 1986). Today all USA and most
Western European manufacturers use the "balanced process" described
below. However, many Eastern European countries such as Poland and the
countries of the former Soviet Union still use acetylene to
manufacture VC because of relatively cheap raw materials such as
calcium carbide and natural gas. Mercury has been used as a catalyst,
although a new catalyst has now been developed in Russia, based on
platinum metal salts instead of mercury, which has increased yields
with acetylene conversion from 95 to 99% (Randall, 1994).
The second major production process involves thermal cracking
[reaction iii] (at about 500°C) of 1,2-dichloroethane (EDC), produced
by direct chlorination [reaction i] of ethylene or oxychlorination
[reaction ii] of ethylene in the "balanced process".
[i] CH2=CH2 + Cl2 -> Cl CH2-CH2 Cl
[ii] CH2=CH2 + 2 HCl + 5 O2 -> Cl CH2-CH2 Cl + H2O
[iii] Cl CH2-CH2 Cl -> CH2=CHCl + HCl
After the cracking (pyrolysis), HCl and unconverted EDC are separated
from VC by two steps of distillation and recycled. The VC is stored
either under pressure at ambient temperature or refrigerated at
approximately atmospheric pressure (European Council of Vinyl
Manufacturers, 1994). More than 90% of the VC produced today is based
on this route (Rossberg et al., 1986; Allsopp & Vianello, 1992).
Other methods of industrial production include:
a) VC from crack gases, where unpurified acetylene and ethylene are
chlorinated together, acetylene being first chlorinated to
1,2-dichloroethane.
b) VC from ethane, which is readily available in some countries. The
major drawback is that ethane must first be functionalized by
substitution reactions giving rise to a variety of side chain
reactions and therefore the reaction must be kinetically
controlled to obtain maximal VC yield.
3.2.1.2 Production of PVC from VC
Many PVC plants are fully integrated beginning with ethylene and
chlorine (or sodium chloride).
VC is a gas at ambient temperatures but is handled as a
compressed volatile liquid in all polymerization operations. PVC
polymerization reactors are thick-walled jacketed steel vessels with a
pressure rating of 1725 kPa. The polymerization of VC is strongly
exothermic. The explosive limits of VC in air are 4-22 vol%, and the
plant must be designed and operated with this in mind, particularly
when handling unreacted VC in the recovery system (Allsopp & Vianello,
1992).
Three main processes are used for the commercial production of
PVC: suspension (providing 80% of world production), emulsion (12%)
and mass or bulk (8%). In Western Europe, the proportion of PVC
produced by the different processes is: 80% suspension; 13% mass; 5%
emulsion and 2% copolymers (Wrede, 1995).
In the suspension (also called dispersion) process,
polymerization takes place at 40-70°C (depending on the type of PVC
being produced) in a reactor (autoclave) of 25-150 m3 capacity fitted
with a jacket and/or condenser for heat removal, as the reaction is
strongly exothermic. Precautions have to be taken in order to avoid
explosive mixtures with air. Liquid VC under its autogenous vapour
pressure is dispersed in water by vigorous stirring to form droplets
of average diameter 30-40 µm. The polymerization takes place within
these droplets and is started by addition of initiators dissolved in
the monomer. Stabilizers are added to prevent the drops rejoining and
to prevent the already polymerized PVC particles from agglomerating.
The reaction conditions can be exactly controlled and the properties
of the product, such as relative molecular mass, can be controlled
exactly. Once polymerization has ended, the autoclave charge is
emptied into degassing tanks, and the non-polymerized VC is degassed,
compressed and stored for reuse (ECETOC, 1988; Allsopp & Vianello,
1992).
During the polymerization process, the PVC is dispersed in the
aqueous phase and it cannot be prevented that a film of PVC forms on
the inside wall of the reactor. This film interferes with the transfer
of heat between the reactor and contents, and the process has to be
interrupted periodically to allow the reactor to be cleaned. The
autoclave, after being emptied, is opened, rinsed and washed either
with solvents or more usually by means of automatic high-pressure jets
(ECETOC, 1988). The latest development in this area is the use of
proprietary build-up suppressants, which are applied before every PVC
batch. After each batch, low pressure rinse with water can remove
loose polymers and the batch cycle is ready to restart. The reactor
needs then to be opened for a thorough cleaning only after 500 or more
batches (Randall, 1994).
Before awareness of the toxicity of VC, it was the autoclave
cleaning personnel who were primarily highly exposed to the compound.
In the past autoclaves were cleaned manually; the inside had to be
scraped with a spatula, or sometimes hammer and chisel to remove the
encrusted polymer adhering to the walls of the vessel and mixing
devices. Lumps of polymer often released monomer when broken,
resulting in high concentrations of VC in the autoclave. Before about
1970, it was usual to check that the level was below 1036 mg/m3 (400
ppm), i.e. two orders of magnitude below the lower explosion limit of
VC. Occupational exposure limits are now 18 mg/m3 (7 ppm) or less
(see Annex I). Further details of workplaces with a former high
exposure to VC are given in section 5.3 and Jones (1981).
Once polymerization has ended, the polymerization batch is
transferred to the stripping unit and then to the slurry tank. The
slurry is a suspension of PVC in water that has to be permanently
stirred; it is then dewatered in a centrifuge decanter and dried. The
resulting dried powder is either stored in silos or bagged. PVC is
then further processed into ready-to-use resins (Depret & Bindelle,
1998).
3.2.1.3 PVC products
PVC is a polymer of VC with 700-1500 monomeric units. It is
relatively inexpensive and is used in a wide range of applications.
PVC is a generic name. Each producer makes a range of PVC polymers,
which vary in morphology and in molecular mass according to the
intended use. PVC resins are rarely used alone but can be mixed with
heat stabilizers, e.g., lead, zinc and tin compounds (Allsopp &
Vianello, 1992), lubricants, plasticizers (e.g., diethylhexyl
phthalate) fillers and other additives, all of which can
influence its physical and mechanical properties (Williamson &
Kavanagh, 1987; Allsopp & Vianello, 1992). Such additives may
constitute up to 60% of the total weight in some finished PVC plastics
(Froneberg et al., 1982). These plastics are formed into a multitude
of consumer products by extrusion, thermoprocessing and rotational
moulding, and into rigid or flexible film by extrusion or calendering.
PVC accounts for 20% of plastic material usage and is used in
most industrial sectors (ECETOC, 1988; European Council of Vinyl
Manufacturers, 1994).
* Packaging
* bottles (produced by blow-moulding) for containing liquid
foods, beverages, cooking oils, vinegar, etc.
* rigid film (calendered or extruded) which is converted into
tubes and shaped containers by subsequent vacuum or
pressure-forming for packaging of various foodstuffs
* flexible film (made by blowing or calendering) for wrapping
solid foods such as cheese, meat, vegetables, fresh fruit,
etc.
* coatings in metal cans
* Building - floor coverings, wall coverings, windows, roller
shutters, piping; 58% of the water supply network and 80% of the
waste water disposal systems in Europe use pipes and fittings
made from PVC.
* Electrical appliances - wires and cables insulation
* Medical care - equipment such as blood bags and gloves;
pharmaceutical and cosmetic packaging. Worldwide, more than 25%
of all plastic-based medical devices used in hospitals are made
from PVC (Hansen, 1991)
* Agriculture - piping; drainage; tubing in dairy industry
* Automobiles - car dash boards and lateral trimming
* Toys
At present, the largest use of PVC is in the building sector
(Rehm & Werner, 1996).
3.2.2 Emissions from VC/PVC plants
3.2.2.1 Sources of emission during the production of VC
Process waste, by-products and unreacted material from a balanced
process and from a PVC plant include (Randall, 1994): a) light/heavy
ends from EDC purification (from direct chlorination and
oxychlorination reactors); b) heavy ends from VC purification (from
pyrolysis reactor); c) pyrolysis coke/tars (from thermal reactions in
the pyrolysis reactor); d) spent catalyst (from direct chlorination,
oxychlorination); e) recovery of unreacted VC (from PVC reactor); f)
offspec batches (from PVC reactor); g) aqueous streams (from EDC
washing, vent scrubbing, oxychlorination water; from centrifuge, VC
stripping, slurry tanks); h) vent gases (from distillation columns,
flash drums, reactor vents, storage tanks, vessel openings; from VC
recovery systems, dryer stacks, centrifuge vent, blending, storage
facilities); i) spills and leaks (sampling, pumps, flanges, pipes,
loading/unloading, valves; bag filling, agitator seals); and j)
equipment cleaning (tanks, towers, heat exchangers, piping; PVC
reactor, product dryer, recovery system).
a) Strategies for minimizing emissions
During the sixties, some control of exposure levels was
introduced resulting in an important decrease of estimated ambient
levels of VC.
In the 1970s, efforts were focused on controlling emissions at
the most significant emission points: reactors, filters and storage
tanks. Elementary modifications of equipment, room and local
ventilation by fans, provisional operation procedures, etc., enabled
the reduction of exposure levels in working areas.
Technical developments have achieved further reductions:
* removal of residual VC from the PVC suspension by stripping
between polymerization and drying with a flow of steam or in
closed-loop systems;
* appropriate collection of residual vents to thermal oxidizers or
other abatement systems;
* reduction of all sources of fugitive emission by maintenance and
upgraded equipment;
* high pressure internal cleaning of the autoclaves to remove PVC
crusts;
* intensive removal of VC before opening or a closed process
design.
Appropriate working procedures, personnel awareness and high
standard equipment, associated with good maintenance practices, are
the recommended ways to reduce fugitive emission to very low levels
(Depret & Bindelle, 1998).
b) Disposal of by-products
The main waste streams of EDC/VC production process in Europe are
light ends (gases: VC, EDC, HCl, ethylene, dioxins; aqueous effluents:
EDC, copper, dioxins) and heavy ends (viscous tars). According to the
PVC Information Council the total amount produced is approximately
0.03 tonnes of by-products per tonne of VC produced (European Council
of Vinyl Manufacturers, 1994). The fractions are either used as
feedstocks for other processes or combusted under controlled
conditions (> 900°C) or by catalytic oxidation to produce CO2, CO,
HCl and water which can be recycled (see above). Exit gases should be
treated using HCl absorbers and gas scrubbers. Spent catalyst, metal
sludges and coke from EDC cracking should be disposed of in controlled
hazardous waste dumps (HWD) or incinerated under controlled
conditions. Sludges from effluent purification should be either
combusted under controlled conditions or deposited in HWD (European
Council of Vinyl Manufacturers, 1994).
3.2.2.2 Emission of VC and dioxins from VC/PVC plants during
production
a) VC
An estimated 550 tonnes of VC was released into air, 451 kg to
water and 1554 kg to soil from manufacturing and processing facilities
in the USA in 1992, although this was not an exhaustive list (ATSDR,
1997). From a VC capacity of 6 200 000 tonnes, an average emission of
80 g/tonne can be calculated (Depret & Bindelle, 1998).
Euro Chlor (1999) reported emissions of VC from suspension PVC
plants in Europe to be 448 tonnes/year, 22 tonnes/year being released
in waste water. An estimated emission value of 300 tonnes VC/year was
provided by German producers (BUA, 1989), i.e., about 55 g VC/tonne
PVC. Total VC emissions in England and Wales were reported to be
3800 tonnes in 1993 and 18 990 tonnes in 1994 (HMIP, 1996).
An estimated 200 000 tonnes of VC was released into the worldwide
atmosphere during 1982 (based on a worldwide PVC production of
17 million tonnes), i.e., 12 kg/tonne of PVC produced (Hartmans et
al., 1985). In 1974, a VC emission of 0.5 kg/tonne PVC was estimated
(Kopetz et al., 1986).
b) Dioxins
PCDDs/PCDFs, some of which are classified as human carcinogens
(IARC, 1997), are formed during EDC/VC production. Concentrations in
VC distillation residues (PU4043) (probably "heavy ends") for two
production factories were 3192 and 5602 ng ITEQ (international toxic
equivalent)/kg, which was estimated to be 12 to 30 g of dioxin/year at
a production level of 200 000 tonnes VC/year in heavy ends (Stringer
et al., 1995). However, the VC/PVC industry argues that, although
dioxins may be found in production wastes, these are incinerated and
are not ultimately emitted to the environment (Fairley, 1997), at
least in those countries where waste streams are regulated (not all
countries possess such high standards of waste incineration). Some
companies do not incinerate their waste products but dispose of them
in other ways, e.g., deep-well injection (Stringer et al., 1995).
Wastewater from VC production can also be contaminated with
PCDDs/PCDFs, in particular if they contain suspended solids.
Installation of filtration devices should lower solid levels and
subsequently PCDD/PCDF emissions (Stringer et al., 1995). Using such
filtration devices, PCDD/PCDF concentrations in wastewater from
EDC/VC/PVC facilities in the USA ranged from not detectable to 6.7
pg/litre TEQ (Carroll et al., 1996).
Annual global emission of dioxins into the environment from
EDC/VC manufacture has been estimated to be 0.002-0.09 kg TEQ by the
European PVC industry but 1.8 kg by Greenpeace (Miller, 1993).
Virgin suspension PVC resin from 11 major production sites in
Europe was found not to contain any process-generated PCDDs/PCDFs at
concentrations above the limits of quantification (2 ppt) (Wagenaar et
al., 1998).
3.2.3 Accidental releases of VC
3.2.3.1 PVC plant and transport accidents
An explosion occurred at a VC recovery plant in 1978 in Germany,
which was set off by vinyl chloride peroxides (Terwiesch, 1982; see
also section 2.2).
A freight train with 12 tank cars of VC was derailed in McGregor,
Manitoba, Canada in March, 1980 under near-blizzard conditions and
-20°C and two of the tanks released VC (Charlton et al., 1983). One
car lost 47 500 litres in the first hour and the other car lost
23 100 litres at an initial rate of 1400 litres/hour, decreasing to
45 litres/hour after 31 h. In the first 15 min, 5680 litres of VC
vapourized by free surface evaporation; thereafter only 2.5% of the
discharging liquid evaporated rapidly. The remaining liquid in the
snow bank was assumed to have evaporated at a rate of 15% per hour,
leaving about 900 litres of VC in the snow bank after 36 h. Although
there was an explosion hazard, no fire occurred.
Two incidences, in 1988 and 1996, occurred in Germany involving
accidental release of VC due to derailment of trains transporting the
liquid substance (Neuhoff, 1988; Anon, 1996). Both accidents were
followed by explosion and fire. Derailment of a goods train occurred
1 km from Schönebeck (near Magdeburg in Germany) and the subsequent
explosion and fire produced a 600- to 800-m black column of smoke. In
all, 1044 tonnes of VC were involved of which 261 tonnes burnt and
350 tonnes could be reclaimed after the fire; 153 tonnes of HCl were
released. Median measurements of the numerous air samples taken at the
place of accident or surroundings did not exceed the German technical
guidance level of 5 vol ppm, although these were not taken until 14 h
after the fire. Maximum concentrations of VC measured were 78 mg/m3
(30 ppm) near the train and 26 mg/m3 (10 ppm) at a distance of 200 m
from the centre of the fire (Hahn et al., 1998). Levels in nearby
industrial sewage pipes were up to 1250 ppm (Anon, 1996). A cytogenic
analysis was carried out on some of the general population exposed to
VC from this accident (Hüttner & Nikolova, 1998; see section 8.1 and
Table 42).
3.2.3.2 Leakage and discharge from VC/PVC plants
Leakage from a waste liquor basin of a VC/PVC plant in Finland
caused high concentrations of VC and dichloroethene in groundwater in
1974. Concentrations of up to 484 mg/litre of the chlorocarbons were
measured in groundwater in the mid-1980s (Nystén, 1988).
It should be noted that in the 1960s and before, when the
toxicity of VC was not known, large amounts of PVC production sludges
containing VC were dumped onto landfills (and possibly still are in
countries where there are no adequate restrictions).
3.2.4 VC residues in virgin PVC resin and products
3.2.4.1 VC residues in different PVC samples
VC is not soluble in PVC nor is it absorbed or adsorbed in the
resin particle. It is entrapped and can escape to the ambient air
(Wheeler, 1981). PVC in a bulk-container loses its residual monomer at
a rate of 25 to 50% per month. Heating tends to accelerate this step,
but when the residual monomer has disappeared PVC is not a significant
source of VC.
Since the 1970s when VC was confirmed as a human carcinogen, it
has been mandatory in many countries to "degas" PVC after
polymerization (see section 3.3.1) and before further processing.
There are limit values for VC content in PVC (see Annex I). For
example, in 1974 raw PVC usually contained more than 1000 ppm of
residual VC. This was subsequently reduced to 10 ppm by regulation
(German Environmental Office, 1978). In a survey of 45 samples of raw
PVC from various countries carried out in 1976-1977, over a third had
VC residues of > 1 ppm, but a third had residues of over 50 ppm, with
4 samples over 200 ppm. The samples with the highest residue level
came from Hungary, Rumania, Italy and the USA (German Environmental
Office, 1978).
3.2.4.2 VC residues in PVC products
In a survey of PVC products carried out in 1976-1977, the
following indoor articles had a VC content of > 0.05 ppm: bathroom
tiles, piping, plastic bottle for table oil, and kitchen film. The
highest concentrations were found in music records, those bought
recently having a VC content of up to 210 ppm and one record 10 years
old having a content of 970 ppm. In record shops and other rooms
containing many records, this could have been an important source of
VC. In contrast, the VC content of toys, kitchen utensils, food
wrappings, wallpaper and car interiors was < 0.05 ppm (German
Environmental Office, 1978).
Whereas in 1974 the typical level of residual VC in PVC bottles
was 50 mg/kg, introduction of improved manufacturing practices at the
polymer resin processing stage reduced this to 3 mg/kg by 1975. In a
1978 survey, 22 out of 24 PVC bottles contained VC at less than
0.4 mg/kg (UK MAFF, 1978).
In a more recent survey, VC residues in various PVC samples were
given as follows: rigid water bottle (850 ppb); thin plasticized food
film (3 ppb); monopolymer powder (10 ppb); copolymer film (15 ppb)
(Poy et al., 1987). PVC film is still used widely for food packaging.
For example, in Denmark, in 1990, 129/239 samples of cling-film used
for cheese wrapping were of PVC (Svensson, 1994).
Residual VC could not be detected (< 0.1 ppm) in two PVC
products from Thailand (Smerasta et al., 1991) or in PVC and products
from it in Poland (Stareczek, 1988). PVC medical devices are regulated
in the USA and have to meet certain requirements (a maximum of 5 ppb
residual monomer for flexible compounds and a 10 ppb ceiling for rigid
compounds (Rakus et al., 1991)).
Levels of VC found in food and pharmaceutical articles are given
in section 5.1.4. Annex I gives current regulations for VC content in
various PVC products.
3.2.4.3 VC formation as a result of heating PVC
a) Thermal degradation of PVC
PVC is thermally stable below 225°C. Between 225°C and its
ignition temperature of 475°C, thermal degradation results in the
release of about 50 compounds (Boettner et al. 1969). PVC does not
degrade back to VC. Thermal degradation of two types of bulk PVC
samples at 148-232°C resulted in the release of long-chain aliphatic
alcohols, toluene, benzene, various chlorinated species, and a major
peak of HCl. The main components released at 260°C-315°C were aromatic
hydrocarbons such as benzene, phenol and adipates, along with various
aliphatic alcohols, alkenes, anhydrides, some of them chlorinated, and
carbon monoxide (Froneberg et al., 1982). When 1 kg of PVC is heated
to 300°C, it releases about 13 g HCl and 5 g CO.
b) Release of VC from heating PVC
Various PVC resins from different producers were tested for VC
evolution over the 130 to 500°C range using a heating rate of
10°C/min. A consistently low level of VC, amounting to 15-30 ppm
(based on resin), was found in the volatile decomposition products
from all of the samples tested, regardless of resin type or
manufacturing source (USA) (Wakeman & Johnson, 1978). A 100-mg sample
was programmed for heating from 200 to 450°C at 3°C/min; this resulted
in the formation of a total of 23.2 ppm VC, the major portion being
generated in the 275-350°C region. Dehydrochlorination occurred most
rapidly between 250 and 275°C. During this period only 2.3 ppm of VC
was formed. The VC evolved by heating PVC is the VC monomer entrapped
in the PVC resin.
At temperatures required for thermoforming PVC for food packaging
applications (90-120°C for a few seconds), no detectable VC was formed
in up to 1 h of exposure at 130°C (detection limit of analysis in air
- 1 ppb). Temperatures for calendering and extrusion operations are
175-210°C. Maximum VC levels determined at 210°C were 0.5 ppm (resin
basis) after 5 min and 1.2 ppm after 30 min (Wakeman & Johnson, 1978).
In more recent studies into VC formation during the thermal
welding of plasticized PVC sheeting (about 225°C), in normal field
situations such as piping in sewers, VC concentrations were usually
not above the detection limit of 0.05 ppm. Only where there was poor
ventilation were higher levels detected (0.2 ppm VC; 1.0-3.5 ppm HCl)
(Williamson & Kavanagh, 1987).
3.2.5 Other sources of VC
3.2.5.1 VC as a degradation product of chlorinated hydrocarbons
VC as a gas, in leachate and groundwater (see Table 4), has been
found in landfills and surroundings where there were no VC/PVC
production facilities in the vicinity. It was found that VC can be
formed, under anaerobic conditions, from the reductive halogenation of
the more highly chlorinated chloroethenes: tetrachloroethylene (PCE),
trichloroethene (TCE), and the dichloroethene isomers ( cis-1,2-DCE,
trans-1,2-DCE, and 1,1-DCE) (Parsons et al., 1984; Vogel & McCarty,
1985; McCarty, 1996, 1997, see Fig. 1). PCE and TCE are widely used as
industrial solvents in particular for degreasing and cleaning metal
parts and electronic components, and in dry cleaning. Production
levels for 1984 were 260 and 200 thousand tonnes for PCE and TCE,
respectively (Wolf et al., 1987). Careless handling, storage and
disposal, as well as the high chemical stability of these compounds,
have made them, and consequently VC, some of the most frequently
encountered groundwater contaminants (Arneth et al., 1988). Although
VC may be further degraded to less chlorinated and non-chlorinated
ethenes, and possibly finally to carbon dioxide and ethane, this
proceeds only at a slow rate under highly reducing conditions
(Freedman & Gossett, 1989; DiStefano et al., 1991; De Bruin et al.,
1992; see also section 4.2). As a consequence, VC can be detected in
landfill sites in and surrounding areas through spreading. Reports
from several countries show high levels of VC contamination of soil
and groundwater, aquifers and wells (see Table 4 and section 5.1).
There have recently been several field studies in
PCE/TCE-contaminated landfill sites and aquifers (Major et al., 1991,
1995; Fiorenza et al., 1994; Lee et al., 1995, see Table 5). These
have shown that under anaerobic conditions, PCE and TCE can be
intrinsically biodegraded to ethene by indigenous methanogenic,
acetogenic and sulfate-reducing bacteria. Furthermore, under aerobic
conditions there is a potential for direct or co-metabolic oxidation
of DCE and VC. Therefore, an efficient bioremediation of chlorinated
ethene-contaminated aquifers may occur in contaminant plumes
characterized by upgradient anaerobic and downgradient aerobic zones,
such as where anaerobic, chlorinated ethene plumes discharge to
aerobic surface water bodies. However, this depends on the ability of
the stream-bed microbial community to degrade efficiently and
completely DCE and VC over a range of contaminant concentrations (Cox
et al., 1995; Bradley & Chapelle, 1998a). It should be noted that this
bioremediation occurs under specific conditions. The biodegradation
studies listed in chapter 4 give conflicting results.
Table 4. Vinyl chloride found in landfill/waste disposal sites as a gas, in leachate and in groundwater
formed probably from degradation of higher chloroethenes
Sample Place (year) of sampling Valuea Concentrations Reference
Landfill gas 2 landfills USA max 230 mg/m3 Lipsky & Jacot (1985)
average 34 mg/m3
Landfill gas landfill, UK max 11 mg/m3 Ward et al. (1996)
plume, 100 m from boundary 40 mg/m3
due to subsurface migration (1991)
Landfill gas landfill, Braunschweig, Germany mean 9 mg/m3 Henning & Richter (1985)
Gas effluents garbage dump, Berlin, Germany 0.27 mg/m3 Höfler et al. (1986)
Gas Germany, average Janson (1989)
industrial waste disposal site 41 mg/m3
municipal waste disposal site 10 mg/m3
Gas Germany, waste disposal site range 0.03-0.3 mg/m3 Bruckmann & Mülder (1982)
Landfill gas UK, 7 waste disposal sites range < 0.1-87 mg/m3 Allen et al. (1997)
Soil air Germany, solvent waste sites 3 max out 128 mg/m3, Köster (1989)
of 200 47 mg/m3,
5 mg/m3
Leachate MSW, Wisconsin, USA (1982) range 61 µg/litre Sabel & Clark (1984)
Leachate USA sites established before 1980 range 8-61 µg/litre Chilton & Chilton (1992)
(6 chosen sites)
Leachate or industrial landfill range 140-32 500, Brown & Donnelly (1988)
groundwater municipal landfill 20-61 000
plume µg/litre
Groundwater, Germany, contaminated water range < 5-460 µg/litre Brauch et al. (1987)
Wells range 15-1000 µg/litre
Groundwater Germany, solvent waste site 3 max 1000 µg/litre, Köster (1989)
samples/200 500 µg/litre,
200 µg/litre
Table 4. (cont'd)
Sample Place (year) of sampling Valuea Concentrations Reference
Groundwater Germany max 120 µg/litre Milde et al. (1988)
Groundwater Santa Clara Valley, USA (near range 50-500 µg/litre Wolf et al. (1987)
plants manufacturing electronic
equipment which use
significant amounts of
chlorinated solvents)
Groundwater Germany: 136 samples from max 12 000 µg/litre Dieter & Kerndorff (1993)
down-gradient wells of 100 waste mean 1694 µg/litre
disposal sites
Groundwater sand aquifer near industrial max > 5 µg/litre Semprini et al. (1995)
site, Michigan, USA. Concentration at 10 m;
increased with depth consistent 56 400 µg/litre
with methane at 23 m
Outwash aquifer Gloucester landfill, Canada (1988) range < 1-40 µg/litre Lesage et al. (1990)
a This column indicates whether the concentration is a maximum (max), average or range value
Each landfill site has individual conditions (e.g., presence of
other solvents such as acetone and methanol), so that the degradation
rates cannot be directly compared. The most extensively studied site
of intrinsic chlorinated solvent biodegradation is the St Joseph
(Michigan, USA) Superfund site where groundwater concentrations of TCE
as high as 100 mg/litre have been found, with extensive transformation
to cis-DCE, VC and ethene. Conversion of TCE to ethene was most
complete where methane production was highest and where removal of
nitrate and sulfate by reduction was most complete (McCarty, 1996;
Weaver et al., 1996). At another site in the USA (Dover Air Force
Base), half-lives of 1 to 2 years have been estimated for each stage
in the reaction chain (e.g., DCE to VC; VC to ethene) (Ellis et al.,
1996). The degradability of chlorinated aliphatic compounds was
studied under methanogenic conditions in batch reactors with leachate
from eight landfill sites in Denmark. PCE and TCE were found to be
degraded in only three of the eight leachates, with significantly
different conversion rates. In one leachate, complete conversion of
chlorinated ethenes, including conversion of VC, was observed within
40 days, while another leachate showed only 50% conversion of PCE
(Kromann et al., 1998).
No known microorganism can aerobically destroy PCE. Laboratory
studies have shown that some anaerobic bacteria (e.g., Dehalobacter
restrictus ) use chlorinated solvents for respiration
(halorespiration), breaking them down in the process to form
cis-dichloroethene, although restricted diet conditions are
necessary (Sharma & McCarty, 1996). Recently, a coccoid bacterium has
been isolated (provisionally named Dehalococcoides ethenogenes
strain 195) which, together with extracts from mixed microbial
cultures, can dechlorinate PCE, removing further chlorine atoms to
form vinyl chloride and finally ethene (Maymó-Gatell et al., 1997).
Escape of landfill gas from the disposal site can take place via
the surface (emission) or into the ambient soil (migration). VC is
emitted from the landfill surface into the ambient air (Wittsiepe et
al., 1996). Awareness of this problem has encouraged the development
of in situ bioremediation of chlorinated solvents and VC using
anaerobic or aerobic co-metabolic processes (Dolan & McCarty, 1995b;
Jain & Criddle, 1995; Semprini, 1995; see section 4.2).
The estimated emission of vinyl chloride from landfill sites in
the USA is 60-33 000 tonnes/annum (Lahl et al., 1991).
3.2.5.2 VC formation from tobacco
VC was identified in the smoke of all 13 cigarettes tested
(1.3-16 ng/cigarette) and in both small cigars tested (14-27
ng/cigar). The level correlates directly with the chloride content of
the tobacco. Filter tips with charcoal reduce selectively the VC
content of cigarette smoke (Hoffmann et al., 1976).
Table 5. Some examples of formation of VC through biodegradation of tetrachloroethene
in landfill sites (concentrations in mg/litre unless stated otherwise)a
Site Sample PCE TCE cis-DCE VC Ethene Reference
Chemical transfer groundwater 4.4 1.7 5.8 0.22 0.01 Major et al.
factory facility, downgradient well n.d. none 76 9.7 0.42 (1991)
North Toronto,
Canada
Carpet backing groundwater n.d. 56 4.2 0.076 Fiorenza et al.
manufacturing downgradient from n.d. 4.5 5.2 low (1994)
plant, Ontario, lagoon
Canada
Refuse landfills 7.15 5.09 not 5.6 McCarty &
(average of 8) (ppmv) (ppmv) measured (ppmv) Reinhard (1993)
Landfill groundwater 0.54 2.6 2.2 2.7 33 Lee et al. (1995)
well 3.4 14 44 54 43
15 270 140 48 14
Heavily polluted groundwater 20 70 20 2 Middeldorp et
site (solvent al. (1998)
distributor) in
Netherlands
a PCE = tetrachloroethene; TCE = trichloroethene; DCE = dichloroethene; n.d. = not detected
3.3 Uses
About 95% of the world production of VC is used for the
production of PVC. The remainder is used for the production of
chlorinated solvents, primarily 1,1,1-trichloroethane (10 000 tonnes
per year; European Council of Vinyl Manufacturers, 1994), via the more
toxic 1,1,2-trichloroethane and 1,1-dichloroethane.
VC was previously used as a refrigerant (Danziger, 1960) and as a
propellant in aerosol sprays for a variety of products, such as
pesticides, drugs and cosmetics (Wolf et al., 1987). These uses have
been banned since 1974 in the USA and in other countries.
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION AND TRANSFORMATION
4.1 Transport and distribution between media
Depending on the sources, VC can enter the environment via air,
water or soil. The most critical matrices are probably air and
groundwater. Euro Chlor (1999) calculated the partitioning of VC into
environmental compartments, based upon the Mackay Level 1 model, to be
99.99% air, 0.01% water, < 0.01% soil and < 0.01% sediment.
4.1.1 Air
Owing to its high vapour pressure (saturation vapour pressure
P0 > 10-4 mmHg; see also section 2), VC released to the
atmosphere is expected, based on calculations of Eisenreich et al.
(1981), to exist almost entirely in the vapour phase. The atmospheric
life-time of VC is limited by its reaction with photochemically
produced OH radicals (see section 4.2.2).
VC is volatilized to the atmosphere from area sources such as
landfill sites. Models have been developed to predict such short-range
dispersion. These models were validated by VC concentrations ranging
from < 5 to 31 µg/m3 (< 2 to 12 ppb) measured in the vicinity of a
landfill in Los Angeles 3 years after it last received any waste
containing VC (Chitgopekar et al., 1990).
No data are available on wet deposition.
In order to model distribution processes, liquid : air partition
coefficients have been determined (Gargas et al., 1989). Coefficients
of 0.43 and of 24.4 were obtained for 0.9% saline : air and for olive
oil : air, respectively.
4.1.2 Water and sediments
VC has a relatively low solubility in water (see section 2), and
the solubility can be increased by the presence of salts (see section
4.2.3).
Experimental data on adsorption to particulate matter in the
water column or to sediment are not available. A partition constant
(unitless) of 8.2 for a sediment-water system was calculated (from a
Kow value of 17) by Mabey et al. (1982), indicating a low adsorption
capacity. A high input of VC into water may lead to low-level
long-term storage in the associated sediment (Hill et al., 1976b).
Volatilization of VC acts as a significant transport mechanism.
It is considered to be the most rapid route for removal of VC from
surface water, but to be an unlikely pathway for disappearance from
groundwater that is not directly exposed to air (Smith & Dragun,
1984). Volatilization parameters such as vapour pressure and Henry's
Law constants indicate that VC is highly volatile. Another factor, the
reaeration rate ratio (rate constant for loss of VC from aqueous
solution divided by the rate constant for oxygen uptake by the same
solution), was reported to be 0.675 at 25°C (theoretical calculation
by Mabey et al., 1982) and approximately 2 (experimental measurement
by Hill et al., 1976b). The results of measurements or calculations of
volatilization half-lives of VC from water bodies are given in Table
6; they range from < 1 h to 5 h (measured in models for disturbed or
quiescent water) and from 2.5 to 43 h (calculated for natural water
bodies). In groundwater, however, VC may remain for months or years
(ATSDR, 1990).
More or less complex models have been developed to describe the
stability of VC in aquatic ecosystems (Hill et al., 1976b; Miller,
1992).
4.1.3 Soil and sewage sludge
Owing to its high vapour pressure, VC can be expected to
volatilize rapidly, especially from dry soil surfaces. No experimental
data are available. However, volatilization of VC from soil can be
predicted based upon its physicochemical properties. The amount of VC
volatilized from a soil depth of 1 m in 1 year was reported to range
from 16 to 45% in a sandy soil and 0.1 to 0.7% in a clay soil. These
calculations were based upon a Henry's Law constant of 0.44 and a
degradation half-life of between 30 and 180 days (Jury et al., 1992).
It has been observed at landfill sites that subsurface migration
of VC is a significant transport mechanism (Hodgson et al., 1992;
Little et al., 1992; Ward et al., 1996). Quantitative experimental
data on the potential of VC for gaseous subsurface migration (see
section 5.1.1) were not found.
Because of its solubility in water, VC can be leached through the
soil to groundwater. Additionally, the high solubility of VC in many
organic solvents may increase its mobility at special locations, e.g.,
landfills or waste disposal sites. Standard experimental studies on
soil sorption of VC are lacking. The soil adsorption coefficients of
VC were estimated from its water solubility, octanol/water partition
coefficient and from the molecular topology and quantitative
structure-activity relationship analysis method according to equations
given by Chiou et al. (1979), Kenaga & Goring (1980), Lyman et al.
(1982, 1990) and Sabljic (1984). The Koc values obtained ranged from
14 to 240, indicating a low adsorption tendency and therefore a high
mobility of VC introduced into soil (US EPA , 1985a; Stephens et al.,
1986; ATSDR, 1997).
A field study performed to determine non-extractable (bound)
residues (NER) of highly volatile chlorinated hydrocarbons gave a low
value of 2.4% for VC, as measured in a lysimeter after one growth
period in the upper (10 cm) soil layer (Klein et al., 1989).
VC was assumed not to appear frequently in sewage sludge due to
its low adsorption potential (log Kow < 2.0) and its high
volatilization tendency (Wild & Jones, 1992).
Table 6. Half-lives reported for volatilization of VC from water
Source Method Conditions Half-life Reference
Dilute aqueous experimental 22-25°C, Dilling et al. (1975);
solution (200 ml) (laboratory) rapid continuous 25.8 min Dilling (1977);
stored in an stirring
open container continuous stirring (200 rpm) 26-27.6 min Callahan et al. (1979)
discontinuous stirring approx. 80-90
(5% of time) min
quiescent, no stirring 290 min
Flowing channel experimental water input: 35 litre/second 0.9 h Scherb (1978)
(field) flow velocity: < 0.50 m/second
depth: 30 cm
Stream calculation based on Hc = 243 kPa.m3/mol 2.5 h Lyman et al. (1990)
depth: 1 m
flow velocity: 1 m/second
wind current: 3 m/second
calculation based on reaeration rate ratio of US EPA (1985a)
2; assumed oxygen reaeration ratesa:
Pond 0.008 h-1 43.3 h
River 0.04 h-1 8.7 h
Lake 0.01 h-1 34.7 h
a According to Tsivoglou (1967); Lyman et al. (1990)
4.1.4 Biota
VC has been identified in environmental samples of fish tissue
(see section 5.1.5) and in several species of aquatic laboratory
animals (molluscs, crustaceans, insects and vertebrates) and algae
experimentally exposed to VC-containing water (see section 4.3).
Reports on the presence of VC in terrestrial plants and animals have
not been found in the literature.
Further experimental data, e.g., in what way and to what extent
VC is capable of entering the biota, are lacking. However, a few
estimations have been made on the basis of the physicochemical
properties of VC. They refer to uptake of VC by terrestrial plants and
animals. Plant uptake was considered to be unlikely (Ryan et al.,
1988; Shimp et al., 1993) because, at an assumed half-life of
< 10 days (Ryan et al., 1988), VC should be lost from the system
rather than be taken up by the plant. Another study (Wild & Jones,
1992) screened organic contaminants for possible transfers into plants
and animals by summarizing approach data. Within the three categories
used (high, moderate, low potential) VC was classified in the
following way:
* retention by root surface: low;
* uptake and translocation: moderate;
* foliar uptake: high;
* transfer to animal tissues by soil ingestion: low;
* transfer to animal tissues by foliage ingestion: moderate.
4.2 Transformation
4.2.1 Microbial degradation
There have been many studies on the biodegradation of VC under
various simulated environmental conditions and these are listed in
Tables 8 to 10. The biodegradation studies have given contradictory
results, with no evidence of degradation under some aerobic conditions
such as surface water (Hill et al., 1976b) and sewage (Helfgott et
al., 1977) and under anaerobic conditions such as groundwater
(Barrio-Lage et al., 1990). Unacclimated biodegradation half-lives of
VC were generally estimated to be of the order of several months or
years (Howard et al., 1991). However, other studies reported complete
degradation in 3 months in simulated aerobic groundwater (Davis &
Carpenter, 1990). Where degradation is reported care must be taken to
ensure that the loss of VC is due to degradation and not from other
losses such as volatilization from the test system. Phelps et al.
(1991b) reported that > 99% of VC was lost from a bioactive reactor
compared to 60% from the control reactor.
Significant microbial degradation of VC under aerobic and
anaerobic conditions has been detected in studies using enrichment or
pure cultures isolated mostly from sites contaminated with different
organic chemicals (Tables 8 to 10). Vinyl chloride cannot be used by
most microorganisms as sole carbon source, but it can be
degraded/metabolized in the presence of propane, methanol,
3-chloropropanol, propylene, isopropene and glucose. However, in some
cases VC can even serve as sole substrate, as is seen with
Mycobacterium sp. (Table 9). The main degradation products include
glycolic acid or CO2 after aerobic conversion (Tables 7 to 9) and
ethane, ethene, methane or chloromethane after anaerobic
transformation (Table 10). Anaerobic mineralization of VC to CO2 has
been demonstrated (Table 10) and may occur under special conditions
(Bradley & Chapelle, 1998b).
A complete mass balance was given by degradation studies with
radiolabelled VC, for example, in aerobic resting cell suspensions of
Rhodococcus sp. A starting concentration of 1 mg [1,2-14C] VC
produced more than 66% 14CO2 and 20% 14C aqueous phase products, and
10% was incorporated into the biomass (Malachowsky et al., 1994).
Aerobic cultures of Mycobacterium aurum growing on a special filter
material were reported to mineralize VC quantitatively according to
the following equation:
VC + microorganisms -> biomass + HCl + CO2 + H2O (Meier, 1994).
The underlying reaction mechanisms for the aerobic and anaerobic
degradation of VC have been postulated to be oxidative and reductive
dehalogenations, respectively, involving a variety of pathways (Vogel
et al., 1987; Barrio-Lage et al., 1990; Ensley, 1991; Castro et al.,
1992a; Leisinger, 1992; Castro, 1993; Meier, 1994; Hartmans, 1995;
Jain & Criddle, 1995).
Frequently, the degradation reaction of VC proceeds more readily
with aerobes than with anaerobes (Tables 7 to 10). The reverse occurs
(Freedman & Gossett, 1989; Semprini et al., 1995) with PCE, an
important precursor of VC in the environment (see chapter 3). Thus,
two-stage treatment systems consisting of anaerobic (first stage) and
aerobic (second stage) cultures have been proposed to achieve the
complete degradation of a range of alkenes having different degrees of
chlorination (Leisinger, 1992; Murray & Richardson, 1993; Nelson &
Jewell, 1993). Recently, a chlorobenzoate-enriched biofilm reactor
using Desulfomonile tiedjei DCB-1 was developed which degraded PCE
under anaerobic conditions without any detectable VC remaining
(Fathepure & Tiedje, 1994).
There have been many efforts to use the VC-degrading capacities
of microorganisms in practical applications such as the purification
of waste gases (Meier, 1994) or of municipal waste waters (Narayanan
et al., 1995) and the remediation of landfill leachates (Lesage et
al., 1993), groundwaters (McCarty, 1993; Holliger, 1995) and
contaminated soils (Schulz-Berendt, 1993). Possible limitations arise
from physical (temperature, accessibility of substrate), chemical (pH,
redox state, concentration of VC and other contaminants, presence of
additional secondary substrates, salinity) and biological (presence of
predators, competition phenomena, adsorption of microbes to surfaces)
factors (Van der Meer et al., 1992).
Table 7. Aerobic degradation of vinyl chloride by mixed microbial consortia from different sites
Inoculum Test design/ Measured Initial Duration Efficiency of Reference
conditionsa parameter concentration degradation
Surface water room VC 20 ml/2.9 ml 41 h no degradation Hill et al. (1976b)
samples temperature
Mixed consortium 21°C; VC 20-120 mg/litre several no degradaion Hill et al. (1976b)
from natural + / - nutrients weeks
aquatic systems
Mixed consortium 20°C oxygen 25 days no degradation Helfgott et al.
from domestic + nutrients demand (1977)
sewage
Mixed consortium 25°C; 14CO2 0.05 mg/litre 5 days 21.5% degradation Freitag et al.
from activated + nutrients (1982, 1985)
municipal sewage
sludge
Naturally simulated 14CO2, 1 mg/kg 108 days > 99% degradation Davis &
occurring aquifers: VC soil-water 65% mineralization Carpenter
consortium from soil-water (CO2) (1990)
groundwater microcosms 0.1 mg/kg 109 days 50% mineralization
(prepared with soil-water (CO2)
sub-surface soil
and groundwater
(20°C)
Consortium aquifer sediment 14CO2, 17 µmol/litre 84 h 22-39% Bradley &
indigenous to microcosms VC mineralization Chapelle (1996)
anaerobic aquifer (CO2)
systems
(contaminated
with CHs)
Table 7. (cont'd)
Inoculum Test design/ Measured Initial Duration Efficiency of Reference
conditionsa parameter concentration degradation
Consortium creek bed sediment 14CO2, 0.2-57 µmol/litre 24 h 6.2-58% Bradley &
indigenous to microcosms VC mineralization Chapelle (1998a)
creek sediment (CO2)
(contaminated
with DCE)
Consortium from soil microcosms VC 5.3 mg/litre 95 h little change Dolan & McCarty
aquifer material in VC (1995a)
(from a concentration
VC-contaminated
site)
a CH = chlorinated hydrocarbons; DCE = dichloroethene
Table 8. Elimination of vinyl chloride in aerobic tests with mixed microbial consortia utilizing special substratesa
Inoculum Additional Test Efficiency of Remarks Reference
substrate degradation
Mixed methane laboratory removal of up inhibition by Fogel et al. (1986,
methanotrophs studies to 100% within methane and 1,1-DCE 1987); Strandberg et al.
4 h-30 days possible; toxic (1989); Uchiyama et al.
effects of VC and VC (1989); Nelson & Jewell
products possible (1993); Dolan & McCarty
(1995a); Chang & Alvarez-Cohen
(1996)
methane field study about 95% inhibition by Semprini et al. (1990, 1991)
(groundwater) in-situ methane possible
transformation
i.c. =
0.03 mg/litre
Mixed propane laboratory > 99% loss after > 60% loss in Phelps et al. (1991b);
microbial study 30 days control Lackey et al. (1994)
consortia i.c. =
4-20 mg/litre
methane laboratory 82 to > 99% Phelps et al. (1991b);
plus study loss after 10-21 Lackey et al. (1994)
propane days; i.c. =
1-20 mg/litre
Mixed methane laboratory 2.3 µmol VC per relative TCs from Dolan & McCarty (1995a)
methanotrophs plus study mg of cells during highest to lowest:
formate 26 h i.c. = trans-DCE;
14.8 mg/litre cis-DCE; VC;
TCE; 1.1-DCE
Table 8. (cont'd)
Inoculum Additional Test Efficiency of Remarks Reference
substrate degradation
Mixed methane laboratory removal of up to Deipser (1998)
consortia study > 99% within 15
days i.c. =
0.55 mg/litre
a i.c.= Initial concentration; DCE = dichloroethene; TCE = trichloroethene;
TC = transformation capacity
Table 9. Survey on isolated bacterial cultures capable of degrading vinyl chloride under aerobic conditions
Inoculum Additional Major degradation Remarks Reference
substrate producta
Mixed culture n. sp. CO2 (> 67%) Malachowsky et al. (1991)
consisting of
Rhodococcus rhodochrous
and 2 bacteria of the
order Actinomycetales
Bacterium of the order propane, CO2 (> 67%) Phelps et al. (1991a)
Actinomycetales glucose or
acetate
Alcaligenes denitrificans isoprene n. sp. Ewers et al. (1990)
ssp. Xylosoxidans
Methylosinus trichosporium methane n. sp. inactivation Tsien et al. (1989); Chang
OB3b possible & Alvarez-Cohen (1996)
methane glycolic Castro et al. (1992a)
acid (44%;
determined at
68% conversion)
Mycobacterium sp. ; no CO2 ; initially: VC as primary Hartmans et al. (1985, 1992);
M. aurum chlorooxirane substrate; Hartmans & DeBont (1992);
(epoxide) inhibition Meier (1994)
possible
Mycobacterium no CO2, HCl maximum growth Hauschild et al. (1994)
aurum L1 rates at
1 mmol
VC/litre
Mycobacterium vaccae propane n. sp. Wackett et al. (1989)
(JOB 5)
Table 9. (cont'd)
Inoculum Additional Major degradation Remarks Reference
substrate producta
Nitrosomonas europaea ammonia n. sp. Vannelli et al. (1990)
Pseudomonas sp. 3-chloro-propanol glycolic acid Castro et al. (1992b)
(71%; determined
at 25%
conversion)
Rhodococcus sp. propane CO2 (> 66%) Malachowsky et al. (1994)
Rhodococcus isoprene n. sp. Ewers et al. (1990)
erythropolis
Xanthobacter propylene n. sp. Ensign et al. (1992)
(strain Py 2)
a The degradation efficiency is given in parentheses; n. sp. = not specified
Table 10. Survey on anaerobic microbial degradation of vinyl chloridea
Inoculum Test design/ Conditions Major degradation Efficiency of Remarks Reference
products degradation
Mixed methanogenic liquid cultures n. sp. i.c. = 400 µg/litre Brauch et al.
consortium (from (20°C) groundwater (1987)
PCE-, TCE-, with sterile sand approx. 50% (100%)
VC-contaminated after 4 (11) weeks
groundwater)
without sterile sand approx. 20% (55%)
after 4 (11) weeks
Mixed consortium incubation of n.sp. 94% within 16 days Nerger &
(from groundwater plus i.c. = 18 µg/litre Mergler-Völkl
TCE-contaminated waste water (1988)
water) (9 : 1)( 21°C)
Mixed consortium digesters filled n.sp. low degradation Deipser (1998)
(from compost) with mature sieved (0.2 mg/m3 compost/h)
compost from private
households
Mixed consortium soil-groundwater (< 1% CO2) no degradation in 5 Barrio-Lage et
(from natural microcosms (25°C) months al. (1990)
sites) i.c. = 2 mg/litre
Mixed consortium flow-through column methane plus 89% degradation traces of Barrio-Lage et
(from natural packed with soil, ethene (82%), (in 9-15 days) chloromethane al. (1990)
sites) a. s.: mixture of CO2 (7%)
phenol, citrate,
ammonium
dihydrogenphosphate,
methanol and methane
Table 10. (cont'd)
Inoculum Test design/ Conditions Major degradation Efficiency of Remarks Reference
products degradation
Consortium aquifer sediment CO2 15-34% mineralization Bradley &
indigenous microcosms, anaerobic in 84 h (versus 3-5% Chapelle (1996)
to anaerobic conditions plus without Fe-EDTA
aquifer systems Fe(III) as Fe-EDTA amendment)
(contaminated i.c. = 17 µmol/litre
with CHs)
Mixed varying conditions chloroethane slow degradation Baek et al.
methanogenic or 15-30°C or ethene (1990); Carter &
methanol-enriched (plus ethane) Jewell (1993);
consortia Skeen et al.
(1995)
Enriched PCE- and batch cultures ethene partial to nearly inhibition by Freedman &
TCE-degrading (35°C) complete degradation PCE possible Gossett (1989);
consortia DiStefano et al.
(1991); Tandoi
et al. (1994)
Mixed anaerobic fixed-bed columnb ethene plus almost complete De Bruin et al.
consortia from 24°C a.s.: ethane conversion of (1992)
river sediment lactate PCE (95-98%)
and wastewater via VC
sludge
Methanobacterium resting cell no degradation Castro et al.
thermoautotrophicum suspensions (60°C) i.c. = 10-3 (1994)
mol/litre
Table 10. (cont'd)
Inoculum Test design/ Conditions Major degradation Efficiency of Remarks Reference
products degradation
"Dehalococcoides anaerobic H2-PCEb ethene 90% conversion decay in rate Maymó-Gatell
ethenogenes strain enrichment culture of PCE via VC of VC et al. (1995,
195"c + mixed conversion 1997)
microbial consortia
a a.s. = additional substrate; i.c. = initial concentration; n.sp. = not specified; CH = chlorinated hydrocarbons;
PCE = tetrachloroethene; TCE = trichloroethene
b starting material PCE
c preliminary name
4.2.2 Abiotic degradation
4.2.2.1 Photodegradation
Studies on the photodegradation of VC are summarized in Table 11.
They include direct and indirect photolysis.
VC in the vapour phase or in water does not absorb wavelengths
above 220 nm or 218 nm, respectively (Hill et al., 1976b). However,
solar radiation reaching the troposphere lacks wavelengths below about
290 nm due to the stratospheric ozone shield. So, direct photolysis of
VC is expected to be insignificant under environmental conditions,
because there is no overlap between the absorption spectrum of VC and
the sunlight radiation spectrum (Callahan et al., 1979). Consistently,
no photodegradation was observed with pure VC in the gas phase or in
water at wavelengths above 220 nm. After irradiation at 185 nm, VC was
photolysed (Table 11).
In the environment, indirect photolysis occurs and includes
reactions of VC in the presence of photosensitizers and those
(Table 12) with photochemically produced reactive particles.
A variety of photolytic products was formed after irradiation of
VC under several experimental conditions (Table 11). Some
intermediates, e.g., chloroacetaldehyde, were of considerable
photochemical stability. Therefore, the photooxidation is unsuitable
as a means of removing VC from waste gases (Gürtler et al., 1994). On
the other hand, treatment of water contaminated with VC and other
halogenated organic compounds by means of UV-enhanced oxidation
(UV/ozone or UV/hydrogen peroxide) was reported to be successful (Zeff
& Barich, 1992).
The atmospheric fate of VC depends on its reaction with reactive
particles such as free OH and NO3 radicals, Cl atoms, ozone and
singlet oxygen. As can be seen from Table 12, the reaction with OH
radicals is the dominant transformation process, showing calculated
tropospheric half-lives of 1-2 days or more. Factors influencing
indirectly (via OH radical concentration) the lifetime of VC are the
degree of air pollution and solar radiation, leading to spatial,
diurnal and seasonal variations (e.g., Hesstvedt et al., 1976).
Reaction products include formaldehyde (HCHO) and formyl chloride
(HCOCl), the latter being a stable potential toxicant (Tuazon et al.,
1988; Pitts, 1993).
Rate constants for the photodegradation of VC in aqueous
solutions have been reported to range from 6.99 × 109 mol-1 second-1
(Grosjean & Williams, 1992) to 7.1 × 109 mol-1 second-1 (Klöpffer et
al., 1985). Mabey et al. (1982) reported that photolysis of VC was not
an environmentally relevant process. They reported oxidation rate
constants of < 108 mol -1h -1 and 3 mol -1h -1 for reactions with
singlet oxygen (1O2) and peroxy radicals (RO2), respectively. On
the basis of an average OH radical concentration of 10-17 mol in
Table 11. Survey on vinyl chloride photolysis studies
Medium Irradiation Photolytic degradation Photolytic productsa Reference
(Duration)
VC in a high-vacuum medium pressure yes primary products: radicals Fujimoto et al. (1970)
system arc (C2H3, Cl); C2H2, HCl
VC in air sunlight (outdoors) yes n.sp. Pearson & McConnell
(half-life = 11 weeks ± 50%) (1975)
xenon arc (> 290 nm) yes CO (90%); HCl
VC in air high pressure yes (> 99% within 15 min) chloroacetaldehyde, Kagiya et al. (1975)
mercury lamp HCl, CO2
outdoors yes (55% within 2 days)
VC in air (dry) > 230 nm (4 h) yes chloroacetaldehyde (primary Müller & Korte (1977)
product), HCl, CO, formyl
chloride
VC in air < 400 nm (45 min) yes CO2, CO, H2O, HCl, HCOOH, Woldbaek & Klaboe
C2H2 (1978)
sunshine (3-45 h) very slow
VC (adsorbed on > 290 nm (n.sp.) yes n.sp. Freitag et al. (1985)
silica gel) (15.3% of applied amount)
VC in an oxygen 185 nm (up to 50 min) yes formyl chloride, Gürtler et al. (1994)
atmosphereb (quantum yield: 2-3) monochloroacetaldehyde,
acetylene, CO, CO2,
monochloroacetyl chloride,
HClc, formic acidc
254 nm (up to 6 h) no
VC in air xenon lamp yes n.sp. Haag et al. (1996)
(initial k: 0.09 second-1)
Table 11. (cont'd)
Medium Irradiation Photolytic degradation Photolytic productsa Reference
(Duration)
VC in air plus UV (> 290 nm) yes formic acid, HCl, CO, Cox et al. (1974);
nitrogen oxides (up to 22 h) (half-life = 1-7 h, + NO formaldehyde, ozone, Dilling et al. (1976);
half-life = 18 h, - NO) (other minor products) Gay et al. (1976);
Carassiti et al.
xenon lamp yes formaldehyde, Hcl (1977) Kanno et al.
(0-120 min) (1977)
< 400 nm (25 min) yes at low NO2 conc.: the same Woldbaek & Klaboe
(increase in reaction rate products as observed in air (1978)
as compared to NO2/NO (see above);
being absent) at high NO2 conc.:
additionally nitrosyl
chloride, N2O
VC in air plus xenon lamp yes n.sp. Haag et al. (1996)
1,1-DCE (initial k: 0.15 second-1)
VC in pure water > 300 nm no Hill et al. (1976b)
(10 mg/litre) (90 h)
VC in natural water > 300 nm no
samples (20 h)
(10 mg/litre)
VC in water plus > 300 nm yes (rapid) various products
photosensitizers
VC in PVC plant > 300 nm yes (half-life = 40 h) n.sp.
effluent sunlight (25 h) very little
a n.sp. = not specified; DCE = dichloroethene
b direct photolysis
c in the presence of water vapour
Table 12. Rate constants and half-lives for gas-phase reactions of vinyl chloride with OH radicals and other reactive particles
VC reaction Rate constant Temperature Assumed atmospheric Calculated Reference
with a (in units of (°C) c concentration of the half-life c,d
cm3/molecule-sec) b reactive particle c
* OH 5.6 × 10-12 27 1 × 106 molecules/cm3 1.4 days Cox et al. (1974);
(measured) US EPA (1985a)
* OH 4.5 × 10-12 23 (296 K) 1 × 106 molecules/cm3 1.8 days Howard (1976); US
(measured) EPA (1985a)
* OH 6.60 × 10-12 26 (299 K) n.sp. n.sp. Perry et al. (1977)
5.01 × 10-12 85 (358 K)
3.95 × 10-12 149 (423K)
(measured)
* OH 6.60 × 10-12 26 1 × 106 molecules/cm3 1.2 days US EPA (1985a)
(Perry et al., 1977)
* OH 6.60 × 10-12 room temperature 1 × 106 molecules/cm3 (3.5 days) Atkinson et al.
(Perry et al., 1977) (298 ± 2 K) (12-h daytime average, (1979, 1987);
Crutzen, 1982) Atkinson (1985)
* OH 6.60 × 10-12 room temperature n.sp. n.sp. Atkinson (1987)
(measured)
5.3 × 10-12
(calculated)
* OH 6.60 × 10-12 25 ± 2 5 × 105 molecules/cm3 (approx 3 days) Tuazon et al. (1988)
(Perry et al., 1977) (298 ± 2 K)
* OH 6.60 × 10-12 26 5 × 105 molecules/cm3 2.2-2.7 days BUA (1989)
(Perry et al., 1977) (Crutzen, 1982)
* OH 6.8 × 10-12 27 (300 K) 5 × 105 molecules/cm3 approx 2.4 days BUA (1989)
(Becker et al., 1984) (Crutzen, 1982)
Table 12. (cont'd)
VC reaction Rate constant Temperature Assumed atmospheric Calculated Reference
with a (in units of (°C) c concentration of the half-life c,d
cm3/molecule-sec) b reactive particle c
* OH 6.60 × 10-12 26 8 × 105 molecules/cm3 1.5 days Howard (1989)
(Perry et al., 1977)
* OH 4.0 × 10-12 26 (299 K) n.sp. n.sp. Kirchner et al.
cm3/mol-sec (1990)
(measured)
* OH 10.6 × 10-12 n.sp. n.sp. n.sp. Klamt (1993)
(calculated)
* OH n.sp. n.sp. 1 × 106 molecules/cm3 (42 h) Pitts (1993)
(24-h average)
* OH 6.60 × 10-12 n.sp. 6.5 × 105 molecules/cm3 (2.7 days) Helmig et al.
(Perry et al., 1977) (estimated global mean; (1996)
Tie et al., 1992)
* OH 6.60 × 10-12 26 7.5 × 105 molecules/cm3 1.6 days Palm (1997)
(Perry et al., 1977) (24-h average, BUA, 1993) personal
communication
* NO3 2.3 × 10-16 room temperature 2.4 × 109 molecules/cm3 (42 days) Atkinson et al.
(measured) (298 ± 2 K) (12-h nighttime average, (1987)
Platt et al., 1984)
* NO3 1.4 × 10-16 23 (296 ± 1 K) n.sp. n.sp. Andersson &
(measured) Ljungström (1989)
Cl 12.7 × 10-11 25 (298 ± 2 K) n.sp. n.sp. Atkinson & Aschmann
(measured) (1987); Grosjean &
Williams (1992)
Table 12. (cont'd)
VC reaction Rate constant Temperature Assumed atmospheric Calculated Reference
with a (in units of (°C) c concentration of the half-life c,d
cm3/molecule-sec) b reactive particle c
O3 1.9 × 10-19 25 n.sp. n.sp. Gay et al. (1976);
Singh et al. (1984)
O3 2.0 × 10-18 n.sp. n.sp. 4 days Hendry & Kenley
(1979);
ECETOC (1983)
O3 2.45 (± 0.45) × 10-19 room temperature n.sp. n.sp. Zhang et al. (1983)
(approx. 25)
O3 2.5 × 10-19 25 7 × 1011 molecules/cm3 (66 days) Atkinson & Carter
(Zhang et al., 1983) (Singh et al., 1978) (1984); Atkinson et
al. (1987)
O3 2.45 × 10-19 25 1 × 1012 molecules/cm3 33 days US EPA (1985a)
(Zhang et al., 1983)
O3 1.2 × 106 27 1.6 × 1012 molecules/cm3 4.2 days Lyman et al. (1982,
cm3/mole-sec 1990); US EPA (1985a)
O3 1.7 × 10-19 22 7 × 1011 molecules/cm3 67 days Klöpffer et al.
(1988)
O3 2.5 × 10-19 (calculated n.sp. n.sp. n.sp. Meylan & Howard
according to AOP) (1993)
19 × 10-19 (calculated n.sp. n.sp. n.sp.
according to FAP)
O (3P) 8.6 × 10 -13 25 2.5 × 104 molecules/cm3 373 days Sanhueza & Heicklen
(1975); US EPA
(1985a)
Table 12. (cont'd)
VC reaction Rate constant Temperature Assumed atmospheric Calculated Reference
with a (in units of (°C) c concentration of the half-life c,d
cm3/molecule-sec) b reactive particle c
O (3P) 5.98 × 10 -13 25 2.5 × 104 molecules/cm3 532 days Atkinson & Pitts
(1977); US EPA
(1985a)
a O(3P) = oxygen atom
b AOP = Atmospheric Oxidation Program (currently used by US EPA); FAP = Fate of Atmospheric Pollutants (part of US EPA's Graphical
Exposure Modeling system, GEMS)
c n.sp. = not specified
d Values in parentheses reported as 'lifetime'
natural water, the kOH rate constant resulted in a half-life of
approximately 110 days (US EPA, 1985a). Both of the other reactions
appeared to be negligible.
4.2.2.2 Hydrolysis
Observations on chemical hydrolysis of VC derive from experiments
with effluent water from a VC plant (pH = 4.3-9.4; 50°C; 57 h;
Callahan et al., 1979), water of different pH values (85°C; 27 h; Hill
et al., 1976b; Mabey et al., 1982), water saturated with O2 (85°C;
12 h; Hill et al., 1976b), water plus ethanol (120°C; Rappoport & Gal,
1969) and two natural water samples (pH = 6.1/4.2 for river/swamp
water, both at room temperature and at 85°C; 41 h; Hill et al.,
1976b). In all cases, no or only slow hydrolysis occurred. The
hydrolytic half-life was estimated to be < 10 years at 25°C (Hill et
al., 1976b). Hydrolysis experiments under strongly alkaline, high
temperature (and therefore environmentally irrelevant) conditions
resulted in a polymerization of VC (Jeffers & Wolfe, 1996).
4.2.3 Other interactions
Under experimental conditions, VC and chlorine in water form
chloroacetaldehyde, chloroacetic acid and other unidentified compounds
(Ando & Sayato, 1984).
Many salts have the ability to form complexes with VC, thus
possibly leading to an increase in its solubility (Callahan et al.,
1979).
4.3 Bioaccumulation
Owing to its high vapour pressure and low octanol/water partition
coefficient, VC is expected to have little tendency for
bioaccumulation (BUA, 1989). Theoretical calculations resulted in
bioconcentration factors (BCFs) of 2.8 (based on a log Kow of
approximately 0.9) and around 7 (based on a water solubility of
2763 mg/litre) in aquatic organisms (US EPA, 1985a). A BCF of 5.7
(based on log Kow = 1.23) was calculated by Mabey et al. (1982) for
aquatic microorganisms.
Experiments performed with 14C-VC (initial concentration:
250 µg/litre) in a closed laboratory model aquatic ecosystem gave the
following results: after 3 days at 26.7°C 34% of the 14C was found in
the water and 65% in the air. The organisms from different trophic
levels contained 14C residues (in VC equivalents, µg/kg) of 1307
(alga, Oedogonium cardiacum), 621 (waterflea, Daphnia magna), 123
(snail, Physa sp.), 1196 (mosquito larva, Culex pipiens
quinquefasciatus) and 312 (fish, Gambusia affinis) (Lu et al.,
1977). These values led to BCFs ranging from 3 to 31 when compared to
the VC water concentration of 42 µg/litre, indicating some
bioaccumulation, but no biomagnification within the food chain.
Another study determined BCFs for green algae (Chlorella fusca)
after a 24-h exposure to 0.05 mg VC/litre and for fish (golden ide,
Leuciscus idus melanotus) exposed to a constant average
concentration of 0.05 mg VC/litre over 3 days. BCFs of 40 and < 10,
respectively, were found (Freitag et al., 1985).
After 5 days of incubation of 0.05 mg VC/litre in activated
sewage sludge, an accumulation factor of 1100 (based on the
distribution of VC between sludge, dry weight and water) was observed
(Freitag et al., 1985).
4.4 Ultimate fate following use
4.4.1 Waste disposal
Several methods have been employed for removal of VC from waste
water: stripping with air, steam or inert gas (Nathan, 1978;
Cocciarini & Campańa, 1992; Hwang et al., 1992), extraction (Nathan,
1978) or adsorption onto activated charcoal or adsorbent resin
(Nathan, 1978; Dummer & Schmidhammer, 1983, 1984).
Like VC waste gases produced during other processes (see
chapter 3), the recovered VC can be recycled or incinerated (US EPA,
1982; BUA, 1989). Special biological filters have been developed for
degrading VC in waste emissions (Meier, 1994, 1996; see also section
4.2.1).
Incineration leading to the total destruction of VC requires
temperatures ranging from 450°C to 1600°C and residence times of
seconds for gases and liquids, or hours for solids (HSDB, 1995).
Photochemical oxidations (Topudurti, 1992; Zeff & Barich, 1992;
Berman & Dong, 1994; see also section 4.2.2) are further methods of VC
elimination. UV-enhanced oxidation (oxidants used: ozone, hydrogen
peroxide) was applied for purification of polluted waters, (e.g.,
waste, leachate, groundwater) (Zeff & Barich, 1992). Treatment with
sodium dichromate in concentrated sulfuric acid was recommended for
the destruction of small quantities of VC, for instance, from
experimental laboratories (HSDB, 1995).
Chlorinated volatile organic compounds (VOCs) can be removed from
drinking-water/groundwater by treatment with activated charcoal
(Schippert, 1987), air stripping (after water is pumped to the
surface) (Boyden et al., 1992) or by air sparging (applied in situ)
(Pankow et al., 1993). Recently, on-site and in situ bioremediation
techniques, which couple evaporative or other methods with microbial
treatment, have been developed for restoration of groundwater systems
(Roberts et al., 1989; Portier et al., 1992, 1993; Fredrickson et al.,
1993; McCarty, 1993; Lackey et al., 1994) or soils (Schulz-Berendt,
1993) contaminated with VC and other VOCs.
4.4.2 Fate of VC processed to PVC
Most of the VC produced is used for the manufacture of PVC (see
section 3) and will therefore be connected with the fate of PVC. PVC
and articles made from it can be disposed of in landfills,
incineration or feedstock recycling. While rigid PVC is an extremely
persistent material, flexible PVC may be less recalcitrant to
disintegration (Harris & Sarvadi, 1994). At incineration, PVC plastics
do not depolymerize to form VC (Harris & Sarvadi, 1994), but produce
volatile aliphatic hydrocarbons and volatile chlorinated organic
compounds (Nishikawa et al., 1992; see also section 3.2.4). There is
evidence for formation of PCDFs/PCDDs (Theisen et al., 1989, 1991;
IPCS, 1989).
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
5.1 Environmental levels
There is very little exposure of the general population to VC.
Concentrations of VC in ambient air are low, usually less than
3 µg/m3. Exposure of the general population may be higher in
situations where large amounts of VC are accidentally released to the
environment, such as a spill during transportation. However, such
exposure is likely to be transient. Near VC/PVC industry and waste
disposal sites, relatively much higher concentrations, up to
8000 µg/m3 and 100 µg/m3, respectively, have been observed. VC has
only rarely been detected in surface waters, sediment or sewage
sludges. Maximal VC concentrations in groundwater or leachate from
areas contaminated with chlorinated hydrocarbons amount to 60 000
µg/litre.
5.1.1 Air
5.1.1.1 Outdoor air
Atmospheric air levels of VC in rural/remote and suburban/urban
areas range from not detectable to 24 µg/m3 (Table 13). Higher values
were recorded in industrial areas, with maxima in the vicinity of
VC/PVC producing or processing plants, even at distances of 5 km. Peak
concentrations were as high as 86 mg/m3 (33 ppm) and 17 mg/m3 (7
ppm), measured, respectively, in 1974 in the USA and 1983 in China
(Table 13). In many countries, VC concentrations near plants have
decreased with time due to regulatory measures (Table 13).
In more recent years attention has been paid to the occurrence of
VC near waste disposal sites, where levels of up to 0.18 mg/m3
(70 ppb) have been detected (Table 13). This is much less than the
significant amounts of VC found in undiluted landfill gas (see
section 3). Bruckmann & Mülder (1982) assumed that gas discharges of
landfills are diluted by a factor of 104 when entering the
atmosphere.
VC emissions of < 0.5 µg/m3 (remote from VC plants) or of a few
µg/m3 (near VC plants) in Germany and the Netherlands were derived
from a computer model (Besemer et al., 1984; WHO, 1987; LAI, 1992).
Ambient air concentrations in three geographical areas of the USA were
computed to range from trace to about 3 µg/m3 (Pellizzari et al.,
1979).
Table 13. Vinyl chloride measured in ambient (atmospheric) aira
Country; site and Valueb Concentrations Reference
year of sampling (µg/m3)
Rural/remote areas single 0.1 Kruschel et
Canada: pine forest al. (1994)
near Barrie, Ontario
year n. sp.
Germany: Westerland, mean values 6.6-24 Bauer
Schwarzwald, (1981)
Lüneburger Heide,
Bayer Wald; prior to
1977
Germany: Taunus; mean value 0.1 Dulson
1975 (n = 4) (1978)
USA: rural Northwest (n = n. sp.) n.d. (< 0.013) Grimsrud &
USA (Pullman, Rasmussen
Washington); (1975)
1974-1975
USA: over ocean (n = n. sp.) n.d. (< 26) Lillian et
Sandy Hook, NJ al. (1975)
(3 miles offshore);
1974
Suburban/urban areas
Canada: urban (22) meanc 0.06 Dann &
and rural (1) sites; (n = 1370; 4% Wang (1992)
1989-1990 > detection)
Germany: Berlin range (n=78) n.d. (n.sp.) - 3.5 Dulson
(3 sites); 1977 means 0.3-0.4 (1978);
Lahmann
(1980)
Germany: Frankfurt/M. n = 1 2.6 Bergert &
(city); 1974 Betz (1976)
Germany: Frankfurt/M. mean (n = 16) 21.1 Arendt et
(suburban); 1975 al. (1977)
Germany: Merkenich annual mean 3.6 German Expert
(North-Rhine Group on the
Westfalia); 1980 Environment
(1988)
Table 13. (cont'd)
Country; site and Valueb Concentrations Reference
year of sampling (µg/m3)
Germany: Cologne means 0.5-15.3 Anon (1981);
(3 stations); (n = n.sp.) BUA (1989)
1979-1986
Germany: several mean values 0.2-11 Bouscaren
sites; year n. sp. et al. (1987)
Germany: Hamburg annual means 2.4-9.0 (including Bruckmann
(12 sites in the city); (n = 300) 1-butene) et al. (1988)
1986-1987
USA: New Jersey geom. mean 1981: 0 Harkov et
(3 urban sites: (3/113)d and 1982: 0 al. (1983,
Newark, Elizabeth, (0/105)d (detection limit: 1984)
Camden); 1981-1982 0.013)
USA (eastern): urban daily mean 0.62 Dann &
sites; 1989 (n=397; 2% Wang (1992)
> detection)
Industrial sites
Canada: Shawinigan range n.d. - 117 Thériault
(vicinity of VC et al. (1983)
polymerization
plant); year n. sp.
China: dormitories for n = 16×3 Zhao et al.
workers (50, 100 and (4 h/day for (1994)
1000 m from a VC 4 days at
polymerization plant) 3 stations)
1983 maxima 17 400 / 5900 / 1500
daily means 4810 / 1080 / 580
1984 maxima 7800 / 2500 / 1400
daily means 4080 / 720 / 500
1986 maxima 6600 / 2000 / 1100
daily means 3120 / 820 / 520
1988 maxima 12 700 / 1000 / 300
daily means 4430 / 320 / 200
1989 maxima 3100 / 700 / 300
daily means 760 / 220 / 170
Table 13. (cont'd)
Country; site and Valueb Concentrations Reference
year of sampling (µg/m3)
Finland: 1.2 km from Kinnunen
a PVC plant (1996)
1993 range of 0.1-1.3
monthly means
1994 range of 0.1-0.8
monthly means
1995 range of < 0.1-0.1
monthly means
range of single 0.1-11
measurements
(n approx. 4000)
1996 range of 0.1-0.4 Kinnunen
monthly means (1997)
range of single 0.1-13
measurements
(n approx. 7000)
Germany: n.sp.; maximum 252 Bouscaren
prior to 1980 et al. (1987)
Germany: Ruhrgebiet 99th percentile 69 BUA (1989)
West; year n. sp.
Germany: 1978-1982 max. (n = n.sp.) 113 Bauer (1981)
Germany: Marl (80 m mean (n = 24) 213 Rohrschneider
and 500 m above mean (n = 9) 108 et al. (1971)
chemical plant);1970
Germany: Frankfurt/M. n.sp. 4.5 Atri (1985)
(industrial area);
year n.sp.
Netherlands: > 600 m maximum ca. 600 Besemer
from VC plant; mean 210 et al.
1976-1977 (1984)
0-500 m distant range (n = 200) < 21-504
500 m distant; 1978 range (n = 100) 13-55
mean 18
Netherlands: range 7.8-181 Guicherit &
VC/PVC plants; (n = n.sp.) Schulting
1980 (1985)
Table 13. (cont'd)
Country; site and Valueb Concentrations Reference
year of sampling (µg/m3)
UK: 5 VC plants; overall means < 13-228 Turner et al.
1984 (April-July) daily means 23-218 (1984)
(ranges related
to the 5 plants,
total n = 440)
Plant VI (7 stations, max. (n = 28); 20 202 (0 km),
0-4.8 km distant) 880 (0.8 km)
Plant IX (15 stations, means (n = 180); 311-1373
0.8-5 km distant) maximum 8806 (5 km)
USA: 1974-1975; VC (n = 708) US EPA
plant (Narco) means 3.4-262 (1975);
(4 stations, geom. means 0-28 Dimmick
< 250 - > 1000 m maximum 27 045 (< 250 m) (1981)
distant)
PVC plant (Aberdeen) (n = 438)
(4 stations, means 11.6-958
< 250 - > 1000 m geom. means 3.7-372
distant) maximum 23 430 (300 m)
PVC plant (Louisville) (n = 712)
(4 stations, means 11.4-101
< 250 - > 1000 m geom. means 5-30
distant) maximum 814 (250-400 m)
USA: residential areas range (n = 30) n.d. - 104 US EPA
near chemical plants (1975)
(n = 15); 1974
USA: Houston, Texas; range (n = 18.) 8-3238 Gordon &
1974 Meeks
(1977)
USA: residential max. (n = n.sp.) > 2560 Fishbein
areas in the vicinity mean (n = n.sp.) 44 (1979)
of VC/PVC plants;
prior to 1975
USA: vicinity of max. 34 McMurry &
VC/PVC plants (Texas, Tarr (1978)
7 stations); 1977
Table 13. (cont'd)
Country; site and Valueb Concentrations Reference
year of sampling (µg/m3)
Areas in the vicinity
of waste sites
Belgium: Mellery; prior no details 19 Lakhanisky
to 1990 et al. (1993)
Germany: surroundings no details < 3 Pudill (1993)
of a hazardous waste
site; about 1988
Germany: 2 landfills range < 0.082-0.65 Wittsiepe
near Bochum; 1990-1991 (n = 16) et al. (1996)
USA: West Covina, range (24-h av; 26-104 Camarena &
California; 1981-1984 n = 32) max 130 Coy (1984)
USA: West Covina, 24-h averages 13-31 (n = 5 days) Baker &
California; 1984 5-day average 23 MacKay
site A (200 m distant 24-h averages 5.7-18 (1985)
from landfill) 5-day average 10.4 (n = 5 days)
site B (20 m distant
from landfill)
USA: southern max. 18 Stephens
California, 1981 range 5-18 et al. (1986)
after 1981 (6 stations)
USA: California; max. 30 Little et
prior to 1990 al. (1992)
USA: New York range n.d. (5.4) - 16 Lipsky &
(2 sites) 1982 Jacot
(1985)
Miscellaneous
USA: all outdoor site mean 8.5 Shah &
types; prior to 1988 (n = 574) Singh
(1988)
a n.d. = not detected (detection limit in parentheses, if specified);
n.sp. = not specified
b This column indicates whether the concentration is a maximum (max),
average or range value
c Values below detection set to 0.5 method detection limit (0.1 µg/m3)
d Values in parentheses = no. detected / no. sampled
5.1.1.2 Indoor air
Indoor air concentrations of VC in houses near landfills in the
USA reached concentrations of up to 1 mg/m3 air (Little et al., 1992:
2 landfills, maxima of 0.13 and 0.3 mg/m3; Stephens et al., 1986:
1 landfill, maximum of 1 mg/m3), thus exceeding the maximum outdoor
levels reported in Table 13 for areas adjacent to landfills. Moreover,
the Californian monitoring programme, collecting a total of 500 air
samples at two outdoor and four indoor sites downwind of a landfill,
revealed that the 120 samples containing the most VC
(> 0.025 mg/m3; 10 ppb) were taken inside homes (Little et al.,
1992). It is assumed that in addition to atmospheric transport,
subsurface migration of VC accounts for the elevated indoor air levels
of VC (Wood & Porter, 1987; Hodgson et al., 1992; Little et al.,
1992).
A room being painted with a red latex paint based on a terpolymer
of vinyl chloride, vinyl acetate and ethylene showed VC levels of
75 and 10 µg/m3 (29 and 4 ppb), respectively, during and some time
(less than one day) after painting (Going, 1976).
In the early 1970s it was investigated whether VC was present in
car interiors as a result of volatization from PVC. Measurements of VC
concentrations in the interior of seven different new 1975 automobiles
gave positive results for two of them. Levels of 1036 to 3108 µg/m3
(0.4 to 1.2 ppm) were detected (detection limit: 130 µg/m3; 0.05 ppm)
(Hedley et al., 1976). Another study (Going, 1976) did not find VC in
the interior ambient air of 16 new and used cars and 4 mobile homes
(detection limit: 26 µg/m3; 10 ppb). It should be noted that since
this time the levels of VC in PVC resins have been drastically reduced
(see section 3.2.4).
5.1.2 Water and sediment
Owing to its high volatility, VC has rarely been detected in
surface waters. The concentrations measured generally do not exceed
10 µg/litre, with a maximum of 570 µg/litre from contaminated sites
(Table 14).
Much higher levels of up to 56 000 µg/litre have been found in
groundwater samples from contaminated sites (Table 15).
The levels in drinking-water supplies ranged from not detected to
2 µg/litre in samples collected in 100 German cities in 1977. In a
state-wide USA study performed in 1981-1982, random samples (taken
from randomly selected water systems) had concentration ranges of n.d.
to 1.1 µg/litre, and non-random samples (taken from systems that were
likely to be contaminated with VOCs) varied from n.d. to 8 µg/litre.
Prior to 1980 single maximum values of up to 380 µg/litre were
reported from the USA (Table 16).
Table 14. Vinyl chloride concentrations measured in surface watera
Country, source Year of sampling Valueb Concentrations Remarks Reference
(µg/litre)
Germany
River Rhine (prior to) 1978 typical conc. 1 Anna & Alberti (1978)
(n = many)
River Rhine 1982 < 0.2 Malle (1984)
River Rhine 1990 range (n = 78) < 0.01-0.031 Wittsiepe (1990)
Tributaries of Rhine (prior to) 1978 typical conc. < 1-5 Anna & Alberti (1978)
(Northrhine-Westfalia) (n = many)
Surface water from n.sp. range < 0.0004-0.4 > 150 samples Wittsiepe et al. (1990)
unspecified sites (n = n.sp.)
in former FRG
River Main 1990 range (n = 22) < 0.004-0.008 Wittsiepe (1990)
River Lippe 1989 range (n = 54) 0.12-0.4 receiving Wittsiepe (1990)
wastewater from
VC/PVC plants
River Ruhr (plus 1990 range (n = 60) < 0.0004-0.005 Wittsiepe (1990)
artificial lake) (lake: up to
0.06)
River Wupper 1989 range (n = 36) up to 0.069 Wittsiepe (1990)
River Saale 1990 range (n = 4) up to 69 receiving Wittsiepe (1990)
wastewater from
an industrial
area of the
former GDR
Table 14. (cont'd)
Country, source Year of sampling Valueb Concentrations Remarks Reference
(µg/litre)
Japan
Rivers in Osaka 1995 range (n = 28) up to 1.2 (3/28)c Yamamoto et
al. (1997)
USA
Delaware River 1976-1977 (n = 11) n. d. Sheldon & Hites
(1978)
Surface water from 1977-1979 maximum 566 (21/606)c Page (1981)
different sites (n = 606)
in New Jersey median 0
Surface water from n.sp. maximum 9.8 Burmaster (1982);
9 states (n = n.sp.) Dyksen & Hess (1982)
Final effluent from a 1980-1981 mean 6.2 Gossett et al. (1983)
waste-water treatment (n = 5)
plant in Los Angeles
Surface waters prior to 1984 median < 5 (63/1048)c Staples et al. (1985)
(n = 1048)
Indian River Lagoon n.sp. (n = n.sp.) n.d. (< 1.0) Wang et al. (1985)
(near water discharge
of VC), several
stations
Surface water near 1985-1990 (n = n.sp.) n.d. Hallbourg et al. (1992)
3 landfills in
Florida
Table 14. (cont'd)
Country, source Year of sampling Valueb Concentrations Remarks Reference
(µg/litre)
Surface water at a 1989-1990 range 0.23 Chen & Zoltek (1995)
landfill in Florida 1992-1993 (n = 5) n.d.
(Orange County)
a n.d = not detected (detection limit in parentheses, if specified); n.sp. = not specified;
GDR = German Democratic Republic
b This column indicates whether the concentration is a maximum (max), average or range
value
c Values in parentheses = no. detected / no. sampled
One reason for the occurrence of VC in drinking-water may be that
residual VC can migrate from PVC pipes used in some water distribution
systems into the water flowing through them. This has been found out
by field (Dressman & McFarren, 1978) and experimental (Banzer, 1979;
Nakamura & Mimura, 1979; Ando & Sayato, 1984) studies. The extent of
leaching depended on the VC concentration in the pipe material. In the
field study the highest VC concentrations (1.4 µg/litre) consistently
occurred in water from new pipes, whereas the lowest level (0.03
µg/litre) was found in the oldest (9 years of age) distribution
system. VC concentrations in landfill leachate samples amounted to up
to 61 mg/litre (Table 17). A gross analysis of water (no
specification) available for the USA and based on 5553 observations
reported maximum and median concentrations of VC as high as
202.6 mg/litre and 107 µg/litre, respectively (US EPA, 1985a).
No VC was detected in urban stormwater run-off from 15 cities in
the USA (n = 86) during a monitoring project concerning priority
pollutants (Cole et al., 1984).
Generally, a time trend cannot be derived from the water analysis
data available.
Most sediment samples contain very low VC concentrations, even at
rather contaminated sites. Sediment was monitored for VC during
1981-1982 in Florida (USA) at several stations (n = 8) of the Indian
River Lagoon and a conveying canal. Although the latter received
discharged water having VC concentrations of 34-135 µg/litre, no VC
was detected in the sediment samples (3 from each station, collected
monthly over a year), the detection limit being 2 ng/g (Wang et al.,
1985). The same was true for surface water (Table 14) and oyster
(section 5.1.5) samples from this site. Sediment samples (n = 2) taken
near the discharge zone of a wastewater treatment plant in Los Angeles
County (California, USA) contained < 0.5 µg VC/kg dry weight. The
corresponding water concentration of VC was 6.2 µg/litre (Gossett et
al., 1983). A survey of 343 sediment samples from the USA gave a
median VC concentration of < 0.5 µg/kg dry weight (Staples et al.,
1985). However, higher VC concentrations were also reported. According
to US EPA (1985a), VC was detected in sediment samples (no further
details given) in the USA at levels ranging from 0-580 µg/kg (n = 649;
median = 23 µg/kg).
5.1.3 Soil and sewage sludge
5.1.3.1 Soil
Subsurface soil samples near the waste pit of an abandoned
chemical cleaning shop in southern Finland showed VC concentrations as
high as 900 mg/kg (Salkinoja-Salonen et al., 1995). After an
accidental spillage of VC into snow in 1980, VC levels as high as
500 mg/kg were measured in the soil at up to 2 m depth (Charlton
et al., 1983).
5.1.3.2 Sewage sludge
VC has been detected in municipal sewage sludges in the USA. The
concentrations ranged from 3 to 110 mg/kg dry weight (corresponding to
145 to 3292 µg/litre), being detected in 3 of 13 samples, with a
median concentration of 5.7 mg/kg dry weight (corresponding to
250 µg/litre) (Naylor & Loehr, 1982). Another study reported a mean
concentration of 35.4 mg/kg dry weight for 6 of 44 samples (Fricke
et al., 1985). A range of 8-62 000 µg/litre was found in 35 of 435 raw
sludge samples (Burns & Roe, 1982).
5.1.4 Food, feed and other products
VC is not a general contaminant of foodstuff and pharmaceutical
or cosmetic products, but it can be detected after contact of these
products with PVC packaging materials. The use of PVC as packaging
material for food, drink and drugs began in the early sixties (chapter
3), whereas legislative action for safeguarding consumers from
exposure to VC did not begin until the early seventies (starting with
a ban on the use of PVC containers for packaging alcoholic beverages
in the USA; Anon, 1973). Current EC and US FDA regulations on the
level of VC in PVC materials intended to come into contact with
foodstuffs are listed in Annex 1.
VC concentrations measured in PVC-packed food and drink of
several countries are compiled in Table 18. A maximum value of
20 mg/kg was found in liquors. Other positive samples included
vegetable oils (up to 18 mg/kg), vinegars (up to 9.8 mg/kg),
margarines (up to 0.25 mg/kg), fruit drinks (> 0.2 mg/kg) and bottled
water (< 0.6 µg/litre).
Retail surveys of foods showed a significant reduction in VC
levels and/or in the number of positive samples since 1974 (UK MAFF,
1978, 1984; van Lierop, 1979; Codex Committee, 1984).
The latest data were from the 1990s (Table 18) and refer to
bottled drinking-water. In addition to small amounts of VC, there were
also indications for the presence of possible reaction products of VC
with chlorine (Fayad et al., 1997).
Pharmaceutical and cosmetic products were less frequently
monitored. The highest concentration, amounting to 7.9 mg/kg, was
detected in mouthwashes (Table 19).
Reports on analyses of animal feed were not available.
The potential for leaching of residual VC from PVC packaging into
the contents has been demonstrated by a variety of product analyses
(Tables 18 and 19) and by experimental studies using food or food
simulants (Daniels & Proctor, 1975; Hocking, 1975; Tester, 1976;
Diachenko et al., 1977; Pfab & Mücke, 1977; UK MAFF, 1978; Chan et
al., 1978; vom Bruck et al., 1979; van Lierop, 1979; Benfenati et al.,
1991; Thomas & Ramstad, 1992). Altogether, the results indicated that
Table 15. Vinyl chloride concentrations measured in groundwatera
Country; source Year of Valueb Concentrations Remarks Reference
of groundwater sampling (µg/litre)
Canada
from a VC spill site 1980 maximum; 10 weeks 10 000 Charlton et al.
after the spill < 20 (1983)
beneath a landfill 1988 range (n = 37) < 1-40 (5/37)c Lesage et al.
near Ottawa (1990)
Finland
from village wells mid-1980 range 5-200 mg/litre Nystén (1988);
(contaminated by leakage of (including DCE) Salkinoja-Salonen
a waste liquor basin of et al. (1995)
VC/PVC industry - detected
in 1974)
Germany
from different sites mean values n.d. (< 5) - 460 Brauch et al.
(4 sites) (1987)
from different wells mean values 15-1040
of a large surface
contamination (5 wells)
from a catchment area of range (n = 30) < 1-120 Milde et al.
a water-works (1988); Nerger &
(TCE / PCE-contaminated) Mergler-Völkl
(1988)
contaminated by waste until 1988 range (n = 113) < 1-12 000 (14/113)c Schleyer et al.
disposal sites (92 mean 2700 (1988)
sites) median 475
Table 15. (cont'd)
Country; source Year of Valueb Concentrations Remarks Reference
of groundwater sampling (µg/litre)
from a site contaminated n.sp. (n = 3) corresponding air Köster (1989)
with chlorinated pockets in soil
hydrocarbons 1000 128 000 µg/m3
500 47 000 µg/m3
200 5000 µg/m3
from a site contaminated ca. 1989 max. (n = 5) 110 Leschber et
with chlorinated al. (1990)
hydrocarbons (in Berlin)
from a contaminated site n.sp. range (n = 3) 710-1670 Kästner (1991)
(in Braunschweig)
of a catchment area of a 1989 (n = n.sp.) up to 0.130 Wittsiepe et al.
waterworks (1990)
Contaminated by accidental 1989 (n = n.sp.) up to 3
spillage of TCE and PCE
Contaminated by waste n.sp. mean (n = 136) 1693 18% positive Dieter &
disposal sites (approx. max. 12 000 Kerndorff (1993)
100 sites)
USA
from different sites in 1977-1979 max. (n = 1060) 9.5 (4/1060)c Page (1981)
New Jersey
from 9 states max. 380 7% positive Dyksen & Hess
(1982)
monitoring wells near n.sp. max. 635 present in Stuart (1983)
industrial waste sites in 3 out of 9 wells
Connecticut
Table 15. (cont'd)
Country; source Year of Valueb Concentrations Remarks Reference
of groundwater sampling (µg/litre)
from Nassau County, 1980 range (n = >100) 1.6-2.5 Connor (1984)
Long Island
from Miami, Florida n.sp. mean (n = 3) 6.8 Parsons et al.
(1984)
from a TCE spill site in n.sp. mean (n = 3) 82 Parsons et al.
Vero Beach, Florida (1984)
monitoring wells near MSW n.sp. present (in Sabel & Clark
landfills in Minnesota 5/20 sites), (1984)
but not quantified
near 3 plants manufacturing n.sp. range (n = 3) 50-500 chlorinated White Paper
electronics equipment solvents stored in (1984) cited in
(Santa Clara Valley) underground tanks Wolf et al.
(1987)
near solvent recovery range (n = 4) Cline & Viste
facilities (1985)
Connecticut 1980 n.d. (< 10) - 2700
Wisconsin 1983 n.d. (< 10) - 210
from a residential area 1983 maximum 2800 contamination Andreoli
in Long Island, New York from dry-cleaning (1985)
(several wells) shop
near a landfill in (about max. 692 Shechter (1985)
New Jersey 1981) (n = approx. 100)
monitoring wells around 1985 max. 2600 seasonal, spatial Stephens et al.
hazardous waste landfill and analytical (1986)
in S. California variations
Table 15. (cont'd)
Country; source Year of Valueb Concentrations Remarks Reference
of groundwater sampling (µg/litre)
monitoring wells near 1990 range (n = 64) n.d. (1-2) - 12 (5/64)c EMO (1992)
an airforce base in Ohio
near 3 landfills in Florida 1985-1990 mean values < 0.22-26.5 Hallbourg et al.
(10 locations) (1992)
monitoring wells at a 1989-1990 range (n = 3) n.d. - 0.23 increase of VC Chen & Zoltek
landfill in central 1992-1993 4.8-48.4 with time, decrease (1995)
Florida (Orange County) in total VOC
from an industrial site; (n = n.sp.) 51-146 Topudurti (1992)
recycling operations
(1940-1987); California
near a former waste disposal (n = n.sp.) US EPA (1992)
facility in Wisconsin:
on-property max. 77
off-property max. 5
Contaminated by 1991 maxima < 5-56 400 Semprini et al.
TCE > 10 years before minima < 5-321 (1995)
(17 sites)
near Plattburg, New York n.sp. (n = 2) 8 and 384 Bradley & Chapelle
(2 locations) (1996)
a n.d. = not detected (detection limit in parentheses, if specified); n.sp. = not specified; TCE = trichloroethene;
PCE = tetrachloroethene; VOC = volatile organic compounds; max. = maximum; MSW = municipal solid waste
b This column indicates whether the concentration is a maximum (max), average or range value
c Values in parentheses = no. detected / no. sampled
Table 16. Vinyl chloride measured in drinking-water suppliesa
Country; source Year of sampling Valueb Concentration Remarks Reference
(µg/litre)
Germany
drinking-water supplies 1977 range n.d. - 1.7 Bauer (1981)
of 100 cities
drinking-water n.sp. (n = n.sp.) ca. 1.6 ng/litre Wittsiepe et al.
(no details) (1993)
USA
finished drinking-water (prior to) 1975 max. (n = n.sp.) 10 Fishbein (1979)
water supplies from (prior to) 1975 max. (n = n.sp.) 5.6 & 0.27
Florida and Philadelphia
raw drinking-water 1975-1979 2 positives 2.2 & 9.4 (2/13)c CEQ (1981)
(13 cities)
finished drinking-water 1 positive 9.4 (1/25)c
(25 cities)
drinking-water wells (prior to) 1980 max. (n = n.sp.) 50 CEQ (1981);
(state New York) Burmaster (1982);
Craun (1984)
drinking-water of (prior to) 1981 range 0.05-0.18 Kraybill (1983)
113 cities mean 0.052
drinking-water 1977-1981 range of trace - 380 (82/1288)c Cotruvo et al.
supplies (166) positives (1986)
Table 16. (cont'd)
Country; source Year of sampling Valueb Concentration Remarks Reference
(µg/litre)
finished water supplies 1981-1982 max.: Westrick et al.
(using groundwater random samples: 1.1 (1/466)c (1984)
sources) from 51 states (n = 466)
in the USA non-random 8.4 (6/479)c
samples:
(n = 479)
private wells in 1982 max. 4.5 (1/63)c Goodenkauf &
Nebraska (n = 63) Atkinson (1986)
a n.d. = not detected (detection limit in parentheses, if specified); n.sp. = not specified
b This column indicates whether the concentration is a maximum (max), average or range value
c Values in parentheses = no. detected / no. sampled
Table 17. Vinyl chloride concentrations measured in leachatea
Country; source Year of sampling Valueb Concentrations Remarks Reference
(µg/litre)
Canada
MSW landfill leachate 1988 n.sp. 14 Lesage et al. (1993)
from Guelph (Ontario) 1989 n.sp. 23
USA
landfill leachates n.sp. (n = 6) not quantified present in 1/6 Sabel & Clark (1984)
from Minnesota samples
landfill leachates prior to 1982 max. (n = 4) 61 (1/4)c
from Wisconsin
landfill leachates prior to 1985 range (n = 5) n.d. Cline & Viste (1985)
(municipal and (< 10) - 120
industrial)
landfill leachates: prior to 1988 range total number of Brown & Donnelly
(n = n.sp.) landfills: 58 (1988)
municipal 20-61 000
industrial 140-32 500
MSW leachates (several (prior to) 1988 range 8-61
states in the USA) (n = n.sp.)
median 40 (6/?)c Chilton & Chilton
(1992)
a n.d. = not detected (detection limit in parentheses, if specified); n.sp. = not specified;
MSW = municipal solid waste
b This column indicates whether the concentration is a maximum (max), average or range value
c Values in parentheses = no. detected / no. sampled
the amount of VC migrating into food or solvent was proportional to
the VC concentration in the PVC packaging, to storage time and to
increasing temperature.
Until 1974 the use of VC as a propellant in aerosol sprays was
allowed in the USA (chapter 3). Realistic application of such aerosol
products (hairspray, deodorant, insecticide, disinfectant, furniture
polish or window cleaner) resulted in high (ppm range) indoor air
concentrations of VC (Gay et al., 1975).
There have been no reports on VC levels found in food,
pharmaceutical or cosmetic products in recent years. This may be due
to regulatory measures for PVC packaging in several countries (see
Annex 1).
5.1.5 Terrestrial and aquatic organisms
Oysters ( Crassostrea virginica) from the Indian River Lagoon in
Florida (USA) contained no detectable levels of VC (detection limit:
0.4 ng/g). The samples were collected weekly (1981) and monthly (1982)
at three sites (Wang et al., 1985). VC was also not detected in the
corresponding water and sediment samples (section 5.1.2).
Another study (Gossett et al., 1983) did find low concentrations
of VC (< 0.3 µg/kg wet weight) in a sample of small invertebrates
from just above the bottom sediments (n = 1), in a muscle sample of
shrimp (n = 1) and in liver samples of several fish species (n = 4).
The animals were collected in 1981 in final effluent waters from a
wastewater treatment plant in Los Angeles County (Palos Verdes,
California, USA) that had VC concentrations of 6.2 µg/litre (section
5.1.2).
Data on fish tissue (no further details given) available from the
US EPA STORET database were reported by US EPA (1985a). VC levels
ranged from 0-250 mg/kg (n = 530; median: 6 mg/kg).
5.2 General population exposure
Exposure of the general population to VC is possible by several
routes. They include inhalation of air polluted with VC (section
5.1.1), mainly in the vicinity of VC/PVC plants or waste disposal
sites, intake of contaminated drinking-water (sections 5.1.2 and
5.1.4), ingestion of food, beverages and medicines packed in PVC
(section 5.1.4), and absorption through skin from PVC-wrapped
cosmetics (section 5.1.4).
Normally, the general population is exposed to only small amounts
of VC, if at all. However, the exposure varies according to the
countries' regulatory measures, the occurrence of accidents or the
spread of precursor substances.
Table 18. Levels of vinyl chloride in food and drink packaged, stored or transported in PVC articles
Country Yeara Product No.a,b Concentrationsc Reference
(µg/kg)
Canada n.sp. alcoholic beverages 22 < 25-1600 Williams & Miles
vinegars 28 n.d. (10) - 8400 (1975)
peanut oil 10 300-3300
Canada n.sp. alcoholic beverages 10 n.d. (10) - 2100 Williams (1976a)
vinegars 10 300-7800
vinegars 9 14-9800 Williams (1976b)
sherry 3 500-2400
peanut oil 3 3800-18 000
Canada n.sp. oil 5 80-2100 Page & O'Grady (1977)
vinegars 5 10-5700
Canada 1981-1982 vinegars n.sp. 27-43 Codex Committee (1984)
other foods n.sp. < 10
Italy n.sp. drinking-water 10 0.013-0.083 Benfenati et al. (1991)
(PVC-bottled)
Netherlands 1975 oil samples 8 > 50 Van Lierop (1979)
salad dressing 1 250
margarine 4 60-250
ready-made salads n.sp. > 50
(whole
batch)
wine n.sp. 760
Netherlands 1976 margarine 1 60 Van Lierop (1979)
fish 1 90
biscuit 3 20-130
Table 18. (cont'd)
Country Yeara Product No.a,b Concentrationsc Reference
(µg/kg)
1977 peanut butter 1 4100 Van Lierop (1979)
(very small
individually wrapped
portions)
1978 various foods: 67 (3 +) n.d. (0.1) - 2 Van Lierop (1979)
soya oil 1 0.3
vinegar 1 0.6
salmon salad 1 2
Norway 1975 butter/margarine 16 n.d. (2) Ehtesham-Ud Din
salad 14 2-15 et al. (1977)
juices 8 6-25
vinegar 27 6-2790
mustard 5 9-14
Saudi Arabia n.sp. drinking-water 9 × 48d < 0.6 Fayad et al. (1997)
(PVC-bottled)
Sweden 1974 edible fats 127 n.d. (2) - 127 Fuchs et al. (1975)
Sweden 1975-1976 various foods 104 n.d. (2) - 600 Albanus et al. (1979)
(edible fats and oils, (mostly: < 10)
ketchup, vinegar, lime
juice, fruit syrup)
Switzerland 1973-1975 edible oils 41 n.d. (5) - 1750 Rösli et al. (1975)
United Kingdom n.sp. spirits n.sp. 0-250 Davies & Perry (1975)
Table 18. (cont'd)
Country Yeara Product No.a,b Concentrationsc Reference
(µg/kg)
United Kingdom 1974 fruit drinks 25 10 - > 200 UK MAFF (1978)
1977 13 < 10
1974 cooking oil 23 10 - > 200
1977 7 < 2
1975 butter, soft margarine 51 < 2 - 200
1977 9 < 2
USA 1971 vegetable oil 1 7000 Breder et al. (1975)
1974 3 700
USA 1973 spirits n.sp. up to 20 000 Anon (1973);
US FDA (1973)
USA 1973-1975 alcoholic beverages n.sp. 11 000-25 000 Codex Committee (1984)
a n.sp. = not specified
b + = number of samples positive
c n.d. = not detected (detection limit in parentheses, if specified)
d 9 brands (locally produced and imported)
Table 19. Vinyl chloride detected in pharmaceutical and cosmetic
products packaged in PVC materials
Product No.a Concentrationb Reference
(µg/kg or µg/litre)
Blood coagulant 30 n.d. (15) Breder et al.
solutions (1975)
Mouthwashes 11 n.d. (20-30) - 7900
Several capsules, tablets 11 n.d. (10-30) Watson et al.
and mouthwashes (1977)
Large volume 5 n.d. (0.1) - < 1 Watson et al.
parenterals (1979)
Mouthwashes 5 7-120
Shampoos 4 9-17
Body oils 4 n.d. (0.1) - 41
Intravenous solutions n.sp. n.d. (1) Arbin et al.
(many) (1983)
Cefmetazole sodium 4 n.d. (0.3) - <1 Thomas &
(Zefazone(R) sterile Ramstad
powder) (1992)
a n.sp. = not specified
b n.d. = not detected (detection limit in parentheses, if specified)
5.2.1 Estimations
Estimations of the respiratory intake of VC reported for the USA
ranged from 0 to 48.3 mg per person per day, based on exposure values
of 0 to 2.1 mg/m3 and assuming that 23 m3 of air are inhaled per day
(US EPA, 1985b). According to Seiber (1996) over 100 000 Californians,
particularly those living near landfills, may be exposed to VC levels
of 2.59 µg/m3 (1 ppb) or more.
A European study (Besemer et al., 1984) evaluating the exposure
of the Dutch population to VC in ambient air assumed an average
exposure of about 0.2 µg/m3, which resulted in a calculated daily
intake of 4 µg VC per person. Small fractions of the population were
concluded to be exposed to higher average levels: 0.01% (> 8.5
millions) to > 5 µg/m3, 0.04% to 4-5 µg/m3, and 0.05% to
3-4 µg/m3. The corresponding estimated daily intakes of VC were
> 100 µg, 80 µg and 60 µg, respectively.
Intake via drinking-water from public water supplies in the USA
was estimated to exceed 1 µg/litre for 0.9% of the population,
5 µg/litre for 0.3% and 10 µg/litre for 0.1% of the population. These
would result in daily VC intakes (assuming a 70-kg man and 2 litres of
water/day) of > 2, 10, and 20 µg/day. Maximal values were estimated
to be approximately 120 µg/day (US EPA, 1985b).
Another study (Benfenati et al., 1991) estimated the daily intake
of VC from PVC-bottled drinking-water bought in Italian supermarkets.
Based on the analytical results (Table 18) and assuming a consumption
of 2 litres of water per person per day and a storage time of 2 months
for the PVC bottles, the authors calculated that the oral intake could
exceed 100 ng VC per person per day.
Evaluation of the results of food surveillance programmes from
the United Kingdom led to a calculated maximum likely VC intake of
1.3 µg/day per person in 1974, of 0.1 µg/day per person in 1976, and
less than 0.02 µg/day per person by 1978 (UK MAFF, 1978, 1984). These
calculations were based on the typical daily consumption of fruit
drink, cooking oil and soft margarine.
Evaluating and summing up possible maximum user intakes of VC
from PVC-bottled liquor, wine and oil, and food packaged in PVC
materials, the US FDA calculated a maximum lifetime-averaged exposure
of 25 ng/person per day (US FDA, 1986). An earlier review reported an
estimated dietary intake of VC of 40 ng/day per person in the USA
(Codex Committee, 1984).
The intake by food and drinking-water in the Netherlands was
estimated (without specifying details) to be about 0.1 µg/day per
person or less (Besemer et al., 1984). Estimates (no details given) of
the average human intake for Switzerland were reported to be 3 ng/kg
body weight per day (Lutz & Schlatter, 1993).
5.2.2 Monitoring data of human tissues or fluids
Monitoring human tissues or fluids as an indirect measure of
exposure to VC has not frequently been applied. As with workers in the
plastics industry (section 5.3), the urine of premature babies was
found to contain large amounts of thiodiglycolic acid (thiodiacetic
acid), a metabolite of VC (chapter 6), but the relationship to
possible VC exposure was questionable (Pettit, 1986).
5.3 Occupational exposure
The main route of occupational exposure to VC is via inhalation
(Sittig, 1985), while dermal absorption is considered to be negligible
(ECETOC, 1988).
Industrial environments associated with VC exposure include VC
production plants, VC polymerization (PVC production) plants and PVC
processing factories. Estimates of numbers of workers exposed to VC
were, for example, in the USA (1981-1983) in the range of 80 000
(ATSDR, 1997) or in Sweden (1975-1980) more than 5000 (Holm et al.,
1982). Since, at the onset of VC/PVC production in the USA and Western
Europe, VC was not recognized as a toxic compound, no precautions
against contact were provided for nor was regular workplace monitoring
performed. Therefore, only sporadic measurements or retrospective
estimates (Table 20) of exposures are available for the period prior
to 1975. Published data from various countries on VC contamination of
workplace air throughout the early and later periods are compiled in
Table 21 and 22. Highest exposures occurred in the VC/PVC production
plants, with peak concentrations of several thousand ppm whereas much
lower exposure levels were measured in processing plants. Owing to
standard-setting and legislative regulations by national authorities
and technical improvements, levels dropped markedly to values of a few
ppm in many countries. Official exposure limits can serve as an
additional indication of approximate VC concentrations occurring in
plants in many countries. These limits have declined gradually (IARC,
1979). Generally, standards require that exposures do not exceed 13 to
26 mg/m3 (5 to 10 ppm) (Torkelson, 1994; Rippen, 1995; ACGIH, 1999).
However, even in the 1990s the standards were not always realized
in all countries (Table 21).
Whereas in industrialized countries factories that could not
satisfy the rigorous regulations of the early 1970s to reduce VC
emissions were forced to close down, in the countries of eastern
Europe and developing countries this was not possible for
socioeconomic reasons and large plants with old-fashioned technologies
continued to function (Hozo et al., 1996).
Table 20. Retrospective estimates of daily occupational
exposures to vinyl chloride prior to 1975a
Country Period VC exposure Reference
(mg/m3)
Germany "first years" > 2600 Szadkowski & Lehnert
prior to 1971 1300 (1982)
1971 260
1974 5.2-7.8
Norway 1950-1954 5200 Hansteen et al. (1978);
1955-1959 2600 Heldaas et al. (1984)
1960-1967 1300
1968-1972 260
1973-1974 207
Sweden 1945-1954 1300 Holm et al. (1982)
1955-1964 780
1965-1969 520
1970-1974 130
United Kingdom 1945-1955 2600 Barnes (1976);
1955-1960 1040-1300 Anderson et al. (1980);
1960-1970 780-1040 Purchase et al. (1987)
mid 1973 390
1975 13
1940-1955 1300-2070 Jones et al. (1988)
1956-1974 390-1300
USA 1945-1955 2600 Wu et al. (1989)
1955-1970 780-1300
1970-1974 260-520
1975 < 2.6-13
a Exposure levels during autoclave cleaning may have been as high as
7800 mg/m3 (Barnes, 1976)
Table 21. Levels of vinyl chloride reported for workplace air samples in VC/PVC production plants
Country Workplace Yearc Concentrations reported Reference
(mg/m3)c
China PVC production plant n.sp. 30-210 Bao et al. (1982)
Croatia plastics industry n.sp. mean = 13 Fucic et al. (1990a)
5200 (occasional peak)
Croatia VC/PVC plant 1949-1987 mean = 543 Hozo et al. (1996,
up to 1300 (occasional peak) 1997)
Former n.sp. n.sp. 2-41 Hrivnak et al. (1990)
Czechoslovakia
Egypt VC/PVC plant n.sp. 0.05-18 (8-h TWA) Rashad et al. (1994)
Finland PVC production plant, n.d. range
breathing zone 1981-1985 1.6 < 0.3-57 Viinanen (1993)
concentrations (TWA) 1986-1989 1.6 < 0.3-46
1993 0.3 < 0.3-26
France PVC production plant 1977-1978 2.3-7.3 (range of monthly Haguenoer et al.
means) (1979)
Germany PVC production 1974 < 65-181 Fleig & Thiess (1974)
department
PVC production plant 1977 1.3-91 German Environmental Office (1978)
PVC production plant 1979 12 (12-h TWA; stationary); Heger et al. (1981)
15.5 (12-h TWA; personal)
Germany 24 plants 1981-1984 3% of 33 samples: > 5 Coenen (1986);
(90 percentile: < 1); BIA (1996)
(shift means)
Table 21. (cont'd)
Country Workplace Yearc Concentrations reported Reference
(mg/m3)c
46 plants 1989-1992 all of 117 samples: < 5
(90 percentile: < 0.1);
(shift means)
Italy VC/PVC plants 1950-1985 < 13- > 1300 Pirastu et al. (1991)
The Netherlands PVC plant 1976-1977 2.6-26 (8-h TWA) De Jong et al. (1988)
Norway PVC plant 1974 65 Hansteen et al.
(1978)
Poland VC/PVC plant 1974 (30-600)a Studniarek et al.
(several departments) 1975 (30-270)a (1989)
1976 (15-60)a
1977 (6-150)a
1978 (1-30)a
1979 (1-15)a
1981 (0.1-36)a
1982 (0.1-12)a
(autoclave cleaners) 1974 (990)a
1982 (9-180)a
breathing zone of VC 1986 21.3 Dobecki &
synthesis mechanic 1987 66.9 Romaniwicz (1993)
1988 43.7
1989 0.7
1990 0.2
Romania PVC production plant 1965-1967 112-554 Anghelescu et al.
(1969)
Russia VC/PVC plant Gįlikovį et al. (1994)
16 probes (whole plant) 1990-1993 1-9 (range of annual means)
under the reactor 1990-1993 up to 200 (range of annual
means)
Table 21. (cont'd)
Country Workplace Yearc Concentrations reported Reference
(mg/m3)c
in compressor room 1990-1993 up to 400 (range of annual
mean)
Singapore PVC production plant 1976
after 1983 2.6-54 (15.3)a Ho et al. (1991)
up to 26 (short-term) (3.9)a
Sweden PVC production plant 1974-1981 0.26-114 (8-h TWA) Holm et al. (1982)
PVC production plant 1974-1980 0.26-5.7 (6-h TWA)
Taiwan PVC plants (n = 5): n.sp. Du et al. (1996)
15 different operation range (n=114): n.d.
units (e.g. outside (0.13) - 1009 range (n=4):
reaction tank)b 6-1009 (mean: 296; median: 86)
15 different job titles range of TWA (n=85):
(e.g. tank supplier)b n.d. - 3680 range (n=9):
5.7-3680 (mean: 660; median:
23.7)
United Kingdom PVC production plant "early days" 7800 Barnes (1976)
(full-time autoclave
cleaner)
USA PVC plant 1950-1959 up to 10 400; 13-2140 Ott et al. (1975)
(8-h TWA)
1960-1963 up to 1300; 13-620 (8-h TWA)
PVC plant n.sp. up to 650 (weekly TWA) Baretta et al. (1969)
USA VC/PVC plants 1973 up to 390 (TWA); peaks Rowe (1975)
2600-10 400
Table 21. (cont'd)
Country Workplace Yearc Concentrations reported Reference
(mg/m3)c
Former USSR VC/PVC plants early 1950s 100-800 Smulevich et al.
(1988)
PVC producing plant 50-800 Filatova &
(occasionally 87 300) Gronsberg (1957)
Former PVC production plant 1974 > 195 Orusev et al. (1976)
Yugoslavia
a Concentrations in parentheses designate geometric means
b Showing highest mean VC concentration
c n.sp.= not specified; TWA = time-weighted average; n.d. = not detected
Table 22. Levels of vinyl chloride reported for workplace air samples in PVC processing plants
Country Workplace Year Concentrations reported Reference
(mg/m3)
China PVC processing plant n.sp. > 30 Bao et al. (1982)
Germany PVC processing department 1974 < 2.6-67 Fleig & Thiess (1974)
Germany polymer extrusion (17 plants) 1989-1992 all of 33 samples: BIA (1996)
< 8 (90 percentile:
< 0.15) (shift means)
Sweden PVC processing plant 1974 < 0.26-0.8 Holm et al. (1982)
Sweden PVC processing plant prior to 1975 >13 - > 26 (8-h TWA) Lundberg et al.
(1993)
Russia PVC processing plant prior to 1990 0.007-1.26 Solionova et al.
(rubber footwear plant) (1992)
Russia PVC processing plant prior to 1966 < 113.6 Bol'shakov (1969)
(synthetic leather plant)
United Kingdom PVC processing plants n.sp. 0.4-0.9 Murdoch & Hammond
(cable factories) (1977)
USA automotive assembly plant(s) 1970s 0.13-7.8 Nelson et al. (1993)
(2 personal samples)
Autoclave cleaners in Croatia were exposed to extremely high
concentrations of VC (between 1295 and 3885 mg/m3; 500 and 1500 ppm).
A retrospective investigation of exposure to VC has been conducted
with 37 autoclave workers (emptying and cleaning) in Split, Croatia,
who were exposed to VC in a suspension polymerization plant. The
investigation covered the period from 1969 to 1987, when the factory
was closed because of its VC emissions. At the beginning, measurements
were done by simple means (Draeger's tubes) and later, from 1980 on,
by infra-red spectroscopy. The 37 workers were exposed to average VC
concentrations of 1412 mg/m3 (543 ppm) (Hozo et al., 1996; Hozo,
1998).
6. KINETICS AND METABOLISM IN LABORATORY ANIMALS AND HUMANS
6.1 Absorption
Animal and human studies have shown that VC is readily and
rapidly absorbed. The primary route of exposure to VC is inhalation.
However, the net uptake after exposure by inhalation is only 30-40% of
inspired VC. This is due to the fact that VC is taken up rapidly until
it reaches a blood concentration in equilibrium with that based upon
inspired concentration and the blood-to-air partition coefficient.
Uptake then decreases to an amount sufficient to replace that
metabolized. The importance of metabolism was shown by Bolt et al.
(1977) who showed that uptake of VC ceased once equilibrium was
reached.
While uptake by the oral route is near 100%, any VC not
metabolized during first pass through the liver will be expired. Thus,
the net dose may be less than the uptake, especially at high doses
resulting in saturation of metabolizing enzymes.
6.1.1 Oral exposure
In a study reported by Watanabe & Gehring (1976) and Watanabe et
al. (1976a), male rats received a single dose by gavage of 0.05, 1.0,
20 or 100 mg/kg body weight and 14C-labelled VC and the excreted
radioactivity was determined for 72 h. As only 2.4, 2.2, 1.0 and 0.5%,
respectively, of the administered radioactivity (Table 23) was
recovered in the faeces (total recovery 91, 89, 81 and 82%,
respectively), it can be assumed that there is nearly complete
absorption of VC in the gastrointestinal tract. Likewise, oral
administration of 0.25 or 450 mg/kg body weight 14C-VC to male rats
resulted in excretion of only 4.6 or 0.7%, respectively (Table 23), of
the applied radioactivity via faeces 0-72 h after application (Green &
Hathway, 1975).
Similar results were obtained by Feron et al. (1981), who fed
rats with different amounts of VC monomer in PVC powder via the diet.
The average amount of VC detected in faeces was 8, 10 and 17% for oral
intakes of 2.3, 7.0 and 21.2 mg/kg body weight per day, respectively.
Since the VC excreted in the faeces was considered by the authors to
be still enclosed in PVC granules and was not bioavailable, it was
concluded that the available VC was nearly completely absorbed in the
gastrointestinal tract.
Studies conducted by Withey (1976) have shown the rapid
absorption of VC from the gastrointestinal tract in male rats after a
single gavage application of 44 to 92 mg/kg body weight in aqueous
solution. Highest blood levels were always measured within 10 min
after administration. After an oral dose of 10 mg VC/kg body weight,
peak levels in brain, liver, kidney and lung were measured 5 min after
dosing, indicating rapid absorption from gastrointestinal tract
(Zuccato et al., 1979).
Table 23. Excretion of radioactivity in rats (in % of applied dose) 72 h after a single oral dose of VCa
Dose in mg/kg VC in CO2 in Urine Faeces Carcass and Total recovery Reference
body weight exhaled air exhaled air tissues
0.05 1.4 9.0 68.3 2.4 10.1 91.2 Watanabe et al. (1976a)
0.25 3.7 13.5 75.1 4.6 n.g. n.g. Green & Hathway (1975)
1.0 2.1 13.3 59.3 2.2 11.1 88.8 Watanabe et al. (1976a)
20 41.6 4.8 22.6 1.0 11.0 81.0 Watanabe & Gehring (1976)
100 66.6 2.5 10.8 0.5 1.8 82.3 Watanabe et al. (1976a)
450 91.9 0.7 5.4 0.7 n.g. n.g. Green & Hathway (1975)
a n.g. = not given
6.1.2 Inhalation exposure
Bolt et al. (1977) blocked the metabolism of VC in male rats by
i.p. injection of 6-nitro-1,2,3-benzothiadiazole (50 mg/kg body
weight), a compound that inhibits most microsomal monooxygenases. The
rats were exposed to approximately 1.2 mg/m3 14C-labelled VC in a
closed system 30 min after the injection. VC was taken up by the
animals until equilibrium was reached between VC in the atmosphere and
in the organism after 15 min, suggesting a rapid uptake of VC.
Similarly, equilibrium blood levels were observed by Withey (1976) in
male rats 30 min after the start of exposure to 18 200 mg VC/m3 (head
only).
Bolt et al. (1976) measured the decline of 14C-VC in a closed
system due to uptake by 3 male rats using an initial concentration of
130 mg/m3. The radioactivity in the air of the exposure system
decreased with a half-life of 1.13 h. Calculation of the clearance of
VC revealed an absorption of inspired VC of about 40%. These results
are in accordance with the data of Hefner et al. (1975a), who reported
in kinetic studies on male rats a similar decline at the same exposure
concentration.
Krajewski et al. (1980) measured the retention of VC in the lungs
of five human male volunteers exposed for 6 h to 7.5, 15, 30 or
60 mg/m3 through a gas mask. The percentage of retention (mean 42%)
was independent of the VC concentration and reached the highest level
of 46% in the first 15 min of exposure.
Buchter et al. (1978) reported toxicokinetic experiments on
themselves. In an open system, about 3-5 min after the start of
inhalation of 6.5 mg VC/m3 in the exposure atmosphere, an equilibrium
was reached between VC inspired and VC expired; 26% (person A) and 28%
(person B) of the VC inspired were removed by the body, probably by
absorption and subsequent metabolism. Metabolism started within the
first minutes of exposure, suggesting fast absorption of VC.
Breath samples were measured of volunteers who had showered in
VC-contaminated well water (Pleil & Lindstrom, 1997). For a brief
10-min shower exposure of 25 µg/m3 (inhalation) and 4 µg/litre
(dermal contact in water), 0.9 µg absorbed dose of VC and a blood
concentration of 0.01 µg/litre was calculated.
6.1.3 Dermal exposure
Hefner et al. (1975b) exposed the whole body (excluding the head)
of male rhesus monkeys to 2080 (2.5 h) or 18 200 mg/m3 (2.0 h)
14C-labelled VC. Very little VC was absorbed through the skin.
Although minuscule, the quantity absorbed appeared
concentration-dependent. The major portion of the VC absorbed was
eliminated by the lungs. The authors argued that no significant
percutaneous absorption would be expected under occupational exposure
conditions.
6.2 Distribution and retention
Data from inhalation and oral studies on rats indicate rapid and
widespread distribution of VC. Rapid distribution of VC was also
reported in humans after inhalation exposure. Rapid metabolism and
excretion limit accumulation of VC in the body. Data on distribution
after oral exposure showed similar results. No studies are available
concerning distribution after dermal exposure. Under conditions of
blocked VC metabolism, VC has been found to accumulate in adipose
tissue.
6.2.1 Oral exposure
Green & Hathway (1975) investigated the distribution in the
tissues of young rats 0.25, 2 and 4 h after oral administration of
30 mg/kg body weight [14C]-VC using whole-animal autoradiography. At
15 min after gavage, most radioactivity was detected in the liver,
followed by the gut (no data about the stomach), and there were small
amounts in the lung and kidney. After 2 h, label was observed in the
liver, kidney, small intestine, stomach, skin, para-auricular region
(probably Zymbal glands), and there were small amounts in lung and
heart. A similar distribution was determined after 4 h, with
additional label in the thymus, salivary gland and Harders gland, but
no radioactivity in the stomach lumen.
The distribution of VC and VC metabolites remaining in the body
(11.1% of recovered radioactivity) 72 h after a single oral
administration via gavage of 1 mg/kg body weight 14C-VC was described
as follows (expressed as % of administered radioactivity per gram
tissue): liver 0.182, skin 0.076, carcass 0.046, plasma 0.053, muscle
0.031, lung 0.061, fat 0.045 (no data about amount in the kidney). A
similar distribution pattern was observed using doses of 0.05 or
100 mg/kg body weight (Watanabe et al., 1976a).
6.2.2 Inhalation exposure
Whole body autoradiograms of male rats showed radioactivity in
the liver, bile duct, digestive lumen and kidneys when animals were
exposed for 5 min to 52 g/m3 14C-labelled VC and sacrificed 10 min
after the end of the exposure period. Animals sacrificed 2-3 h after
the exposure period showed a wider distribution of radioactivity, with
most of the labelled substances being observed in the liver, urinary
system, digestive lumen, lacrimal glands, skin and thymus (Duprat et
al., 1977).
Buchter et al. (1977) pretreated male rats i.p. with 50 mg/kg
body weight 6-nitro-1,2,3-benzothiadiazole (to block VC metabolism)
and exposed the rats in a closed system for 5 h to VC concentrations
of between 65 and 26 000 mg/m3. Authors observed an equilibrium
between 14C-labelled VC in the gas phase and the exposed organism.
The distribution of VC was independent of the exposure concentration.
The following distribution of VC (labelled and unlabelled) in
different organs (expressed as mol VC in 1 g tissue per mol VC in 1 ml
air) was reported immediately after the exposure period: blood 0.65,
liver 0.62, spleen 0.59, kidney 0.59, muscle 0.68 and adipose tissue
8.3. These results indicated that unmetabolized VC is accumulated in
adipose tissue due to its lipophilicity. In contrast, without blockage
of VC metabolism, most radioactivity was detected in kidney and liver
(5 h, 260 mg/m3).
In similar studies Bolt et al. (1976) measured radioactivity
(representing mostly VC metabolites) in different organs immediately
after exposure of male rats to 14C-VC (initial concentration 130
mg/m3) for 5 h. Highest levels were detected in the kidney (2.13%;
expressed as % of incorporated VC per g tissue) and liver (1.86%),
followed by spleen (0.73%), muscle (0.32%), adipose tissue (0.22%) and
brain (0.17%).
Distribution pattern changed with longer post-exposure
observation periods. Watanabe et al. (1976b) reported the following
percentages of 14C activity (mostly VC metabolites) in rats per gram
tissue 72 h after inhalation exposure to 26 or 2600 mg/m3
14C-labelled VC, respectively, for 6 h: liver (0.139; 0.145), kidney
(0.079; 0.057), skin (0.072; 0.115); carcass (0.048; 0.049), plasma
(0.051; not detected), muscle (0.052; 0.038), lung (0.065; 0.046), fat
(0.026; not detected). There was no significant difference between low
and high dose. 72 h after the exposure period most of radioactivity
had been excreted; 13.8% (low dose), 14.5% (high dose) of total
recovered radioactivity remained in carcass and tissues (Table 25). A
similar distribution pattern were presented by Watanabe et al. (1978b)
using the same experimental design but rats exposed once or repeatedly
to 13 000 mg/m3. Furthermore, the authors detected no significant
difference between single and repeated exposure.
Ungvįry et al. (1978) presented evidence for the permeability of
the placenta to VC. After exposure of pregnant rats to 5500, 18 000 or
33 000 mg VC/m3 for 2.5 h on day 18 of gestation, VC was detected in
fetal (13, 23 and 31 µg/ml, respectively) and maternal blood (19, 32
and 49 µg/ml), as well as in amniotic fluid (4, 5 and 14 µg/ml).
Toxicokinetic experiments on humans (self-experiments) were
reported by Buchter et al. (1978). Using an open system with a
concentration of 6.5 mg VC/m3 inspired air, the concentration in the
expired air reached a constant concentration after about 5 min
exposure in subject A and 7 min in subject B, indicating the end of
the distribution phase.
6.2.3 Partition coefficients in vitro
In vitro studies using the vial equilibration method (3 h
incubation, blood and tissue homogenates from male Sprague-Dawley
rats) revealed the following VC partition coefficients for male rats:
blood/air 2.4; fat/blood 10.0; muscle/blood 0.4; liver/blood 0.7; and
kidney/blood 0.7 (Barton et al., 1995). In similar experiments
tissue/air partition coefficients were obtained for different rodent
species (Gargas et al., 1989; Clement International Corporation, 1990;
Table 24). These data suggested that the concentration of VC in
adipose tissue is higher than in other tissues. Furthermore, in all
species in which both sexes were tested, partition coefficients for
fat/air were greater in females than in males (Table 24).
6.3 Metabolic transformation
The main route of metabolism of VC in the liver into non-volatile
compounds after inhalative or oral uptake involves 3 steps: a) the
oxidation by cytochrome P-450 to form chloroethylene oxide (CEO, also
known as 2-chlorooxirane), a highly reactive, short-lived epoxide that
rapidly rearranges to form chloroacetaldehyde (CAA); b) the
detoxification of these two reactive metabolites as well as
chloroacetic acid, the dehydrogenation product of CAA, through
conjugation with glutathione catalysed by glutathione S-transferase;
c) the modification of the conjugation products to substituted
cysteine derivatives, which are excreted via urine. The main metabolic
pathways are shown in Fig. 2. At high dose levels the metabolism of VC
is saturable.
The first step in VC metabolism requires microsomal
mixed-function oxidases (cytochrome P-450 enzymes) together with
oxygen and NADPH as cofactors. This was confirmed by studies in vitro
(Barbin et al., 1975; Guengerich et al., 1979) and in vivo
(Reynolds et al., 1975a,b; Guengerich & Watanabe, 1979; Bartsch et
al., 1979; Guengerich et al., 1981). The major catalyst of the
oxidation is CYP2E1 in humans. This has been demonstrated by in vitro
systems using purified human CYP2E1 or by inhibition of catalytic
activity in human liver microsomes with rabbit anti-human CYP2E1. In
liver microsomes from uninduced rats, VC is activated solely by
CYP2E1, at concentrations ranging from 1 to 106 ppm in the gas phase,
according to Michaelis-Menten kinetics (El Ghissassi et al., 1998).
The following kinetic constants were determined:
Km = 7.42 ± 0.37 µmol/litre;
Vmax = 4674 ± 46 nmol.mg protein -1 min-1.
Comparison of the Vmax obtained in this study to the Vmax determined
in vivo in rats (Gehring et al., 1978; Filser & Bolt, 1979) shows
that virtually all the metabolic activation of VC in vivo occurs in
the liver.
Table 24. Tissue/air VC partition coefficients for rodent tissuesa
Species; strain Sex Blood/ Liver/ Muscle/ Fat/
air air air air
Rat; F344 male 1.60 1.99 2.06 11.8
female 1.55 2.05 2.39 21.1
Rat; CDBR male 1.79 3.0 2.18 14.6
female 2.12 1.66 1.28 19.2
Rat; Wistar male 2.10 2.69 2.72 10.2
female 1.62 1.48 1.06 22.3
Mouse; B6C3F1 male 2.83 n.g. n.g. n.g.
female 2.56 n.g. n.g. n.g.
Mouse; CD-1 male 2.27 n.g. n.g. n.g.
female 2.37 n.g. n.g. n.g.
Hamster; male 2.74 3.38 2.56 14.3
Syrian golden female 2.21 1.31 1.96 21.1
a Data from Clement International Corporation (1990); vial
equilibration method (Gargas et al., 1989), blood or tissue
homogenates incubated for 1-4 h until equilibrium was achieved,
as indicated by two consecutive time points without significant
difference; n.g. = not given
Table 25. Percentage of 14C activity eliminated during 72 h following
inhalation exposure to [14C]-vinyl chloride for 6 h in male ratsa
Exposure groups 26 mg/m3 2600 mg/m3 13 000 mg/m3
(number of animals) (4) (4) (2)
Expired as unchanged VC 1.6 12.3 54.5
Expired as CO2 12.1 12.3 8.0
Urine 68.0 56.3 27.1
Faeces 4.4 4.2 3.2
Carcass and tissues 13.8 14.5 7.3
a Expressed as percentage of the total 14C activity recovered
(similar experimental design; Watanabe & Gehring, 1976;
Watanabe et al., 1976b, 1978b)
Applying the pharmacokinetic model developed by Andersen et al.
(1987) to describe the metabolism of inhaled gases and vapours, the
uptake of VC by rats in vivo, as determined by Gehring et al. (1978)
and by Filser & Bolt (1979), could be accurately predicted.
Using S9 extracts from human liver samples, Sabadie et al. (1980)
observed a great interindividual variability in the capacity to
activate VC into mutagenic metabolites. This is in agreement with the
observation of Guengerich et al. (1991) who found that levels of
CYP2E1 varied considerably among individual humans. Sabadie et al.
(1980) noted that the average activity of human samples is similar to
that of rat samples.
Chloroethylene oxide (CEO) has a half-life of only 1.6 min at pH
7.4 and 37°C (Malaveille et al., 1975). It can spontaneously rearrange
to CAA (Barbin et al., 1975) or hydrolyse to glycolaldehyde
(Guengerich et al., 1979; Guengerich, 1992). The latter reaction can
also be catalysed by epoxide hydrolase (Fig. 2).
Evidence for the detoxification of reactive VC metabolites
through conjugation with hepatic glutathione catalysed by glutathione
S-transferase (GST) has been shown by measuring the decrease of the
hepatic non-protein sulfhydryl content (primarily glutathione) and the
excretion of thiodiglycolic acid via urine in rats after inhalative
exposure to high levels (> 260 mg/m3) of VC (Watanabe & Gehring,
1976; Watanabe et al., 1976c, 1978a; Jedrychowski et al., 1984).
The conjugation products S-carboxymethyl glutathione and
S-formylmethyl glutathione are excreted in the urine of animals as
substituted cysteine derivatives [ N-acetyl- S-(2-
hydroxyethyl)cysteine, S-carboxymethyl cysteine and thiodiglycolic
acid] and the metabolite CO2 in exhaled air (Green & Hathway, 1975,
1977; Watanabe et al., 1976a,b; Müller et al., 1976; Bolt et al.,
1980). Thiodiglycolic acid has been detected in the urine of workers
occupationally exposed to 0.36-18.2 mg/m3 (Müller et al., 1978).
An alternative pathway has been suggested from inhibition studies
of VC metabolism with ethanol (Hefner et al., 1975c). Pretreatment of
rats with ethanol (5 ml/kg body weight) significantly reduced the
depression of the concentration of non-protein sulfhydryl in the liver
caused by exposure to 2780 mg VC/m3 for 105 min. This inhibition was
less pronounced in rats exposed to < 260 mg/m3. It is postulated
that at low concentrations, a sequential oxidation to 2-chloroethanol,
2-chloroacetaldehyde and 2-chloroacetic acid, involving alcohol
dehydrogenase, takes place. It is speculated that ethanol inhibits
specific P-450 enzymes. However, this hypothesis has not been
substantiated by further experimental data and this pathway has not
been recognized as a direct pathway in recent physiologically based
pharmacokinetic (PBPK) models and risk assessments based on them
(sections 6.6 and 10).
VC exposure does not result in enzyme induction but, on the
contrary, causes destruction of the cytochrome P-450 protein
responsible for its biotransformation (Pessayre et al., 1979; Du et
al., 1982). The impaired rate of oxidative metabolism associated with
P-450 destruction may partly explain the phenomenon of saturation of
the VC metabolism already at relatively low dosage levels and on the
other hand explain the tolerance to liver damage in experimental
animals subjected to continuous intermittent exposure to high VC
concentrations (73 000 mg/m3, 7 h/day, 5 days/week for 6 weeks).
Enzymes metabolizing VC were shown to be saturated in rats at a
concentration of 650 mg/m3 (rats exposed in a closed system; Bolt et
al., 1977; Filser & Bolt, 1979). In rhesus monkeys saturation of
metabolic elimination of VC was observed at atmospheric concentrations
greater than 520 mg/m3 (closed system; Buchter et al., 1980).
Saturation of metabolism occurred in rats at a single oral dose of
20 mg/kg body weight by gavage (Watanabe & Gehring, 1976; Table 23).
Saturation conditions were not reached in humans at an inhalation
exposure to 60 mg/m3 for 6 h (Krajewski et al., 1980).
In closed systems, after the initial absorption of VC until
equilibrium between atmosphere and organism, the continued absorption
is attributed to metabolism (Bolt et al., 1977). Rats exposed to VC
concentrations that did not exceed the above-mentioned threshold of
saturation metabolized VC in accordance with first-order kinetics with
a half-time of 86 min. At concentrations above saturation, the
elimination followed zero-order kinetics (Hefner et al., 1975a,c;
Filser & Bolt, 1979).
Although VC is primarily metabolized in the hepatocyte
(Ottenwälder & Bolt, 1980), the primary target cell for
carcinogenicity in the liver is the sinusoidal cell, as can be seen
from the incidence of ASL in both animals and humans. Non-parenchymal
cells have only 12% of the activity of hepatocytes in transforming VC
into reactive, alkylating metabolites (Ottenwälder & Bolt, 1980). VC
does not induce DNA damage in isolated non-parenchymal liver cells, as
measured by an alkaline comet assay (Kuchenmeister et al., 1996),
whereas it does in isolated hepatocytes. However, the majority of the
DNA adduct studies have been conducted or related to the hepatocyte.
It can be postulated that the majority of reactive metabolites can
leave the intact hepatocyte to produce tumours in the sinusoidal cells
(Laib & Bolt, 1980). The greater susceptibility of the sinusoidal
cells to the carcinogenic effects of VC may also result from the
inability of the sinusoidal cells to repair one or more of the DNA
adducts produced by VC as efficiently as the hepatocyte.
6.4 Elimination and excretion
After low doses, VC is metabolically eliminated and non-volatile
metabolites excreted mainly in the urine. At doses that saturate the
metabolism, the major route of excretion is exhalation of unchanged
VC. Independent of applied dose, the excretion of metabolites via
faeces is only a minor route. The metabolic clearance of VC is slower
in humans than in experimental animals, on a body weight basis.
However, it is comparable in several mammalian species, including
humans, when calculated on a body surface area basis.
6.4.1 Oral exposure
Male rats were gavaged with different doses of 14C-labelled VC,
and the radioactivity excreted was determined during the following
72 h (Green & Hathway, 1975; Watanabe & Gehring, 1976; Watanabe et
al., 1976a). Results are presented in Table 23. With low doses
radioactivity was mainly excreted as conjugated metabolites via the
urine or exhaled as 14C-labelled CO2 (section 6.3), but with doses
of 20 mg/kg body weight or more the main elimination route was
exhalation of unchanged VC (Table 23), suggesting saturated
metabolism. A minor route of excretion at all doses tested is via the
faeces.
Measuring the elimination of radioactivity in the urine as a
function of time revealed biphasic elimination at dose levels up to
100 mg/kg body weight with a half-life of approximately 4.6 h in the
initial rapid phase (first-order kinetics) (Watanabe et al., 1976a).
At low doses (1 mg/kg body weight) pulmonary elimination (as
CO2) during the first 4 h was monophasic with a half-life of 58 min,
but with a dose of 100 mg/kg body weight elimination of mainly
unchanged VC was biphasic with an initial rapid phase (half-life
14.4 min) followed by a slow phase (half-life 41 min) (Watanabe et
al., 1976a).
Most of the radioactivity excreted via urine or exhaled as the
metabolite CO2 was eliminated during the first 24 h after gavaging of
0.25 or 450 mg/kg body weight, whereas elimination of unchanged VC via
the lung was complete within 3-4 h (Green & Hathway, 1975).
Pretreatment of rats with unlabelled VC (up to 300 mg/kg body
weight per day orally for 60 days) had no effect on the rate of
elimination of a single oral dose of 14C-VC (oral application on
day 1 and 60) (Green & Hathway, 1975), suggesting that VC did not
induce its metabolism.
6.4.2 Inhalation exposure
Metabolic elimination of VC has been investigated in different
species, measuring the decline of VC in the gas phase of a closed
system into which VC was initially injected (Buchter et al., 1978;
Filser & Bolt, 1979; Buchter et al., 1980). Using VC concentrations
that did not exceed the saturation threshold (section 6.3), the
following first-order metabolic clearance rates for VC (expressed in
litre/h per kg body weight; initial concentration in mg/m3 in
parentheses) were determined: rat 11.0 (< 650), mouse 25.6 (130),
gerbil 12.48 (130), rabbit 2.74 (130), rhesus monkey 3.55 (< 520),
humans 2.08 (26). Because the metabolism of VC is perfusion-limited
(Filser & Bolt, 1979), comparison of clearance rates should be made on
a body surface area basis rather than a body weight basis. In this
case, these six mammalian species exhibit similar clearance rates.
With exposure concentrations above the "saturation point"
(> 650 mg/m3), the maximum velocity of metabolic elimination in rats
was 110 µmol/h per kg body weight (Filser & Bolt, 1979) or 3.6 mg/h
per kg body weight (Barton et al., 1995).
Elimination and excretion of 14C in rats within 72 h after a 6-h
exposure to 26, 2600 or 13 000 mg/m3 14C-labelled VC is shown in
Table 25 (similar experimental design; Watanabe et al., 1976b, 1978b;
Watanabe & Gehring, 1976). The amount of expired VC increased with the
exposure concentration, whereas the relative urinary excretion of
metabolites decreased, indicating a saturation of metabolism. Minor
decreases were seen in the proportion excreted via the faeces or
expired as CO2.
Measuring the time course of expiration in these experiments, the
pattern of pulmonary elimination of unchanged VC was similar at all
exposure concentrations, following first-order kinetics with
half-lives of 20.4, 22.4 (Watanabe et al., 1976b) and 30 min (Watanabe
et al., 1978b), respectively. After a 6-h exposure to 26 and
2600 mg/m3, elimination of the 14C-label via urine as a function of
time revealed a biphasic excretion of radioactivity with estimated
half-lives for the first (rapid) phase of 4.6 and 4.1 h (Watanabe et
al., 1976b). Because of extremely variable excretion curves in the
second phase, no attempts were made to estimate the half-lives of the
slow phase, which accounted for less than 3% of the radioactivity
excreted in the urine. Similar results were presented by Bolt et al.
(1976). Rats exposed for 5 h to 130 mg/m3 14C-VC excreted 70% of
incorporated radioactivity during the first 24 h after exposure in
urine and less than 3% in the following 3 days.
The rate of elimination of a single inhalative exposure to
14C-VC is not influenced by prior repeated exposure to the same
concentration (13 000 mg/m3) of unlabelled VC 6 h/day, 5 days/week
for 7 weeks (Watanabe et al., 1978b).
In human volunteers, the mean concentration of VC in the expired
air up to 30 min after a 6-h exposure to 7.5-60 mg/m3 reached no more
than 5% of the inhaled concentration (Krajewski et al., 1980).
When male volunteers were exposed to 130 (n=6), 650 (n=4) or
1300 mg/m3 (n=4) for 7.5 h, the VC concentration in expired air was
2.6, 23 or 52 mg/m3, respectively, 1 h after exposure (Baretta et
al., 1969).
6.5 Reaction with body components
6.5.1 Formation of DNA adducts
In vitro, both CEO and CAA can form etheno adducts with nucleic
acid bases (Fig. 3; Bolt, 1986; Bartsch et al., 1994; Barbin, 1998),
but the former exhibits greater reactivity (Guengerich, 1992). In
addition, 7-OEG has been characterized as a major reaction product of
CEO with guanine (Scherer et al., 1981), whereas CAA does not yield
this adduct (Oesch & Doerjer, 1982). 1, N6-Ethenoadenosine and
3, N4-ethenocytidine were characterized as reaction products of VC
with ribonucleosides in the presence of a microsomal activation system
(Barbin et al., 1975; Laib & Bolt, 1978). Analysis of DNA incubated
in vitro with rat liver microsomes, an NADPH-regenerating system
and [14C]-VC revealed the formation of 7-OEG, the major DNA adduct,
and of 1, N6-etheno-2'-deoxyadenosine (Epsilon dA) and
3, N4-etheno-2'-deoxycytidine (Epsilon dC) (Laib et al., 1981).
More recently, Müller et al. (1997) quantified six adducts in DNA
treated with CEO, including 7-OEG, the four ethenobases and
5,6,7,9-tetrahydro-7-hydroxy-9-oxoimidazo[1,2-alpha]purine. The
reactivity of CAA towards double-stranded B-DNA is very low
(Guengerich, 1992). CAA reacts with unpaired A and C bases to yield
Epsilon A and Epsilon C, respectively. Treatment of DNA with CAA has
also been reported to result in the formation of N2,3-Epsilon G and
1, N2-Epsilon G moieties (Oesch & Doerjer, 1982; Kusmierek & Singer,
1992; Guengerich, 1992).
The formation of Epsilon dC and tentatively of Epsilon dA in
liver DNA from rats exposed to VC in their drinking-water for 2 years
was reported by Green & Hathway (1978). In subsequent studies, Epsilon
A and Epsilon C were found in the nucleotides in hydrolysates of rat
liver RNA and 7-OEG but not Epsilon A or Epsilon C in DNA after
exposure to [14C]-VC (Laib & Bolt, 1977, 1978; Laib et al., 1981).
Similar results were found in mice (Osterman-Golkar et al., 1977).
More recent studies, using analytical methods (HPLC and fluorescence
spectrophotometry, monoclonal antibodies and negative-ion chemical
ionization with mass spectrometry using electrophore labelling; Fedtke
et al., 1989, 1990a; Eberle et al., 1989) or experimental designs with
greater sensitivity (young rats, short delay between exposure and
analysis), have led to conclusive demonstration of Epsilon dA and
Epsilon dC as DNA adducts in different rat organs after inhalation
exposure to VC (Table 26). The concentration of DNA adducts was 3- to
8-fold higher in the liver than in the lung and kidney, reflecting the
higher capacity of the liver for metabolic activation of VC (Fedtke et
al., 1990b; Swenberg et al., 1992). Both etheno bases (Epsilon C and
Epsilon A) accumulated in rat liver DNA during intermittent exposure
to VC (1300 mg/m3). Only Epsilon C accumulated in rat lung and
kidney, Epsilon A appearing to accumulate principally in the target
organ, the liver (Guichard et al., 1996). Subsequently, analysis of
further tissues showed increased levels of Epsilon A in the testis,
but not in the brain and spleen of rats exposed intermittently for 8
weeks; levels of Epsilon C increased in the testis and spleen but not
in the brain (Barbin, in press).
The rate of reaction of CAA with nucleic bases is slower than
with CEO (Zajdela et al., 1980). In in vitro studies, CEO but not
CAA was shown to be the main entity giving rise to etheno adducts
(Guengerich, 1992). Furthermore, as discussed in section 7.8, CEO but
not CAA show similar toxicity/mutagenic profiles to VC in a
metabolically competent human B-lymphoblastoid line (Chiang et al.,
1997). These findings seem to corroborate the original suggestion that
it is CEO rather than CAA that is the main source of etheno adducts
(Van Duuren, 1975; Guengerich et al., 1981; Gwinner et al., 1983).
In pre-weanling rats exposed to VC, 7-OEG had a half-life of
approx. 62 h, while the etheno adducts are highly persistent with a
half-life for Epsilon G of approx. 30 days (Swenberg et al., 1992).
Studies on the persistence of Epsilon A and Epsilon C in the liver of
adult rats have shown that there is no significant decrease in adduct
levels 2 months after termination of VC exposure (Guichard et al.,
1996). This is in contrast with the known repair of etheno adducts in
vitro (Dosanjh et al., 1994). The rat and human 3-methyladenine DNA
glycosylases can excise Epsilon A from DNA (Saparbaev et al., 1995).
Epsilon C can be released by the human mismatch-specific thymine-DNA
glycosylase (Saparbaev & Laval, 1998).
DNA adduct formation seems to be age-dependent; about 5- to
6-fold more DNA adducts were determined in young animals compared to
adults (Laib et al., 1989). Similar results were presented by Fedtke
et al. (1990b) who exposed adult and 10-day-old Sprague-Dawley rats
(Table 26). Ciroussel et al. (1990) exposed 7-day-old and 13-week-old
rats (strain BD IV) for 2 weeks to VC and detected 6 times more DNA
adducts in the liver of young rats compared to adults (Table 26).
It should be noted that background levels of etheno bases have
been found in various tissues in unexposed rodents (Guichard et al.,
1996; Fernando et al., 1996; Barbin, in press) and humans (Nair et
al., 1995, 1997). Lipid peroxidation products have been shown to react
with nucleic acid bases yielding etheno adducts (El Ghissassi et al.,
1995a; Chung et al., 1996; Bartsch et al., 1997). The role of lipid
peroxidation products in the endogenous formation of background levels
of etheno bases is further supported by the finding of elevated levels
of Epsilon dA and Epsilon dC in the liver from Long Evans Cinnamon
rats (Nair et al., 1996) and from patients with Wilson's disease or
with primary haemochromatosis (Nair et al., 1998).
Etheno adducts are also formed via substituted oxiranes formed
from other vinyl monomers, e.g., vinyl bromide (Bolt, 1994) and vinyl
and ethyl carbamate (urethane) (Park et al., 1993).
6.5.2 Alkylation of proteins
Bolt et al. (1980) studied covalent binding of radiolabel to
proteins in rats exposed to 14C-labelled VC. The target of alkylation
is the free sulfhydryl group of proteins. The liver always showed the
highest binding rate. The fraction of VC that was irreversibly bound
to proteins was independent of the VC dose applied, indicating no
threshold effect even at low doses. In vivo studies on rats
(Guengerich & Watanabe, 1979) have shown that the amount of total VC
metabolites bound in the liver to proteins is twice that bound to DNA,
RNA and lipids. Osterman-Golkar et al. (1977) reported the alkylation
of cysteine (S-(2-hydroxyethyl)cysteine) and histidine ( (N-1-and
N-3) hydroxyethylhistidine) of the globin precipitate of haemoglobin
and small amounts of the alkylated histidines in proteins from testis
in mice exposed to 1,2-14C-vinyl chloride.
In vitro studies have shown that incubation of rat liver
microsomes with 14C-labelled VC results in NADPH-dependent microsomal
uptake and covalent binding to microsomal proteins (Kappus et al.,
1975, 1976; Baker & Ronnenberg, 1993). It has been suggested that VC
metabolites might be involved in the destruction of the haem moiety of
cytochrome P-450 in the liver (Guengerich & Strickland, 1977).
6.6 Modelling of pharmacokinetic data for vinyl chloride
There has been progress in recent years in the development of
physiologically based toxicokinetic (PBTK) models describing the
toxicokinetics of chemicals. These theoretical models permit
predictions of the dose of active metabolites reaching target tissues
in different species, including humans, and, therefore, have improved
the toxicokinetic extrapolation in cancer risk assessments. PBTK
models have also been used as a tool to examine the behaviour of VC in
mammalian systems. The description of PBTK models for VC is presented
in Annex 2.
Table 26. Detection of DNA adducts in vivo after VC inhalation in ratsa
Strain; sex; Treatment; Investigated Alkylated bases in Comments References
age post-exposure organs DNA (max. concentration
survival time in pmol/µmol unmodified
base in specified organs)
BD IV; male & 1300 mg/m3, 7 h/day, liver, lung, epsilon-dA (0.131, 0.105, higher sensitivity of Ciroussel
female; 7 days 7 days/week for brain, kidney 0.06, b.d.l.); epsilon-dC young rats compared et al.(1990)
2 weeks; none (0.492, 0.246, 0.216, with adults
b.d.l.) (see section 7.7.4)
BD IV; male; 1300 mg/m3, 7 h/day, liver epsilon-dA (0.019); Ciroussel
13 weeks 7 days/week for epsilon-dC (0.080) et al. (1990)
2 weeks; none
S.-D.; male & 1560 mg/m3, 4 h/day liver, lung, 7-OEG (162, 20, 29, <10, higher sensitivity of Fedtke et
female; 10 days for 5 days; kidney, <10); N2,3-epsilon-G (1.81, young rats compared al. (1990b)
0, 3, 7, 14 days brain, spleen 0.21, 0.31, <0.12, <0.12); with adults (see
measured immediately a.e. section 7.7.4)
concerning liver
adducts
S.-D.; female; 1560 mg/m3, 4 h/day liver, lung, 7-OEG (43, 20, n.g.); DNA adduct formation Fedtke et
adult for 5 days; kidney N2,3-epsilon-G (0.47, max. at the end of al. (1990b)
0, 3, 7, 14 days 0.27, <0.12); measured exposure; most DNA
immediately a.e. adducts in the liver
S.-D.; male & 5200 mg/m3, 7 h/day liver, lung epsilon-dA (0.05, 0.13) more DNA adducts in the Eberle et
female; 11 days on days 1-9 and 24 h epsilon-dC (0.16, 0.33) lung than in the liver al. (1989)
on day 10; none
S.-D.; male & 1560 mg/m3, 4 h/day liver, lung, 7-OEG (162, 20, 29); DNA adduct Swenberg
female; 10 days for 5 days; kidney N2,3-epsilon-G (1.81, 0.21, max. at the end et al.
0, 3, 7, 14 days 0.31); epsilon-dC (0.98, 0.30, exposure; most DNA (1992)
0.29); epsilon-dA (0.21, 0.065, adducts in the liver
0.04); measured immediately a.e.
Table 26. (cont'd)
Strain; sex; Treatment; Investigated Alkylated bases in Comments References
age post-exposure organs DNA (max. concentration
survival time in pmol/µmol unmodified
base in specified organs)
S.-D.; male; 1300 mg/m3, 4 h/day, liver, lung, epsilon-dA (liver: background exposure-time-dependent Guichard
6 weeks 5 days/week for 1, kidney, 0.0004, a.e. up to 0.045; increase of adducts in et al. (1996)
2, 4, 8 weeks; none lymphocytes background lung and kidney up to liver (ca. 100-fold);
0.033, no increase a.e.); no (lymphocytes) or
epsilon-dC (liver: background slight increase (ca.
0.0007, a.e. up to 0.08; kidney: 2-fold in kidney, ca.
background 0.086, a.e. up to 5-fold in lung) of
0.16; lung: background 0.072, adducts in other
a.e. up to 0.38) organs but higher
background levels
8 weeks brain, testis, epsilon-dA (increase in testis, Barbin
spleen not in brain or spleen) (in press)
epsilon-dC (increase in testis
and spleen, but not in brain)
a a.e. = after exposure; b.d.l. = below detection limit; n.g. = not given; S.-D. = Sprague-Dawley; epsilon-dA = 1,N6-ethenodeoxyadenosine;
epsilon-dC = 3,N4-ethenodeoxycytidine; 7-OEG = 7-(2-oxoethyl)guanine; epsilon-G = ethenoguanine
7. EFFECTS ON LABORATORY MAMMALS AND IN VITRO TEST SYSTEMS
7.1 Acute toxicity
VC appears to be of low acute toxicity when administered to
various species by inhalation. A summary of acute toxicity is given in
Table 27. No data are available on acute toxicity after dermal
application.
VC has a narcotic effect (see also section 7.6) after inhalative
administration of high doses. In rats, mice and hamsters, death was
preceded by increased motor activity, twitching of extremities,
tremor, ataxia, tonic-clonic convulsions and accelerated respiration
(Patty et al., 1930; Mastromatteo et al., 1960; Prodan et al., 1975).
In dogs, severe cardiac arrhythmias occurred under narcosis after
inhalative exposure to 260 000 mg VC/m3 (Oster et al., 1947). Similar
results were reported by Carr et al. (1949). Pneumonitis was more
frequent in experimental mice than in controls 8 and 18 months after a
single 1-h exposure to 1000, 3000, 13 000 or 130 000 mg/m3 (Hehir et
al., 1981).
After acute inhalative exposure to VC, pathological findings in
the rat included congestion of the internal organs, particularly lung,
liver and kidney as well as pulmonary oedema (Patty et al., 1930;
Mastromatteo et al., 1960; Lester et al., 1963; Prodan et al., 1975).
Exposure to VC (3900 mg/m3) for 24 h did not cause pathological
changes in male and female rats or female New Zealand rabbits. In
mice, exposure to the same concentration for 4 and 8 h resulted in
circulatory changes, while longer exposure (12 and 24 h) caused
vasomotor paralysis followed by characteristic shock with subsequent
alterations in the liver and lungs (Tįtrai & Ungvįry, 1981).
7.2 Short-term toxicity
7.2.1 Oral exposure
Groups of 15 male and 15 female weanling Wistar rats were gavaged
with VC dissolved in soya-bean oil (0, 30, 100 and 300 mg/kg body
weight, once daily, 6 days/week for 13 weeks). The treatment caused no
noticeable changes in appearance or behaviour, body weight gain or
food intake. The total number of white blood cells and the sugar
content of the blood were slightly decreased by the intermediate and
high dose levels. The activities of serum ASAT and ALAT and of urinary
ASAT were decreased in males given the top dose. There were no other
significant changes in the haematological or biochemical indices and
no treatment-related alterations were observed in the microscopic
constituents of the urine. The relative weight of the liver in males
and females showed a tendency to increase with increasing doses of VC
but the difference from the controls was statistically significant
only at the highest dose level. The NOEL was given as 30 mg/kg body
weight (Feron et al., 1975). In contrast, in another study, all male
and female Wistar rats (presumably a total of 15 rats) receiving once
daily 300 mg VC/kg body weight in peroxide-free corn oil by gavage
died within 60 days of treatment (Knight & Gibbons, 1987).
7.2.2 Inhalation exposure
A summary of the non-neoplastic effects of VC after short-term
inhalation is presented in Table 28. Information on carcinogenic
effects after short-term inhalation are presented in section 7.7 and
Table 30.
In various species, the main target of VC toxicity is the liver.
A dose-related significant increase in relative liver weight was found
in male rats at exposure levels of 26, 260 and 7800 mg/m3 (Bi et al.,
1985). Degenerative effects on liver parenchyma were reported in
rabbits at a dose level of 520 mg/m3 (Torkelson et al., 1961), in
rats at 1300 mg/m3 (Torkelson et al., 1961; Wisniewska-Knypl et al.,
1980), and in mice at 2600 mg/m3 (Lee et al., 1977). Bi et al. (1985)
reported a decreased relative testis weight in rats at exposure levels
of 260 and 7800 mg/m3; however, the effect was not dose-related.
These authors also reported a higher incidence and severity of damage
to the testicular seminiferous tubules at all dose levels tested (26,
260 and 7000 mg/m3), the differences with the controls being
statistically significant only at the two highest dose levels.
However, the severity of the testicular damage was clearly
dose-related (correlation coefficient 0.993; P < 0.01) suggesting
an adverse effect of VC on the testes already at 26 mg/m3. Effects on
relative liver weight were detected in rats at 26 mg/m3 (Bi et al.,
1985) (see also Table 28 and section 7.5.1). Effects on the kidney
(Lee et al., 1977; Feron et al., 1979a,b; Himeno et al., 1983) and the
lung (Suzuki, 1980) were observed in rats and/or mice at higher doses
(Table 28). For mice a LOEL of 130 mg/m3 was given (decreased
survival; Hong et al., 1981). Rats, mice and rabbits seem to be more
sensitive than guinea-pigs and dogs (Torkelson et al., 1961; Hong et
al., 1981).
7.2.3 Dermal exposure
No studies were available on short-term dermal exposure.
7.3 Long-term toxicity - effects other than tumours
7.3.1 Oral exposure
Studies on effects induced by long-term oral application of VC in
rats are presented in detail in Table 29. Studies on other species are
not available.
Long-term feeding studies in male and female rats showed
increased mortality in males at doses > 5.0 mg/kg body weight per
day (Feron et al., 1981) and in females at doses of > 1.3 mg/kg
Table 27. Toxicity of VC after acute inhalation exposurea
Species Duration of Parameter Value References
exposure in g/m3
Rat 2 h LC50 390 Prodan et
LC100 525 al. (1975)
Rat 30 min LC100 780 Mastromatteo
et al. (1960)
Rat 1 h ataxia, 130 Hehir et al.
hyperventilation (1981)
Mouse 2 h LC50 293 Prodan et al.
LC100 375 (1975)
Mouse 30 min LC100 780 Mastromatteo
et al. (1960)
Guinea-pig 2 h LC50 595 Prodan et al.
LC100 700 (1975)
Guinea-pig 30 min death, 780 Mastromatteo
threshold dose et al. (1960)
Guinea-pig 2-6 h deep narcosis, 260 Patty et al.
no death (1930)
Guinea-pig 18-55 min death, 390-650 Patty et al.
threshold dose (1930)
Guinea-pig 90 min narcosis, 65-130 Patty et al.
threshold dose (1930)
Rabbit 2 h LC50 295 Prodan et al.
LC100 700 (1975)
a Cited studies not conducted according to present-day standards
Table 28. Toxicity of vinyl chloride in animals after short-term inhalation exposures - non-neoplastic effectsa
Species, strain; Doses in Exposure duration; Significant effects References
number of animals mg/m3 frequency;
per dose per post-exposure
exposure period observation period
Rat, Wistar; 8 0, 26, 260, 3 or 6 mo; 26 mg/m3: relative spleen and heart weight Bi et al.
(3 mo) or 30 7800 6 d/week, 6 h/d; * (6 mo); incidence of testicular seminiferous (1985)
(6 mo) m rats none tubule damage * ! (exposure period not specified)
> 26 mg/m3: relative liver weight * (6 mo) 260 mg/m3:
relative heart weight * (3 mo) > 260 mg/m3: relative
testis weight ** (6 mo); incidence of testicular
tubule damage * 7800 mg/m3: relative kidney and
spleen weight * (3 mo)
Rat; n.g.; at high 0, 130, 260, 4.5 (high dose) 130 mg/m3: NOEL (body and organ weight, Torkelson
dose 10 f & 520, 1300 or 6 mo; survival, haematology, clinical chemistry, urine et al.
10 m (control 5 d/week, analysis, histopathology) 260 mg/m3: relative liver (1961)
5 f & 5 m); other 7 h/d; none weight * (m+f) 1300 mg/m3: granular degeneration
groups 20-24 m in centrilobular liver parenchyma #;
& 24 f liver weight * (m)
Rat, Wistar; 0, 130, 1, 3, 6 mo; 130 mg/m3: slight changes such as proliferation Wisniewska-Knypl
8-10 m 1300, 5 d/week, 5 h/d; of hepatocellular SER (3-6 mo)# et al.
52 000 none > 1300 mg/m3: liver weight * (1-6 mo), ultrastructural (1980)
hepatocellular changes (swollen mitochondria, lipid
droplets *) after 3 and 6 mo #
Rat, 0, 2465 24.5 weeks; 2465 mg/m3: mortality *# (m+f), haematology and Groth et
Sprague-Dawley; 7 h/d, 5 d/week; clinical chemistry <-> al. (1981)
110-128 rats up to 19 weeks
per sex
Rat, Wistar; 0, 13 000 4, 13, 26 weeks; 13 000 mg/m3, > 4 weeks: body weight **, Feron et al.
10 f & 10 m 5 d/week, 7 h/d; blood clotting time ** > 13 weeks: liver function (1979a,b)
none (BSB-retention test) **; liver and kidney
weight * (m+f)
Table 28. (cont'd)
Species, strain; Doses in Exposure duration; Significant effects References
number of animals mg/m3 frequency;
per dose per post-exposure
exposure period observation period
26 weeks: spleen weight * (m+f); clear cell foci Feron &
and basophilic foci in the liver * (m+f)# Kroes (1979)
Rat, Sherman; 0, 52 000 92 d; 5 d/week, 52 000 mg/m3: relative liver weight * and Lester et al.
12-15 rats/sex 8 h/d; none spleen weight ** (m+f); white blood cell counts **; (1963)
swelling of hepatocytes with vacuolization,
compression of sinusoids #
Mouse, CD-1; 0, 130, 650, 1, 3, 6 mo; > 130 mg/m3: survival after 6 mo exposure ** # (low Hong et al.
8-28 mice/sex 2600 5 d/week, 6 h/d; dose: m, 1/8 versus 22/28 in control; f, 0/8 versus (1981)
12 mo 23/28; tumour incidences no differences)
Mouse, CD-1; 0, 2600 3-9 exposures; 2600 mg/m3: early deaths (2 m + 1 f)!; pathological Lee et al.
36 mice/sex 5 d/week, 6 h/d; changes in dead animals: acute toxic hepatitis (1977)
none (congestion, diffused necrosis), tubular necrosis
in renal cortex
Mouse, n.g.; 6500 1 or 6 mo; 6500 mg/m3: hyperplastic nodules and dilatated Schaffner
5 m (1 mo) or (no control) 5 d/week, 5 h/d; sinusoids in liver parenchyma after 6 mo (1979)
14 m (6 mo) none exposure #
Mouse, CD-1; 0, 6500, 5-6 mo; > 6500 mg/m3: proliferation and hypertrophy of Suzuki
3-16 m 15 600 5 d/week, 5 h/d; bronchiolar cells, hypersecretion of bronchial (1980)
2-37 d and bronchiolar epithelium, hyperplasia of
alveolar epithelium, bronchiolar inflammation #
(only pulmonary effects recorded; effects not
dose related)
Mouse, CD-1; 13 000 10 weeks; 13 000 mg/m3: focal lung hyperplasia, proliferation Himeno et
10 m (no control) 5 d/week, 4 h/d; of sinusoidal cells of the liver, proliferative al. (1983)
none effects in renal glomeruli, giant cells in testis #
Table 28. (cont'd)
Species, strain; Doses in Exposure duration; Significant effects References
number of animals mg/m3 frequency;
per dose per post-exposure
exposure period observation period
Mouse, ICR; a) 0, 13 000, a) 5 to 6 d, basophilic stippled erythrocytes * (#) in a) and b), Kudo et
n.g. 26 000; b) 62 d; a) 4 h/d, related effect not dose al. (1990)
b) 78 to 104 b) continuously;
none
Hamster, 0, 520 6 mo; 5 d/week, 520 mg/m3: survival ** Drew et
golden Syrian; 6 h/d; life span al. (1983)
56 f
Guinea-pig, 0, 130, 6 mo; 5 d/week, 520 mg/m3: NOEL (body and organ weight, Torkelson
n.g.; 10-12 m 260, 520 7 h/d; none survival, clinical chemistry, histopathology) et al. (1961)
& 8-12 f
Rabbit, n.g.; 0, 130, 6 mo; 5 d/week, 260 mg/m3: NOEL (body and organ weight, Torkelson
3 rabbits/sex 260, 520 7 h/d; none survival, clinical chemistry, histopathology) et al. (1961)
520 mg/m3: degeneration of centrilobular
liver parenchyma (m+f) with periportal cellular
infiltration (f) #
Dog; n.g.; 0, 130, 6 mo; 5 d/week, 130-520 mg/m3: no effects recorded (body and Torkelson
1 dog/sex 260, 520 7 h/d; none organ weight, survival, haematology, clinical et al. (1961)
chemistry, urine analysis, histopathology) #
a d = day; mo = month, m = male; f = female; n.g. = not given; ! = increase not significant; # = no data about significance;
* = increased; ** = decreased; <-> = no change
body weight per day (Feron et al., 1981; Til et al., 1983, 1991). At
14.1 mg/kg body weight per day, blood clotting time was decreased and
alpha-fetoprotein levels in blood serum were increased (Feron et al.,
1981). Skin fibrosis was observed at 30 mg/kg body weight per day
administered by gavage (Knight & Gibbons, 1987).
As for short-term exposure, the primary target organ of VC in
rats after long-term oral exposure is the liver. Female rats appeared
to be more sensitive than males to the hepatotoxicity of VC
(section 7.7.4). Increased relative liver weights were found at
14.1 mg/kg body weight per day after feeding periods of 6 or 12 months
(Feron et al., 1981). Morphological alterations of the liver included
extensive hepatocellular necrosis at doses > 5 mg/kg body weight
per day, foci of haematopoiesis at 14.1 mg/kg body weight per day, and
cysts and liver cell polymorphism (variation in size and shape of
hepatocytes and their nuclei) at doses > 1.3 mg/kg body weight per
day (Feron et al., 1981; Til et al., 1983, 1991). Foci of
hepatocellular alteration (clear cell, mixed cell, eosinophilic and
basophilic foci) were common findings. Clear cell, mixed cell and
eosinophilic foci occurred at doses > 1.3 mg/kg body weight per day
and basophilic foci at doses > 0.014 mg/kg body weight per day (Til
et al., 1983, 1991).
7.3.2 Inhalation exposure
A summary of non-neoplastic and neoplastic effects after
long-term intermittent inhalation of VC is presented in Table 32 with
details of exposure regime (see also section 7.7). Long-term exposure
to VC by inhalation resulted in increased mortality in rats exposed to
a dose of 260 mg/m3 for 12, 18 and 24 months, in mice exposed to
130 mg/m3 for 6, 12 and 18 months and in hamsters exposed to
520 mg/m3 for 6, 12 and 18 months (Drew et al., 1983). Maltoni et al.
(1984) reported increased mortality in rats (BT15) and hamsters (BT8)
at lower dose levels (2.6 and 130 mg/m3, respectively), but no
statistical evaluation was performed. Rats exposed to 130 mg/m3
showed reduced body weight and increased relative spleen weight (Sokal
et al., 1980; see below). At this dose morphological alterations were
reported in rat liver, such as hepatocellular lipid accumulation and
mitochondrial swellings (Wisniewska-Knypl et al., 1980) as well as
proliferation of cells lining the liver sinusoids (Sokal et al.,
1980). Exposure to higher doses revealed degenerative alteration in
the testis (1300 mg/m3; Sokal et al., 1980) and tubular nephrosis and
focal degeneration of the myocardium (13 000 mg/m3; Feron & Kroes,
1979) in rats.
Male Wistar rats (42-80 per group) exposed to 0, 130, 1300 and
52 000 mg VC/m3, 5 days/week, 5 h daily, for 10 months, showed
significantly reduced body weight in all treatment groups; general
condition and behaviour were not altered. Relative organ weights of
spleen and heart (except at the mid dose) were significantly elevated
at > 130 mg/m3, as well as liver and kidney weights at mid- and
high-dose levels and testis weights at the high-dose level. X-ray
analysis did not show any skeletal alterations. Histopathology
revealed statistically significantly increased incidences of nuclear
polymorphism (nuclei of variable size and irregular shape) of
hepatocytes and proliferation of reticuloendothelial cells lining
liver sinusoids at the two highest dose levels. Fatty degeneration of
hepatocytes was found at all exposure levels. Ultrastructural changes
in hepatocytes seen at all exposure levels, but not in controls,
included swollen and giant mitochondria with broken cristae,
proliferation and dilatation of smooth endoplasmic reticulum, nuclear
membrane invaginations, areas of cytoplasmic degradation and increased
numbers of small lipid droplets (Sokal et al., 1980; Wisniewska-Knypl
et al., 1980). In addition, Sokal et al. (1980) reported necrotic foci
of the spermatogenic epithelium and disorders of spermatogenesis
accompanied by large multinuclear syncytial cells in the testis
predominantly at 1300 mg/m3. Haematology, urine analysis and clinical
chemistry did not reveal differences of toxicological significance.
Thus NOEL for rats or mice concerning non-neoplastic effects
could not be derived, since effects were observed at the lowest levels
studied (130 mg/m3).
7.4 Skin and eye irritation; sensitization
No relevant data on skin irritation were identified. No studies
were available on sensitizing effects of VC in animals. Erythema and
second-degree burns were reported in a worker after accidental
exposure to liquid VC (see section 8.3.2.1). Dryness of eyes and nose
was reported by volunteers exposed to 1300 mg/m3 (see section 8.2).
7.5 Reproductive toxicity, embryotoxicity and teratogenicity
7.5.1 Male reproductive toxicity
Inhalation studies on rats showed some evidence of reduced
fertility and morphological alterations of the testis. It should be
noted that none of the studies cited were conducted according to
current guidelines (OECD, 1983a,b).
In dominant lethal studies on mice no reduction in fertility was
observed (Anderson et al., 1976; Table 35). However, reduced fertility
was noted in male CD rats (12 per group) mated once on week 11 of
exposure (0, 130, 650 or 2600 mg/m3; 6 h/day, 5 days per week). VC
treatment decreased dose-dependently the ratio of pregnant to mated
females; this was significant at mid- and high-dose level (Short et
al., 1977; see also section 7.8.2).
Bi et al. (1985; for details see Table 28 and section 7.2.2)
observed decreased relative testis weight at 260 mg/m3 and
morphological alterations in the testis of rats even at the lowest
dose of 26 mg/m3. Morphological alterations in the testis of rats
were also reported by Sokal et al. (1980, see section 7.3.2).
VC was administered to adult female CD rats at 0, 24, 260 and
2860 mg/m3, 6 h/day, 5 days/week for at least 10 weeks prior to
mating until day 4 of lactation (94 + days) in a two-generation study.
Alterations in reproductive performance and fertility were not
detected at any dose level tested. Centrilobular hypertrophy in the
liver and increased relative liver weights, however, were noted at all
dose levels tested in a dose-related manner (Shah, 1998).
7.5.2 Embryotoxicity and teratogenicity
Although the available studies did not follow guideline
standards, the information leads to the conclusion that there is
embryotoxicity or fetal toxicity, including increased numbers of
resorptions, decreased numbers of live fetuses and delayed development
at dose levels producing maternal toxicity. VC treatment did not
induce gross malformations. There is evidence for the permeability of
the placenta to VC (Ungvįry et al., 1978; see section 6.2.2).
John et al. (1977, 1981) investigated mice, rats and rabbits for
teratogenic effects of inhaled VC. In all three species a similar
experimental design was used. Pregnant CF-1 mice were exposed 7 h/day
to 0, 130 or 1300 mg VC/m3 on day 6-15 of gestation. For both
concentrations tested, concurrent control groups were sham-exposed.
Animals were observed daily and maternal body weight recorded at
several intervals (no further data). The mice were sacrificed on day
18 of gestation. After determination of external anomalies, one-third
of each litter (19-26 litters per group) was examined for soft tissue
anomalies and the other mice for skeletal anomalies. Exposure to
1300 mg/m3 led to deaths (5 of 29 bred females, P < 0.05), reduced
maternal body weight gain (P < 0.05) and food consumption
(P< 0.05). No maternal toxicity was apparent in females exposed to
130 mg/m3. The number of live fetuses per litter and fetal body
weight were significantly decreased and the number of resorptions
significantly increased at 1300 mg/m3, but these values were within
the range observed for the second concurrent control group or for
historical controls. No soft tissue or external anomalies were
detected. Significantly increased incidences of three skeletal
variants (delayed skull and sternebrae ossification, unfused
sternebrae) in the high-dose group were indicative of delayed skeletal
development. No developmental toxicity was observed at 130 mg/m3.
Sprague-Dawley rats were exposed to 0, 1300 or 6500 mg/m3 on
gestation day 6-15 and dams were sacrificed on day 21 of gestation.
Low-dose exposure resulted in reduced maternal weight gain
(P< 0.05). At the high-dose level further maternal effects like
reduced food consumption and increased liver weight were observed
(P< 0.05), and one out of 17 pregnant rats died (no further
information). Examination of 16-31 litters per group revealed
significantly decreased fetal weight in the low-dose but not in the
high-dose group. Significantly increased incidences in dilated ureter
were observed at 6500 mg/m3.
Rabbits were exposed on gestation day 6-18 to 0, 1300 or
6500 mg/m3 and sacrificed on gestation day 29. Except for reduced
food consumption in the low-dose group and 1 death in 7 bred females
of the high-dose group, no further evidence of maternal toxicity was
observed. In the high-dose group only 5 litters were examined, but in
other groups 11-19 litters. Compared to the concurrent control, litter
size was significantly reduced in the low-dose group (but not at
6500 mg/m3). However, there was an increase in litter size in this
treatment group compared to controls concurrent to the high-dose
group. The incidence of delayed ossification of the sternebrae was
increased at 1300 mg/m3, but not in the high-dose group (John et al.,
1977, 1981).
Ungvįry et al. (1978) exposed pregnant CFY rats continuously to 0
or 4000 mg VC/m3 during the first, second or last third of pregnancy
(gestation day 0-8, 7-13 or 13-20). Dams were sacrificed on gestation
day 20 and living, dead or resorbed fetuses were recorded. Placenta
and fetuses were weighed and fetuses macroscopically investigated. One
half of each litter was examined for soft tissue anomalies including
histopathology of organs with abnormalities and the other half was
processed for investigation of the skeletal system. Maternal weight
gain was significantly reduced in pregnant rats exposed during the
last third of pregnancy. Increased relative liver weight was observed
in dams exposed during the first or second third, but histopathology
revealed no pathological changes in the liver of any VC-treated rat.
No further signs of maternal toxicity were reported. Examination of 13
to 28 litters per group revealed increased fetal loss in per cent of
total implantation sites after exposure during gestation day 0-8
compared with the concurrent control. However, this value was not
significant compared with other control groups (e.g., control exposed
gestation day 13-20) of the same study. None of the soft tissue or
skeletal anomalies were attributed to VC treatment.
Exposure of 40 pregnant white Wistar rats to VC (mean level of
6.15 mg/m3 during whole gestation) resulted in elevated embryonic
mortality, lowered fetal weight, and induction of external and
internal anomalies in the development of the fetus (Mirkova et al.,
1978).
7.6 Special studies
7.6.1 Neurotoxicity
Profound narcosis was reported in guinea-pigs exposed to
65 000 mg VC/m3 for 90 min (Patty et al., 1930). Ataxia was observed
at this dose level after 5 min of exposure. The anaesthetic action of
VC was also observed in dogs (Oster et al., 1947) and mice (Peoples &
Leake, 1933). Mastromatteo et al. (1960) reported deep narcosis in
rats and mice exposed to 260 000 mg/m3 for 30 min. The narcotic
effect was preceded by increased motor activity after 5 min of
exposure, twitching of extremities (after 10 min), ataxia (after
15 min) and tremor (after 15 min). Rats exposed to 130 000 mg/m3 for
60 min showed ataxia preceded by hyperactivity but no narcotic effect
(Hehir et al., 1981).
Neuropathological alterations were observed in rats exposed to
78 000 mg/m3 (4 h/day, 5 days/week) for 12 months (Viola, 1970; Viola
et al., 1971). During the exposure period, the rats were slightly
soporific. Histopathology revealed diffuse degeneration in the gray
and white matter of the brain and at the level of the white matter
zones of reactive gliosis. In the cerebellum, atrophy of the granular
layer and degeneration of Purkinje cells were most prominent. In
addition, peripheral nerve bundles were often surrounded and invaded
by fibrotic processes.
Reports on neurotoxicity in humans occupationally exposed to VC
are given in section 8.3.2.3.
7.6.2 Immunotoxicity
Sharma & Gehring (1979) investigated mitogen-stimulated
transformation in splenic lymphocytes isolated from mice exposed to
26, 260 or 2600 mg/m3 for 2, 4 or 8 weeks (6 h/day, 5 days/week). The
treatment produced no effects on body or organ weight, except
increased spleen weight in high-dose groups, no effects on
haematological parameters and no pathological alterations at necropsy.
VC exposure caused stimulation of spontaneous lymphocyte
transformation in lymphocyte cultures prepared from mice exposed for 2
weeks to the high dose and from mice exposed for 4 weeks at all dose
levels, but this was not dose-dependent. The response of lymphocytes
to phytomitogens was increased at 2600 mg/m3 after exposure for
2 weeks and at all dose levels after exposure for 4 or 8 weeks, with
more pronounced effects at the mid-dose level. Stimulation of
lymphocyte transformation was not observed in lymphocytes from
unexposed mice cultured in the presence of VC, indicating that
metabolites of VC formed in vivo may be responsible for this effect.
Exposure of mice to 26 mg/m3 for 6 months (Bi et al., 1985;
Table 28) or rats to 130 mg/m3 for 10 months (Sokal et al., 1980;
section 7.3.2) induced increased relative spleen weight, whereas much
higher doses (52 000 mg/m3 for 92 days) produced decreased spleen
weight and reduced white blood cell counts (Lester et al., 1963; see
Table 28).
Reports on immunological and lymphoreticular effects in humans
occupationally exposed to VC are given in section 8.3.2.
7.6.3 Cardiovascular effects
Viola (1970) demonstrated thickening of the walls of small
arterial vessels (in some vessels blockage of lumen) due to
endothelial fibrosis and proliferation of endothelial cells in rats
exposed to 78 000 mg/m3 (4 h/day, 5 days/week) for 12 months.
Exposure of rats to 13 000 mg/m3 for 12 months (Feron & Kroes, 1979;
see Table 32) resulted in thickened walls of arteries and focal
degenerations of the myocardia.
Oster et al. (1947) observed cardiac arrhythmias (e.g.,
ventricular extrasystoles and fibrillation, auriculoventricular block)
in dogs at a dose level of 260 000 mg/m3.
Bi et al. (1985) reported increased relative heart weight in rats
exposed for 6 months to 26 mg/m3 or for 3 months to 260 mg/m3 (Table
28).
Impaired peripheral circulation as well as other cardiovascular
effects in occupational exposed humans are described in section 8.3.2.
7.6.4 Hepatotoxicity
Data on hepatotoxicity are presented in sections 7.2 and 7.3.
7.7 Carcinogenicity
VC causes a wide spectrum of tumours in animals and this spectrum
is similar in a number of different species (see Table 33). For
example in rats, the following tumours have been described after VC
inhalation exposure: liver and other angiosarcomas, other liver
tumours, mammary gland carcinoma, nephroblastoma, neuroblastoma,
stomach tumours and Zymbal gland tumours. The lowest dose at which an
increase in tumour incidences was observed when rats were exposed by
inhalation was 130 mg/m3 for liver angiosarcoma (ASL) and 13 mg/m3
for mammary tumours. There is evidence that animals are more
susceptible to tumour induction early in life. There is also evidence
that liver tumours are induced in female rats at lower doses than in
males.
VC is also carcinogenic in animals after oral application. The
spectrum of tumours is similar to that observed after inhalation
exposure. The lowest observed dose producing a carcinogenic effect
(ASL) in rats was 1.3 mg/kg body weight per day.
7.7.1 Oral exposure
Details of studies on the carcinogenicity of VC in rats after
oral administration are tabulated in Table 29. There are three gavage
studies: two studies with 52 or 59 weeks gavage and a 32- or 18-week
follow-up and a 2-year lifetime gavage study with small numbers of
animals and high mortality at high doses (Maltoni et al., 1981 [BT11
and BT27] and Knight & Gibbons, 1987). The two feeding studies used
PVC powder containing VC incorporated into the diet; the numbers of
animals were large, and the administration comprised the whole
lifetime of the animals (Feron et al., 1981; Til et al., 1983, 1991).
Table 29. Long-term toxicity/carcinogenicity of vinyl chloride in experimental animals after oral administrationa
Species; strain; Dose; 1) Effects other than tumours (dose in mg/kg bw/d); Reference
initial number of route of exposure; 2) number of animals for histopathological evaluation of
animals per dose; exposure period; neoplastic effects;
vehicle frequency of treatment; 3) type of tumour: number of animals with this tumour/dose
post-exposure group (unless otherwise given)
observation period
Rat; Wistar; 0, 1.7, 5.0, 14.1 1) > 1.7 mg/kg: haematology, biochemistry, urine analyis, Feron
60-80 rats/sex; mg/kg bw./d, organ function, body weight and food consumption <-> et al.
PVC powder bioavailable d (oral mortality !* (f); liver clear cell foci (f+m), basophilic (1981) e
(vehicle) intake 0, 1.8, 5.6, foci (f), eosinophilic foci (m+f) !*; liver-cell
containing VC 17.0 mg/kg bw/day); polymorphism !* (m); liver cysts !* (f);> 5.0 mg/kg: general
incorporated oral feed; lifespan condition ” (m+f, at mo 18); mortality !* (f+m); liver
into the diet study, terminated basophilic foci !* (m); extensive liver necrosis !* (f);
at week 135 (m) 14.1 mg/kg: extensive liver necrosis and liver cysts !* (m); focal
and 144 (f); diet haematopoiesis in liver !* (m); liver weight !* (f+m at mo
provided 4 h/d; 6, f at mo 12, interim sacrifice); blood clotting time at mo
none 6 and alpha-fetoprotein at mo 12 ”* (f+m);
2) 55, 58, 56, 59 m and 57, 58, 59, 57 f;
3) ASL in m: 0, 0, 6*, 27* and in f: 0, 0, 2, 9*;neoplastic liver
nodules in m: 0, 1, 7*, 23* and in f : 2, 26*, 39*, 44*;
hepatocellular carcinoma in m: 0, 1, 2, 8* and in f: 0, 4, 19*,
29*; lung angiosarcoma in m: 0, 0, 4*, 19* and in f: 0, 0, 1, 5*
Rat; Wistar; 100 0, 0.014, 0.13, 1.3 1) > 0.014 mg/kg: bw. and food consumption <-> (m+f); liver Til et al.
rats/sex except mg/kg bw./d, basophilic foci !* (f); 1.3 mg/kg: mortality !* (f; at (1983,
high dose (50 bioavailable d (oral week 149); liver glutathione level at week 40 or 80 <-> 1991) f
per sex); PVC intake 0, 0.018, (satellite groups); liver clear cell (m+f), basophilic (m),
powder (vehicle) 0.17, 1.7 mg/kg eosinophilic (f), and mixed cell foci (f) !*; liver cysts !*
containing VC bw./day); oral feed; (f ); moderate to severe liver-cell polymorphism !* (m+f );
incorporated into lifespan study, 2) 99, 99, 99, 49 m and 98, 100, 96, 49 f;3) ASL in m: 0, 0,
the diet terminated at week 0, 1 and 0, 0, 0, 2 in f; neoplastic liver nodules in m 0,
149 (m) and 150 (f); 0, 0, 3 and 0, 1, 1, 10* in f; hepatocellular carcinoma in
diet provided 4 h/d; m 0, 0, 0, 3* and 1, 0, 1, 3 in f
none
Table 29. (cont'd)
Species; strain; Dose; 1) Effects other than tumours (dose in mg/kg bw/d); Reference
initial number of route of exposure; 2) number of animals for histopathological evaluation of
animals per dose; exposure period; neoplastic effects;
vehicle frequency of treatment; 3) type of tumour: number of animals with this tumour/dose
post-exposure group (unless otherwise given)
observation period
Rat; 0, 3.33, 16.6, 50 1) > 16.6 mg/kg: survival ”# (m); 3.33 & 50: bw ”# (m); Maltoni
Sprague-Dawley; mg/kg bw./d; 2) 40 f and 40 m per group; et al.
40 rats/sex; pure gavage; 52 weeks; 3) ASL in m: 0, 0, 4, 8* and in f: 0, 0, 6*, 9*; (1981,
virgin olive oil once daily, Zymbal gland carcinoma in m: 0, 0, 1, 1 and in f: 1, 0, 1984)
4-5 d/week; up to 1, 0; extrahepatic angiosarcoma in m: 0 in all groups and [BT 11] b
84 weeks c in f: 0, 2, 0, 2; nephroblastoma in m: 0, 0, 2, 1 and in
f: 0, 0, 1, 1;
Rat; 0, 0.03, 0.3, 1.0 1) > 0.03 mg/kg: survival and bw <->#; Maltoni
Sprague-Dawley; 75 mg/kg bw./d; 2) 75, 75, 73, 75 m and 75, 75, 73, 75 f; et al.
rats/sex; pure gavage; 59 weeks; 3) ASL in m: 0, 0, 0, 1 and in f: 0, 0, 1, 2; (1981,
virgin olive oil once daily, Zymbal gland carcinoma in m: 0, 0, 0, 2 and in f: 1, 0, 1984)
4-5 d/week; up to 0, 3; extrahepatic angiosarcoma in m: 0 in all groups and [BT 27] b
77 weeks c in f: 0, 0, 0, 1; nephroblastoma 0 in all groups
Rat; Wistar; 0, 3, 30, 300 mg/kg 1) 3 mg/kg: bw <-> , rat skin composition <->, mortality Knight &
10-20 rats/sex; bw./d; gavage; 1/16 (control n.g.) 30 mg/kg: bw <->, mortality 5/15; Gibbons
peroxide-free 95-125 weeks; biochemical parameters for skin fibrosis !* 300 mg/kg: (1987)
corn oil once daily; none mortality 15/15 within first 60 days of treatment;
2) 20, 16, 15, 10;
3) Liver tumors (predominantly angiosarcomas):
0, 1, 11, 10 #
a In all studies vehicle-treated controls; analysis of VC concentration in vehicle in all studies except Knight & Gibbons (1987);
bw. = body weight; d = day; f = females; m = males; mo = months; n.g. = not given; * = effect significant at P < 0.05;
# = no statistical evaluation; <-> = unchanged; ” = decreased; ! = increased
b Study number in experiments done by Maltoni and coworkers
c Animals were kept until spontaneous death or sacrificed at the end of given post exposure observation period
d Bioavailability studied in an ancillary study
e Study comparable to OECD guidelines 451 with acceptable restrictions (OECD, 1981a,b)
f Study comparable to OECD guidelines 453 (OECD, 1981a,b)
A statistically significant increase in the incidence of ASL was
seen in Sprague-Dawley rats at a dose level of 16.6 mg VC/kg body
weight per day (gavage; study BT 11), and some similar tumours were
also observed at dose levels of 0.3 and 1.0 mg/kg body weight per day
(Maltoni et al., 1981). Hepatic angiosarcomas were also observed in
gavage studies in Wistar rats (Knight & Gibbons, 1987). This tumour
type is very rare in untreated rats (4 in several thousand rats of the
colony used by Maltoni et al., 1981). Because of short dosage and
follow-up in the first study, and small number of animals due to early
mortality in the third, these studies probably did not reflect the
total carcinogenic potential of VC.
The results of the two feeding studies carried out by another
working group (Feron et al., 1981; Til et al., 1983, 1991) confirmed
the findings presented by Maltoni et al. (1981) concerning ASL
(induction at 1.3 mg/kg body weight per day, significant at 5.0 mg
VC/kg body weight per day). Furthermore these feeding studies
presented evidence for significantly increased tumour incidences of
neoplastic liver nodules (females) and of hepatocellular carcinoma
(HCC) (males) at 1.3 mg/kg body weight per day (Til et al., 1991).
7.7.2 Inhalation exposure
After the first reports of the carcinogenicity of VC in rats
(Viola, 1970; Viola et al., 1971), Maltoni and coworkers intensively
investigated the effects of inhalation exposure to VC in different
laboratory animals.a Study results were published in Maltoni et al.
(1974), Maltoni & Lefemine (1975) and Maltoni et al. (1979), and the
final report in Maltoni et al. (1981, 1984). Detailed information on
these inhalation studies as well as pertinent studies from other
working groups is tabulated in Table 30 (short-term studies) and Table
32 (long-term studies). A summary of tumour types induced by long-term
inhalation exposure to vinyl chloride in different species is given in
Table 33. Further studies on inhalative carcinogenicity of VC not
discussed in this section, but leading to similar results to the
studies tabulated below, were performed on rats (short-term: Hong et
al., 1981; long-term: Viola et al., 1971; Maltoni et al., 1974; Lee et
al., 1977; Maltoni et al., 1981, 1983, 1984; Bi et al., 1985; Maltoni
& Cotti, 1988; Froment et al., 1994) and mice (short-term: Suzuki,
1978, 1981; Schaffner, 1979; Hehir et al., 1981; Himeno et al., 1983;
Adkins et al., 1986; long-term: Keplinger et al., 1975; Lee et al.,
1977; Holmberg et al., 1979; Drew et al., 1983).
7.7.2.1 Short-term exposure
Several carcinogenicity studies have been performed where animals
were exposed by inhalation for rather short periods (up to 6 months)
and kept for different post-exposure periods up to lifetime
(Table 30). The tumour spectrum in rats and mice is very similar to
a Studies performed at Beautivoglio (BT) Laboratories; studies
numerated BT1-BT27
Table 30. Short-term inhalation studies on carcinogenicity of vinyl chloride in experimental animalsa
Species; strain; Doses in mg/m3; 1) Number of animals for histopathological References
age at start of exposure period; evaluation;
experiment; initial frequency of 2) type of tumour: number of animals with
number of animals per treatment; this tumour/dose group (unless otherwise
dose per exposure post-exposure given); other observations
period; type of observation period
control;
Rat; 0, 15 600, 26 000; 1) 227, 120, 118; Maltoni
Sprague-Dawley; 5 weeks; 4 h/d, 2) (#) ASL: 0, 0, 1; extrahepatic et al.
11 weeks; 120 5 d/week; angiosarcoma: 0, 0, 0; Zymbal gland (1981)
(control 240) 149 weeksc carcinoma: 0, 9, 9; hepatoma: 0, 0, 1; [BT 10]b
f & m; untreated nephroblastoma: 0, 1, 0;
control neuroblastoma: 0, 1, 0
Rat; 0, 15 600, 26 000; 1) 227, 118, 119; Maltoni
Sprague-Dawley; 25 weeks; 1 h/d, 2) (#) ASL: 0, 3, 1; extrahepatic angiosarcoma: et al.
11 weeks; 120 4 d/week; 0, 2, 0; Zymbal gland carcinoma: 0, 5, 9 (1981)
(control 240) 129 weeksc [BT 10]b
f & m; untreated
control
Rat; 15 600, 26 000; 1) 18 m and 24 f at low dose, 24 m and Maltoni
Sprague-Dawley; 5 weeks; 4 h/d, 20 f at high dose; et al.
1-day-old; 5 d/week; 119c 2) ASL in m: 5, 6 and in f:12, 9; (1981)
see number of extrahepatic angiosarcoma in m: 5, 6 [BT 14]b
animals for and inf: 12, 9; hepatoma in m: 9, 13
evaluation; no control; and in f: 11, 7
Rat; 0 or 2465; 1) 84-128 rats per sex per age group; Groth
Sprague-Dawley; 24.5 weeks; 7 h/d, 2) angiosarcomas (mostly in liver) in the 4 et al.
6, 18, 32, or 52 weeks; 5 d/week; scheduled age groups in exposed m: 1.2, 2.2, 7.4, (1981)
110-128 rats/sex sacrifice at 3, 6 and 18% and in exposed f: 2.3, 7.2, 28, 13%;
per age group; 9 months, only 1 subcutaneous angiosarcoma (m) in
treated control termination at all control groups (739 mice) #; older
week 43 adults more susceptible to
angiosarcoma-inducing effect
Table 30. (cont'd)
Species; strain; Doses in mg/m3; 1) Number of animals for histopathological References
age at start of exposure period; evaluation;
experiment; initial frequency of 2) type of tumour: number of animals with
number of animals per treatment; this tumour/dose group (unless otherwise
dose per exposure post-exposure given); other observations
period; type of observation period
control;
Rat; Wistar or 0, 6.5, 13, 26, 52, 1) 2-27 rats of each sex per dose per strain; Laib
Sprague-Dawley; 104, 208; 3 weeks; 2) dose dependent increase in ATPase-deficient et al.
3 days; 2-27 rats 8 h/d, 5 d/week; liver foci (preneoplastic lesion), (1985a)
per sex per strain; 10 weeks presumably significant at > 52 mg/m3
treated control ( m Sprague-Dawley rats at > 104 mg/m3);
f more susceptable than m in both strains
Rat; Wistar; 0 or 5200; 1) 2-10 rats of each sex per dose per exposure Laib
age-dependency 5-83 days, see 2); period; et al.
studied early in life, 8 h/d, 7 d/week; rats 2) age dependent increase in ATPase-deficient (1985b)
see 2); 4-10 rats sacrificed at the age liver foci lesion) (preneoplastic studied,
per sex; of 4 months rats exposed a) during gestation or at
treated control postnatal day 1-5 (b), 1-11 (c), 1-17 (d),
1-47 (e), 1-83 (f), 7-28 (g), or 21-49 (h);
no effect in a & b, foci area steeply
increased in c and d but not further
enhanced in e and f, foci area not lowered
in g but only a few foci in h
Mouse; CD1; 0, 2.6, 26, 260, 780, 1) see 2); Suzuki
5-6 weeks; 1560; 4 weeks; 2) pulmonary tumour incidences (#):week 0 (1983)
30-60 m; treated 6 h/d, 5 d/week; post-exposure: no tumour week 12: 0/18,
control 0, 12, 40 weeks 0/10, 0/9, 0/6, 6/9, 8/9 week 40: 0/17, 1/9,
3/9, 6/9, 5/7, 6/7; study focused on lung
tumours (no details on other tumours);
tumours derived from type II alveolar cells
Table 30. (cont'd)
Species; strain; Doses in mg/m3; 1) Number of animals for histopathological References
age at start of exposure period; evaluation;
experiment; initial frequency of 2) type of tumour: number of animals with
number of animals per treatment; this tumour/dose group (unless otherwise
dose per exposure post-exposure given); other observations
period; type of observation period
control;
Mouse; CD-1; 0, 130, 650, 2600; 1) 60, 40, 44, 38 m and 60, 40, 40, 38 f Hong et al.
2 months; 1, 3, 6 months; for cumulative effects; (1981)
8-28 mice/sex; 6 h/d, 5 d/week; 2) cumulative (f+m); ASL: 1, 2, 13*, 18*;
treated control 12 months extrahepatic angiosarcoma 0, 2, 6*, 2;
bronchioloalveolar tumor 16, 18, 52*, 50*;
cumulative (f): metastatic lung
adenocarcinoma 0, 4*, 2, 6*; mammary
carcinoma 4, 10*, 13*, 6; tumours at
different organ sites in f and m
increased in proportion to dose or duration
of exposure at higher dose levels
a d = day; f = females; m = males; n.g. = not given; # = no statistical evaluation; * = significant at P < 0.05
b Study number in experiments done by Maltoni and coworkers
c Animals were kept until spontaneous death or sacrificed at the end of given post-exposure observation period
that observed in long-term inhalation studies (compare with Table 32).
Hehir et al. (1981) presented evidence that even one single high dose
(> 13 000 mg/m3) resulted in a dose-related increase of pulmonary
tumours in mice. This effect was also demonstrated by repeated
administration to mice at lower dose levels with exposure periods
varying between 1 and 6 months (Hong et al., 1981; Suzuki, 1983).
Incidences of ASL, extrahepatic angiosarcoma and mammary gland
carcinoma were elevated in mice with increasing exposure periods at
doses of 130 or 650 mg/m3 (Hong et al., 1981). Also in rats,
angiosarcomas, predominantly in the liver, were induced by VC
inhalation for 25 weeks (15 600 mg/m3, 1 h/day, Maltoni et al., 1981;
2465 mg/m3, 7 h/day, Groth et al., 1981). Short-term exposure led to
increased incidences of carcinoma of the Zymbal gland, a sebaceous
gland of the ear canal in rats, at high doses (> 15 600 mg/m3;
Maltoni et al., 1981, BT10).
Short-term exposure studies on the effect of age on
susceptibility to tumour induction are discussed in section 7.7.3.
7.7.2.2 Long-term exposure
a) Rats
Although differences in experimental design, notably in the
duration of follow-up, and the low overall frequency of tumours tend
to confuse the picture, a dose-response relationship could be observed
in studies BT1, BT2, BT9 and BT15 on Sprague-Dawley rats for ASL
(Table 31, Fig. 4 and chapter 10) and Zymbal gland carcinomas, while
it was less clear for nephroblastomas, neuroblastomas and mammary
malignant tumours.
Lee et al. (1978) and Drew et al. (1983) reported similar
findings on the incidence of ASL. Female rats seem to be more
susceptible to ASL tumours than males (Lee et al., 1978; Maltoni et
al., 1981). ASL was detected at 26 mg/m3 and the incidence was
statistically significant at 130 mg/m3, which was at a lower level
than other tumour types with the exception of mammary gland tumours
(13 mg/m3; see Table 33) (Maltoni et al., 1981). VC exposure also
caused angiosarcomas at extrahepatic sites (Maltoni et al., 1981, BT14
see Table 30; Lee et al., 1978; Drew et al., 1983).
Increased incidences of other tumour types than those reported by
Maltoni et al. (1981) in Sprague-Dawley rats were shown by Drew et al.
(1983) in Fischer-344 rats (neoplastic liver nodules and
hepatocellular carcinoma) and by Feron & Kroes (1979) in Wistar rats
(tumour of the nasal cavity). These were single dose experiments.
Drew et al. (1983; see Table 32) studied the effect of exposure
duration (6, 12, 18 or 24 months) on tumour incidences and
demonstrated that longer exposure periods, and thus greater cumulative
exposures, led to an increase in tumorigenic responses concerning
liver angiosarcoma, angiosarcoma at all sites, and hepatocellular
Table 31. Incidence of angiosarcomas of the liver (ASL) and mammary
tumours observed in Sprague-Dawley ratsa
Exposure Concentration ASL ASL ASL Mammary
(ppm) (mg/m3) (males) (females) (males + adenocarcinomas
females)
0 0 0/173 0/239 0/412 12/239
1 2.6 0/58 0/60 0/118 12/60
5 13 0/59 0/60 0/119 22/60
10 26 0/59 1/60 1/119 11/60
25 65 1/60 4/60 5/120 15/60
50 130 2/174 13/180 15/354 61/180
100 260 0/60 1/60 1/120 3/60
150 390 1/60 5/60 6/120 6/60
200 520 7/60 5/60 12/120 5/60
250 650 1/29 2/30 3/59 2/30
500 1300 0/30 6/30 6/60 1/30
2500 6500 6/30 7/30 13/60 2/30
6000 15 600 3/29 10/30 13/59 0/30
10 000 3/30
a The data are combined from the experiments BT1, BT2, BT9 and BT15 conducted by Maltoni et al. (1984).
Control animals from several experiments are combined in the 0 ppm group. Animals were exposed
4 h/day, 5 days/week for 52 weeks; the follow-up was until the death of the animals or until week
83 (BT1), 90 (BT2), 90 (BT9) or 95 (BT15). Tumours were scored at the time of death. Adapted from
Reitz et al. (1996)
carcinoma in female F-344 rats. This exposure duration-related effect
was not observed in mice and hamsters, except angiosarcoma at all
sites in hamsters.
The lowest reported dose (13 mg/m3) to cause a statistically
significant increase in tumour incidences was demonstrated for mammary
adenocarcinoma in female rats (Maltoni et al. 1981; BT15). Although
this tumour type is common in untreated rats and no clear
dose-response relationship was shown (probably due to reduced survival
in the high-dose groups, Maltoni et al., 1984; BT1, BT2, BT9, BT15),
the increase in tumour rate was considered by the Task Group to be of
toxicological relevance, since the incidence was increased compared to
concurrent and historical controls of the same colony (Maltoni et al.,
1981, 1984) and similar results on mammary gland tumours were
presented in further studies on rats (Drew et al., 1983) and other
species (Lee et al., 1978; Maltoni et al., 1981, 1984; Drew et al.,
1983) (see also Table 33; Fig. 4).
b) Mice
In mice, the spectrum of tumours induced by long-term inhalation
exposure is similar to that observed in rats, but an increase in lung
tumours was only observed in mice (Lee et al., 1978; Maltoni et al.,
1981; Drew et al., 1983, see also Table 32). At a dose level of
130 mg/m3, incidences in angiosarcoma (liver, extrahepatic or all
sites combined), lung carcinoma, and mammary gland carcinoma showed a
statistically significant increase (Lee et al., 1978; Drew et al.,
1983). Doses lower than 130 mg/m3 were not investigated, but at the
dose levels investigated, the angiosarcoma frequencies were higher in
mice than in rats (Lee et al., 1978). No clear-cut relationship with
time of exposure (6-18 months) and incidence of ASL was observed; this
could, however, have been due to decreased survival and shorter
follow-up after longer exposure time (Drew et al., 1983).
c) Other species
Limited data are available on other species. In experiment BT8
with male Syrian golden hamsters, low frequencies of ASL, acoustic
duct tumours and melanomas were observed. In addition, an increased
incidence in forestomach and skin epithelial tumours was reported, but
these tumours were also observed in controls, there was no
dose-response relationship and a statistical evaluation was not
presented (Maltoni et al., 1981).
Female hamsters were exposed to a single dose for different
exposure periods (Drew et al., 1983). Angiosarcomas (all sites), skin
tumours and increased incidences in mammary gland carcinoma and
stomach adenomas (glandular portion of the stomach) were reported
(Drew et al., 1983).
In a study reported only as an abstract, Caputo et al. (1974)
reported increased incidences in skin acanthoma and lung
adenocarcinoma in rabbits exposed to VC.
Table 32. Inhalation studies on the long-term toxicity/carcinogenicity of vinyl chloride in experimental animalsa
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Rat; 0, 130, 650, 1300, 1) > 650: survival ”# (f, in m at > 1300); Maltoni et
Sprague-Dawley; 13 6500, 15 600, > 15 600: body weight ”# (m, in f at 26 000); al. (1981,
weeks; 30 rats/sex; 26 000; 52 weeks; 2) 29-30 m and 29-30 f per group; 1984)
untreated control 4 h/d, 5 d/week; 3) ASL in m: 0, 0, 1, 0, 6*, 3, 3 and in [BT1]b
83 weeks c f : 0, 1, 2, 6*, 7*, 10*, 4; Zymbal gland
carcinoma in m: 0, 0, 0, 3, 1, 3, 10* and in f:
0, 0, 0, 1, 1, 4, 6*; hepatoma in f:
0, 0, 0, 5, 2, 1, 0;
nephroblastoma in m: 0, 0, 1, 2, 5*, 4, 3
and in f: 0, 1, 4*, 4*, 1, 1, 2;
neuroblastoma in m: 0, 0, 0, 0, 2, 2, 2 and in f:
0, 0, 0, 0, 2, 1, 5*;
mammary adenocarcinoma in f: 0, 2, 2, 1, 2, 0, 3
Rat; 0, 260, 390, 520; 1) > 130: survival ”# (m, in f at 520); Maltoni et
Sprague-Dawley; 52 weeks; 4 h/d, 2) 59-60 rats per treatment group, control al. (1981,
13 weeks; 60 5 d/week; 91 weeks c 65 m and 100 f; 1984)
rats/sex, control 3) ASL in m: 0, 0, 1, 7* and in f: 0, 1, 5*, 5*; [BT2]b
85 m and 100 f; mammary adenocarcinoma in f: 1, 3, 6*, 5;
untreated control nephroblastoma in m: 0, 8*, 8*, 5* and in f:
0, 2, 3, 2
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Rat; Sprague-Dawley; 0, 130; 52 weeks; 1) survival ”# (f); Maltoni et
13 weeks; 150 rats 4 h/d, 5 d/week; 2) 144 m and 150 f, control 48 m and 50 f; al. (1981,
per sex, control 50 m 90 weeks c 3) ASL in m: 0, 2 and in f: 0, 12*; 1984)
and 50 f; mammary adenocarcinoma in f: 5, 59*; [BT9]b
untreated control no significant effects on incidences of other
tumour types
Rat; 0, 2.6, 13, 26, 65; 1) > 2.6 survival ”# (m+f); Maltoni et
Sprague-Dawley; 52 weeks; 4 h/d, 2) 58-60 m and 60 f per group; al. (1981,
13 weeks; 60 rats/sex; 5 d/week; 3) ASL in m: 0, 0, 0, 0, 1 and in f: 0, 0, 0, 1, 1984)
untreated control 95 weeks c 4; mammary adenocarcinoma in f: 6, 12, 22*, [BT15]b
21*, 15*; nephroblastoma in m: 0, 0, 0, 0, 1;
hepatoma in f: 0, 0, 0, 0, 1
Rat; Spague-Dawley; 0, 6500; life time 1) survival & body weight ”# Maltoni &
13 weeks (pregnant (up to 69 weeks); 2) 60, 54 dams, progeny 158, 63 m and 149, 64 f Cotti (1988)
rats) or gestation day 4 h/d, 5 d/week 3) tumour incidences in % (#):
12 (embryos); see for 7 weeks, than ASLd : 0, 50.0 in dams and 0, 56.2 in m and
number of animals for 7 h/d, 5 d/week; 0, 73.0 in f progeny; hepatocellular carcinoma:
for evaluation; none 0, 9.2 in dams, 0.6, 42.2 in m and 0, 60.3 in f
untreated control progeny mammary carcinoma: 7.4, 7.4 in dams, 5.4,
4.7 in f progeny; neuroblastomad: 0, 59.2 in dams,
0, 48.4 in m and 0, 42.8 in f progeny
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Rat; Fischer-344; 0 or 260; 6, 12, 1) survival ”* (exposure periods > 6 mo); Drew et al.
8-9 weeks, 2nd 18, 24 mo, 2nd 2) 112 (control), 76, 55, 55, 55 per exposure (1983)
experiment: 2, 8, experiment 6 or period; 2nd experiment 6 mo exposure: 112
14, 20 mo (6 mo 12 mo exposure (control), 76, 52, 51, 53 per age group; 2nd
exposure) or 2, 8, at different age; experiment 12 mo exposure: 112 (control), 55,
14 mo (12 mo 6 h/d, 5 d/week; 54, 49 per age group;
exposure); n.g. (only life span 3) ASL: 1, 4*, 11*,13*, 19*; angiosarcoma all
f, see number of sites: 2, 4, 12*, 15*, 24*; mammary fibroadenoma:
rats for evaluation); 24, 28*, 28*, 24*, 26*; mammary adenocarcinoma:
n.g. 5, 6, 11*, 9*, 5*; neoplastic liver nodules: 4,
15*, 20*, 7*, 6*; hepatocellular carcinoma: 1, 3,
4*, 8*, 9*; 2nd experiment 6 mo exposure: ASL: 1,
4*, 2, 0, 0; angiosarcoma all sites: 2, 4, 2, 0,
0; mammary fibroadenoma: 24, 28*, 23*, 17, 20;
mammary adenocarcinoma: 5, 6, 2, 3, 2; neoplastic
liver nodules: 4, 15*, 10*, 2, 4; hepatocellular
carcinoma: 1, 3, 6*, 0, 1;
2nd experiment 12 mo exposure: ASL: 1, 11*, 5*,
2; angiosarcoma all sites: 2, 12*, 5*, 2; mammary
fibroadenoma: 24, 28*, 16*, 15; mammary
adenocarcinoma: 5, 11*, 4, 0; neoplastic liver
nodules: 4, 20*, 4, 4; hepatocellular carcinoma:
1, 4*, 1, 0;
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Rat; Wistar; 11 0, 130, 650, 1300, 1) > 130: survival & body weight ”# Maltoni et
weeks; control 40 m, 6500, 15 600, 2) 38, 28, 28, 27, 25, 26, 27; al. (1981,
other groups 30 m; 26 000; 52 weeks; 3) (#):ASL: 0, 0, 1, 3, 3, 3, 8; hepatoma: 0, 0, 1984)
untreated control 4 h/d, 5 d/week; 0, 0, 1, 2, 0; nephroblastoma: 0, 1, 0, 2, 0, [BT17]b
113 weeks c 2, 1; neuroblastoma: 0, 0, 0, 0, 1, 1, 3;
Zymbal gland carcinoma: 0, 0, 0, 0, 0, 2, 2
Rat; Wistar; newly 0 or 13 000; 1) mortality !# (9 m and 10 f still alive at week Feron et al.
weaned; 62 rats/sex; 52 weeks, 10 rats 52); body weights ”* (f+m); blood clotting time (1979a,b)
treated control per dose per sex ”#; liver function ”# (BSB-retention test); Feron &
sacrificed at week relative liver, kidney and spleen weight !*; Kroes
4, 13, 26 (see degree of tubular nephrosis !# (f+m); focal (1979)
section 7.2); 7 h/d, degeneration of myocardium and thickened walls
5 d/week; none of arteries # (f+m); haematopoietic activity
in the spleen !# (f+m); distended liver
sinusoids # (f+m);
2) 62 m and 62 f per group;
3) cumulative tumours (#):ASL: 0, 3 in m and 0, 6
in f ; Zymbal gland tumour: 0, 7 in m and 0, 4
in f; tumour of nasal cavity: 0, 10 in m and
0, 10 in f;
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Rat; CD; 2 mo; 0, 130, 650, 2600; 1) > 650: survival ”# (m+f); Lee et al.
36 rats/sex; 12 mo, 4 rats/dose 2) 35, 36, 36, 34 m and 35, 36, 34, 36 f; (1978)
treated control per sex terminated 3) tumours combined for all exposure periods:
at month 1, 2, 3, 6, ASL: 0, 0, 2, 6 in m and 0, 0, 10*, 15* in f;
9; 6 h/d, 5 d/week; lung angiosarcoma: 0, 0, 0, 4 in m and 0, 0,
none 3, 9* in f; other tumours not related to VC
treatment
Rat [no further data 0, 14, 25, 266, 1) no data Kurlyandski
provided] 3690; 52 weeks; 2) 70, 50, 39, 43, 51 m et al.
4.5 h/d, 5 d/week 3) ASL: 0, 0, 7.7, 9.3, 11.8%; (1981)
other angiosarcomas: 0, 1.0, 2.5, 0, 3.9%
other liver tumours: 2.8, 2.0, 2.6, 11.6, 13.7%
haemoblastoma: 4.3, 14.0, 15.4, 34.9, 2.0%
other tumours: 21.5, 28.1, 15.4, 9.3, 23.5%
Mouse; Swiss; 0, 130, 650, 1300, 1) > 130: survival ”# (m+f); > 1300: body weight ”# Maltoni et
11 weeks; 30 mice 6500, 15 600, (m, in f at > 650); al. (1981,
per sex, control 26 000; 30 weeks; 2) 80, 30, 30, 30, 29, 30, 26 m and 70, 30, 30, 1984)
80 m and 70 f; 4 h/d, 5 d/week; 30, 30, 30, 30 f; [BT4]b
untreated control 51 weeks c 3) combined (f+m) tumours (#): ASL: 0, 1, 18, 14,
16, 13, 10; liver angioma: 0,1, 11, 5, 5, 7, 6;
extrahepatic angiosarcoma: 1, 1, 3, 7, 8, 1, 1;
lung tumour: 15, 6, 41, 50, 40, 47, 46;
mammary carcinoma in f : 1, 13, 12, 9, 10, 9, 14
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Mouse; Swiss CD-1; 0 or 130; 6, 12, 1) survival ”* (all exposure periods); Drew et
8-9 weeks, 2nd 18 mo, 2nd 2) 71 (control), 67, 47, 45 per exposure period; al. (1983)
experiment: 2, 8, experiment 6 or 12 mo 2nd experiment 6 mo exposure: 71 (control), 67,
14 mo (6 or 12 mo exposure at 49, 53 per age group; 2nd experiment 12 mo
exposure); n.g. (only different age; exposure: 71 (control), 47, 46, 50 per age group;
f, see number of 6 h/d, 5 d/week; 3) angiosarcoma (all sites): 1, 29*, 30*, 20*;
mice for evaluation); lifespan mammary gland carcinoma: 2, 33*, 22*, 22*; lung
n.g. carcinoma: 9, 18*, 15*, 11*; 2nd experiment 6 mo
exposure: angiosarcoma (all sites): 1, 29*, 11*,
5; mammary gland carcinoma: 2, 33*, 13*, 2; lung
carcinoma: 9, 18*, 13*, 7;
2nd experiment 12 mo exposure: angiosarcoma
(all sites): 1, 30*, 17*, 3;
mammary gland carcinoma: 2, 22*, 8*, 0; lung
carcinoma: 9, 15*, 9*, 3;
Mouse; CD-1; 2 mo; 0, 130, 650, 2600; 1) > 650: survival ”# (tumour development), all Lee et al.
total 36 mice/sex; 1, 2, 3, 6, 9, and mice in high-dose group and females in mid-dose (1978)
treated control 12 mo (4 mice per group died or were terminated at mo 10-12;
group per sex per 2) 26, 29, 29, 33 m and 36, 34, 34, 36 f;
exposure period); 3) tumours combined for all exposure periods:
6 h/d, 5d/week; bronchioloalveolar adenoma (#):1, 8, 10, 22 in m
none and 0, 4, 12, 26 in f; ASL: 0, 3, 7*, 13* in m
and 0, 0, 16*, 13* in f; extrahepatic angiosarcoma:
0, 5*, 2, 0 in m and 0, 1, 3, 9* in f; mammary
tumours(#): 0, 9, 3, 13 in f; incidences exposure
time-dependent
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Hamster; Syrian 0, 130, 650, 1300, 1) > 130: survival ”#; Maltoni et
golden; 11 weeks; 6500, 15 600, 2) see initial number of hamsters; al. (1981,
30 m, 60 m in 26 000; 30 weeks; 3) (#): ASL: 0, 0, 0, 2, 0, 1, 0; 1984)
control; 4 h/d, 5 d/week; acoustic duct tumour: 0, 0, 0, 3, 1, 2, 1; [BT8]b
untreated control 79 weeks c melanoma: 0, 1, 1, 0, 0, 1, 2, 1;
forestomach tumour: 3, 3, 4, 9, 17, 10, 10;
skin epithelial tumour: 3, 9, 3, 7, 3, 1, 7;
Hamster; Syrian 0 or 520; 6, 12, 1) survival ”* (all exposure periods); Drew et al.
golden; 8-9 weeks, 18 mo, 2nd 2) see tumour incidences; (1983)
2nd experiment: 2, 8, experiment 6 or 3) tumour incidences in control and at different
14, 20 mo (6 mo 12 mo exposure at exposure periods: angiosarcoma (all sites): 0/143
exposure) or 2, 8, different age; (control), 13/88*, 4/52*, 2/103; mammary
14 mo (12 mo 6 h/d, 5 d/week; carcinoma: 0/143, 28/87*, 31/52*, 47/102*;
exposure); n.g. life span stomach adenoma: 5/138, 23/88*, 3/50*, 20/101*;
(only f, see number skin carcinoma: 0/133, 2/80; 9/48*, 3/90; 2nd
of hamster for experiment 6 mo exposure: angiosarcoma (all
evaluation); n.g. sites): 0/143 (control), 13/88*, 3/53*, 0/50,
0/52; mammary carcinoma: 0/143, 28/87*, 2/52*,
0/50, 1/52; stomach adenoma: 5/138, 23/88*,
15/53*, 6/49*, 0/52; skin carcinoma: 0/133, 2/80;
0/49, 0/46, 0/50; 2nd experiment 12 mo exposure:
angiosarcoma (all sites): 0/143 (control),
4/52*, 1/44, 0/43;
mammary carcinoma: 0/143, 31/52*, 6/44*, 0/42;
stomach adenoma: 5/138, 3/50*, 10/44*, 3/41;
skin carcinoma: 0/133, 2/80; 0/38, 0/30;
Table 32. (cont'd)
Species; strain; age Doses in mg/m3; 1) Effects other than tumours Reference
at start of exposure period; (dose in mg/m3);
experiment; initial frequency of treatment; 2) number of animals for
number of animals per post-exposure histopathological evaluation of
dose per exposure observation period neoplastic effects;
period; type of 3) type of tumour: number of animals
control; with this tumour/dose group (unless
otherwise given)
Rabbit; n.g.; n.g.; 0, 26 000; 1) n.g.; Caputo
40 exposed to VC, 15 mo; 4 h/d, 2) see tumour incidences; et al.
20 controls (no data 5 d/week; n.g. 3) tumour incidences: skin acanthoma: 0/20 (1974)
about sex); (control), 12/40*; lung adenocarcinoma 0/20, 6/40
treated control
a * = significant at P < 0.05; #: no statistical evaluation; f = females; m = males; mo = months; n.g. = not given
b Study number in experiments done by Maltoni and coworkers
c Animals were kept until spontaneous death or sacrificed at the end of given post exposure observation period
d Presumably significant in all exposed groups
Table 33. Summary of tumour types induced by long-term inhalation exposure to vinyl chloride
Lowest reported dose that significantly increased tumour incidences by tumour type (dose in mg/m3)
Species Liver Angiosarcoma Other Lung Mammary Nephro- Skin Neuro- Stomach Zymbal
angiosarcoma (other sites) liver carcinoma gland blastoma tumour blastoma tumour gland
tumours carcinoma tumour
Rat 130 in f c extrahepatic neoplastic 13 in 260 in 6500 in fore- 26 000
520 in m c 260 in f b,d nodules f h m b,g f b,k stomach in f + m f
lung 260 in f 650 in papilloma
2600 in f e b,d f g 78 000 in
hepato- f + m i
cellular
carcinoma
260 in
f d
Mouse 650 in 130 in m b,e 130 in 130 in
f + m e f b,d f b,d
all sites
130 in f b,d
Rabbit acanthoma
26 000,
n.d.
about sex
Hamster all sites 520 in b,j adenoma
520 in f b,d carcinoma 520 in
f b,d 520 in f f b,d
b,d
a Exposure in all studies 4-7 h/day, 5 days/week, exposure period at least 6 months
f = females; m = males; n.d. = no data
b The lowest dose tested with the study design described in the cited study
c Maltoni et al. (1981; BT9) f Maltoni et al. (1981; BT1) i Maltoni et al. (1981; BT6)
d Drew et al. (1983) g Maltoni et al. (1981; BT1, BT2) j Caputo et al. (1974)
e Lee et al. (1978) h Maltoni et al. (1981; BT15) k Maltoni & Cotti (1988)
7.7.3 The effect of age on susceptibility to tumour induction
Recently there has been some concern about early-life sensitivity
to vinyl chloride (Hiatt et al., 1994; Cogliano et al., 1996).
However, there is contradictory evidence (Drew et al., 1983 versus
Groth et al., 1981) concerning the effects of age on ASL induction in
rats, but final conclusions on increased susceptibility in 6- to
8-weeks-old animals could not be drawn from these data. The
discrepancy in results with rats is probably due to differences in
strain and/or experimental design. However, there is evidence from
other studies that there is possibly a higher sensitivity to liver
tumour induction in different rat strains in the first weeks of life,
a life-phase much earlier than that studied by Drew et al. (1983).
Studies on DNA adduct formation support these results.
F-344 rats, hamsters and mice (Swiss and B6C3F1) of different
age at the beginning of the exposure period (2, 8, 14 or 20 (not mice)
months old) were exposed to VC using the same experimental design and
same exposure period (Drew et al., 1983; see Table 32 for details).
For ASL in rats, angiosarcoma at all sites and mammary carcinoma
in all three species, neoplastic nodules in rats and lung carcinoma in
Swiss mice, the tumour response was highest in young animals
(2-month-old) exposed for 6 or 12 months. The validity of the
demonstrated age-related effects is limited in this study because only
statistical evaluation in comparison to control or to all other
groups, but not related to each other exposure group, was performed.
In addition, exposure later in life automatically shortens the period
of follow-up, and thus tends to lead to an apparent elevated
sensitivity at young age.
The effect of age on the susceptibility to induction of ASL in
Sprague-Dawley rats was also studied by Groth et al. (1981; see Table
30 for details). Rats aged 6, 18, 32 or 52 weeks were exposed to VC at
the same dose level and exposure period. In contrast to Drew et al.
(1983), the results of this study demonstrated that the older the rats
were at the start of the exposure period, the greater was the tumour
incidence. The maximum incidence was observed in male rats 52 weeks
old at first exposure and in females 32 weeks old at first exposure
(significantly increased compared to 6- or 18-week-old females). For
males the effect of age was statistically significant.
Exposure of 1-day-old rats of the same strain to high VC
concentrations for 5 weeks (BT14, see Table 30) revealed a remarkably
higher incidence of ASL, extrahepatic angiosarcoma and hepatoma
compared with 11-week-old rats exposed within the same experimental
design (BT10; Maltoni et al., 1981; see Table 30). However, these
experiments were not concurrent, and the tumour response in different
series from this laboratory has been variable (Tables 31, 32).
Maltoni & Cotti (1988) exposed 13-weeks-old pregnant
Sprague-Dawley rats and their progeny from the 12th day of gestation
to VC (Table 32). Although no statistical evaluation was performed it
seems that there was no significant difference between the dams and
the progeny concerning incidences in ASL, mammary carcinoma, and
neuroblastoma. The incidence of hepatocellular carcinoma, however, was
9% in dams and 42% in male and 60% in female progeny. In this study
the duration of exposure (and also of latency) was up to 69 weeks for
the progeny, but 56 weeks for the dams.
Laib et al. (1985a; see Table 30) presented evidence for
dose-related increased induction of ATPase-deficient liver foci in
Wistar and Sprague-Dawley rats, discussed as preneoplastic
hepatocellular lesions, after a 3-week exposure to low concentrations
(> 52 mg/m3). The increased induction of these foci was restricted
to a well-defined period of highest sensitivity beginning with rapid
liver growth in 7- to 21-day-old rats. No induction of these foci was
reported in adult rats exposed to 5200 mg/m3 for 70 days after
partial hepatectomy (no further details) (Laib et al., 1985b).
Comparative investigations on the alkylation of liver DNA in
young and adult Wistar rats exposed under the same exposure conditions
confirmed the age-related sensitivity of rats to VC (section 6.5.1 and
Table 26.)
7.7.4 The effect of gender on susceptibility to tumour induction
There is evidence that female rats of various strains are more
susceptible to liver tumour induction than males. Maltoni et al.
(1981, 1984) reported in all studies on Sprague-Dawley rats (BT1, BT2,
BT9, BT15; see Table 32) a higher incidence of ASL in female rats
compared with males after long-term inhalation. Similar results were
presented by Feron et al. (1979a,b) on Wistar rats, Lee et al. (1978)
on CD rats (Table 32) and Groth et al. (1981, Table 30) on
Sprague-Dawley rats. In inhalation experiments with pregnant
Sprague-Dawley rats, the investigators observed a higher incidence of
ASL and hepatocellular carcinoma in the female progeny than in the
males (Maltoni & Cotti, 1988; Table 32).
After long-term oral administration of VC (Table 29), incidences
of ASL (BT11, Sprague-Dawley rats; Maltoni et al., 1981), neoplastic
liver nodules and hepatocellular carcinoma (Wistar rats; Feron et al.,
1981 (not ASL) and Til et al., 1991) were higher in females than in
males. Interestingly, preneoplastic alterations in the liver, like
increased basophilic foci (Til et al., 1991; Table 29) or
ATPase-deficient foci (Laib et al., 1985a; Table 30), were observed in
female rats at lower doses than in males.
Although statistical analysis of sex differences was not
performed, in rats there is a tendency towards a higher susceptibility
to VC-induced ASL in female animals. Data on species other than rats
are not sufficient for an assessment of sex differences (Table 30,
32).
7.7.5 Carcinogenicity of metabolites
CAA was reported to induce hepatocellular tumours in B6C3F1 mice
when administered orally in drinking-water (Daniel et al., 1992). CEO
caused local tumours after repeated subcutaneous injection and skin
tumours in mice in classical initiation-promotion experiments (CEO
used as an initiator and 12-O- n-tetradecanoylphorbol-13-acetate as a
promoter), whereas CAA did not under comparable conditions (Zajdela et
al., 1980).
7.8 Genotoxicity
Genotoxicity studies on VC in vitro and in vivo in laboratory
animals are given in sections 7.8.1 and 7.8.2, respectively.
Genotoxicity studies on the metabolites of VC are described in section
7.8.3, and the mutagenic/promutagenic properties of DNA adducts formed
by the reactive VC metabolites CEO and CAA are discussed in section
7.8.4. Data on gene mutation and cytogenetic damage in humans exposed
to VC are given in section 8.4. A summary on the genotoxicity of VC
in vitro and in vivo, including human data, is presented in Table
36.
7.8.1 In vitro studies
Relevant studies on the genotoxicity of VC in vitro are
presented in Table 34. Genotoxic activity of VC has been detected in
several in vitro test systems, predominantly after metabolic
activation.
VC is mutagenic in the Ames test in the presence of metabolic
activation in Salmonella typhimurium strains TA100, TA1530 and
TA1535 but not in TA98, TA1537 and TA1538 (Rannug et al., 1974;
Bartsch et al., 1975; McCann et al., 1975; De Meester et al., 1980;
Shimada et al., 1985) indicating that the mutations are the result of
base-pair substitutions (transversion and transition) rather than
frameshift mutations. This is in agreement with the finding that
etheno-DNA adducts formed by the reactive metabolites CEO and CAA (see
section 6.5.1) are converted to actual mutations by base-pair
substitutions (see section 7.11.2 and 8.4). Barbin et al. (1997)
examined p53 mutations in VC-induced rat liver tumours and detected in
12 samples (11 ASL, 1 HCC) only one deletion but 12 base-pair
substitutions (transversion and transition).
In some studies, VC has been shown to also exert mutagenic
activity in S. typhimurium without addition of S9-mix (McCann et
al., 1975; De Meester et al., 1980; Shimada et al., 1985), but the
mutagenic effect was enhanced by addition of a metabolic activation
system (De Meester et al., 1980; Shimada et al., 1985; Victorin &
Ståhlberg, 1988). This increase was more pronounced when liver
extracts were derived from animals pretreated with an enzyme inducer
(Aroclor 1254) (De Meester et al., 1980). The reason for the mutagenic
Table 34. Genotoxicity of vinyl chloride in vitroa
Test type Test organism; Exposure conditions; Results Results Reference
species strain comments with MA without MA
Ames test Bact.; 20% VC in atmosphere; Rannug et al.
Salmonella up to 90 min + - (1974)
typhimurium - -
TA1535 - -
TA1536 - -
TA1537
TA1538
Ames test Bact.; 0.2, 2, 20% VC in atmosphere; n.g. Bartsch et al.
S. typhimurium 1.5-48 h; dose- and + (1975)
TA1530 time-dependent effect +
TA1535 -
TA1538 -
G-46
Ames test Bact.; 20% VC in atmosphere; McCann et al.
S. typhimurium 3, 6, 9 h; - - (1975)
TA98 time-dependent effect + +
TA100 + +
TA1535 - -
TA1538
Ames test Bact.; 1) 2-20% VC in De Meester et
S. typhimurium atmosphere; 16 h; + + al. (1980)
TA1530 dose-dependent effect
Ames test Bact.; 0.1-10% VC in atmosphere; Shimada et al.
S. typhimurium 18 h; - - (1985)
TA98 dose-dependent effect + +
TA100 + +
TA1535 - -
TA1537 - -
TA1538
Table 34. (cont'd)
Test type Test organism; Exposure conditions; Results Results Reference
species strain comments with MA without MA
Ames test Bact.; VC in DMSO added to soft Laumbach et
S. typhimurium agar and bacteria n.g. - al. (1977)
TA100
Ames test Bact.; 83 mM VC in liquid n.g. Bartsch et al.
S. typhimurium suspension; 30 min - (1975)
TA1530 -
TA1535 -
G-46
Gene Bact.; 10.6 mM VC in medium; + - Greim et al.
mutation Escherichia 2 h (1975)
assay coli
K12
Forward Yeast 16, 32, 48 mM VC in + - Loprieno et
mutation cells; medium; 1 h; al. (1977)
assay Schizo- dose-dependent effect
saccharomyces
pombe P1
Forward Yeast 16 or 48 mM VC in medium; + - Loprieno et
mutation cells; 5-240 min; time-dependent al. (1976)
assay S. pombe effect
SP.198
Table 34. (cont'd)
Test type Test organism; Exposure conditions; Results Results Reference
species strain comments with MA without MA
Reverse Yeast cells; incubated with 0.275 n.g. - Shahin (1976)
mutation Saccharomyces or 0.55% VC in DMSO;
assay cerevisiae 4-48 h
XV185-14C
Forward Fungi; VC solution in ethanol - - Drozdowicz &
mutation Neurospora for 3-4 h or 25, 50% Huang (1977)
assay crassa VC in atmosphere for
Ema 5297 3.5 or 24 h
Gene Plant; plant cutings exposed n.g. + Van't Hof &
mutation/ Tradescantia to VC in atmosphere; Schairer (1982)
deletion spec. 6 h; positive at
assay clone 4430 > 195 mg/m3
Cell mammalian 5, 10, 20, 30% VC in + - Drevon &
gene cells; atmosphere; 5 h; Kuroki (1979)
mutation Chinese dose-dependent effect
assay hamster
V79
HGPRT human cells; 25-400 µM VC in + n.g. Weisman (1992)
gene B-lymphoblastoid medium for 24 h;
mutation line cells with
assay metabolizing system
Gene yeast cells; 48 mM VC in incubation + - Loprieno et al.
conversion S. cerevisiae medium; 180-360 min (1976)
assay D4
Gene yeast cells; incubation in 0.275 n.g. - Shahin (1976)
conversion S. cerevisiae or 0.55% VC in DMSO;
assay D5 4-48 h
Table 34. (cont'd)
Test type Test organism; Exposure conditions; Results Results Reference
species strain comments with MA without MA
Gene yeast cells; 2.5% VC in atmosphere + - Eckardt
conversion S. cerevisiae for 1 h et al. (1981)
assay D7RAD
Cell mammalian 10, 20, 30, 40, 50% VC + n.g. Styles (1980)
trans- cells; in atmosphere; no data
formation BHK C1 13 about exposure period
assay
Cell mammalian 18, 180, 1315, 2662 n.g. + Tu et al. (1985)
trans- cells; mg/m3 VC in atmosphere;
formation BALB/c-3T3 24 h; dose-dependent
assay C1 1-13 effect
Rec-assay Bact.; initial 22 mM; 24 h n.g. - Elmore et al.
(DNA repair) B. subtilis (1976)
168M or MC-1
Unscheduled mammalian 5.0, 7.5 or 10% VC in + n.g. Shimada et
DNA cells; rat atmosphere; al. (1985)
synthesis hepatocytes 18 h; dose-dependent
effects
SCE assay human cells; 10, 25, 50, 75, 100% VC + - Anderson et
stimulated in atmosphere; 3 h; al. (1981)
lymphocytes dose-dependent effect
a Bact. = bacteria; DMSO = dimethylsulfoxide; MA = metabolic activation; n.g. = not given; SCE = sister-chromatid
exchange; + = positive; - = negative
activity in the Ames test in the absence of S9-mix has been suggested
to be a result of non-enzymic breakdown of VC or an internal bacterial
metabolism (Bartsch et al., 1976; Shimada et al., 1985), but the
origin of the direct mutagenic effect remains unclear.
Positive results in the Ames test were also observed with
metabolic activation by extracts prepared from human liver biopsies
(Bartsch et al., 1975). Mutagenicity of VC was dependent on
concentration (Bartsch et al., 1975; De Meester et al., 1980; Shimada
et al., 1985) and exposure duration (Bartsch et al., 1975; McCann et
al., 1975). VC was mutagenic when plates were exposed to a VC
atmosphere in a closed (Bartsch et al., 1975 and Table 34) or in a
dynamic flow-through system (Victorin & Ståhlberg, 1988). No mutagenic
effect was observed when VC was dissolved in aqueous solution with
(Bartsch et al., 1975) or without (Rannug et al., 1974; Laumbach et
al., 1977) metabolic activation, probably due to rapid loss of VC in
aqueous solution by evaporation (Bartsch et al., 1975).
Other gene mutation assays in bacteria (Greim et al., 1975),
yeast cells (Loprieno et al., 1976, 1977) and mammalian cells (Drevon
& Kuroki, 1979) revealed positive results exclusively in the presence
of metabolic activation. Mutagenic effects were also reported in a
human cell line containing cloned cytochrome P450IIE1, which is
capable of metabolizing VC (Weisman, 1992). Gene mutation was also
detected in plant cuttings (Tradescantia) exposed to VC (Van't Hof &
Schairer, 1982). No mutagenicity was observed in Neurospora crassa
with or without addition of exogenic activation system (Drozdowicz &
Huang, 1977) but the validity of this study is limited by
contradictory documentation.
In gene conversion assays, positive results were reported with
Saccharomyces cerevisiae in the presence of a metabolic activation
system (Loprieno et al., 1976; Eckardt et al., 1981). No mutagenic
effects were observed without metabolic activation (Shahin, 1976).
VC exposure induced unscheduled DNA synthesis in rat hepatocytes
(Shimada et al., 1985) and increased sister-chromatid exchange in
human lymphocytes after addition of an exogenic activation system
(Anderson et al., 1981). No growth inhibition was detected in DNA
repair-deficient bacteria without metabolic activation (Elmore et al.,
1976).
Cell transformation assays revealed positive results with
(Styles, 1980) or without (Tu et al., 1985) metabolic activation.
7.8.2 In vivo studies
Key studies on the genotoxicity of VC in vivo are documented in
Table 35. VC exposure induced gene mutation and mitotic recombination
in Drosophila melanogaster but not gene mutation in mammalian germ
cells. VC showed in vivo clastogenic effects, increased sister
chromatid exchanges and induced DNA breaks. VC induced also gene
conversion and forward mutations in host-mediated assays.
Mutagenic activity of VC was reported in the mitotic
recombination assay (Vogel & Nivard, 1993) and the sex-linked
recessive lethal (SLRL) assay (Magnusson & Ramel, 1976; Verburgt &
Vogel, 1977; Magnusson & Ramel, 1978) on D. melanogaster. The lowest
effective concentration in the SLRL assay was 2210 mg/m3 with a 2-day
exposure period and 78 mg/m3 after 17 days of exposure (Verburgt &
Vogel, 1977). In the same study, negative results were observed at
higher exposure concentrations with a 2-day exposure period in assays
on D. melanogaster for dominant lethals, translocations (not
tabulated) and sex chromosome loss. These results were discussed by
the authors as a consequence of a saturation effect observed in the
SLRL test (Verburgt & Vogel, 1977). However, sex chromosome loss was
observed in studies by Ballering et al. (1996) in D. melanogaster
exposed to higher concentrations.
No mutagenic activity was detected in the dominant lethal assay
with mice (Anderson et al., 1976; Himeno et al., 1983) and rats (Short
et al., 1977). No mutagenicity was reported in the mouse spot test
(Peter & Ungvįry, 1980). Chromosomal aberrations in rats (Anderson &
Richardson, 1981) and hamsters (Basler & Röhrborn, 1980) were reported
and mouse bone marrow micronucleus tests (Jenssen & Ramel, 1980;
Richardson et al., 1983) gave positive results.
VC exposure (260, 650 and 1300 mg/m3) induced single-strand
breaks dose-dependently in the liver DNA of NMRI mice (Walles &
Holmberg, 1984; Walles et al., 1988). Increased frequencies of
sister-chromatid exchange and chromosome aberrations were observed in
the bone marrow of Chinese hamsters after exposure to 1.25, 2.5 or 5%
(v/v) for 24 h (Basler & Röhrborn, 1980).
7.8.3 Genotoxicity of VC metabolites
Metabolic activation of VC is necessary to form the genotoxic
metabolites. The metabolites themselves are genotoxic in the absence
of metabolic activation (see also section 6.3). The VC metabolites,
chloroethylene oxide (chloro-oxirane), chloroacetaldehyde, and
chloroacetic acid were investigated for genotoxicity. In vitro
studies discussed in this section were performed without metabolic
activation unless otherwise stated. Information on the mechanism of
mutagenesis is presented in section 6.5 and 7.11.2.
Chloroethylene oxide (CEO) was found to be the most effective VC
metabolite regarding forward mutation and gene conversion in yeast
(Loprieno et al., 1977), gene mutation in mammalian cells (Huberman et
al., 1975) and reverse mutation in bacteria (Malaveille et al., 1975;
Rannug et al., 1976; Hussain & Osterman-Golkar, 1976). The mutational
specificity of CEO was investigated in Escherichia coli, using trpA
mutant strains. In this system, CEO induced all types of base-pair
substitutions (except one, which was not tested) (Barbin et al.,
1985b). GC -> AT transitions were the most frequent, followed by
AT -> TA transversions. This metabolite inhibited growth in DNA
repair-deficient bacteria (Elmore et al., 1976; Laumbach et al.,
1977).
Table 35. Genotoxicity of vinyl chloride in vivoa
Species/Strain/Sex Test type Test conditions; comments Results References
Gene mutation
Mouse/CD-1/m dominant lethal 20 mice/group exposed to 0, 7800, - Anderson et
assay 26 000 or 78 000 mg/m3 for 6 h/d for al. (1976, 1977)
5 d before 8 wk mating; survival 100,
90, 95 and 45%
Rat/CD/m dominant lethal 12 mice/group exposed to 0, 130, 650, - Short et al.
assay or 2600 mg/m3 for 6 h/d, 5 d/wk; one (1977)
mating during week 11 of exposure;
no mortality
Mouse/CD-1/m dominant lethal a) 13 m exposed to 26 000 mg/m3 for - Himeno et al.
assay 4 h/d for 5 d (11 controls) before 7 wk (1983)
mating; b) 20 m exposed 4 h/d, 5 d/wk
to 13 000 mg/m3 for 10 wk before 3 wk
mating
Mouse/C57BL/f mouse spot test 44 pregnant mice exposed to 12 000 mg/m3 - Peter &
for 5 h on gestation day 10 (51 controls) Ungvįry (1980)
Mouse/Swiss/n.d. host-mediated 4-6 mice exposed orally to 700 mg/kg bw. + Loprieno et al.
forward mutation in olive oil; yeast cells (S. pombe (1976)
assay SP.198) inoculated in peritoneum for
3, 6 or 12 h
D. melanogaster/ dominant lethal m exposed to 0 or 78 000 mg/m3 for 2 d; - Verburgt &
Berlin K/m assay total number of eggs per group at Vogel (1977)
least 6950; increase not significant; no
differences in hatchability (> 80%)
D. melanogaster/ Drosophila a) m exposed to 0, 1, 10, 20% VC in air + Magnusson &
Karsnäs/m SLRL test for 3 h (at least 491 chromosomes tested); Ramel (1978)
b) m exposed to 0, 1, or 10% VC for 3 h
after pretreatment with 1% phenobarbiturate
solution for 24 h;
a) positive at > 1% VC; no dose dependency;
pretreatment in b) increased mutagenicity
D. melanogaster/ Drosophila 50 m/group exposed continuously a) to 0, + Verburgt &
Berlin K/m SLRL test 78, 520, 2210, 26 000, 78 000, 130 000 mg/m3 Vogel (1977)
for 2 d or b) to 0, 78, 2210 mg/m3 for 17 d;
a) positive at > 2210 mg/m3, no clear dose
response;
b) positive at > 78 mg/m3
Mitotic
recombination
D. melanogaster/ mitotic eye mosaic assay; 48- to 72-h-old larvae + Vogel &
LS/f + m recombination exposed to 5200 mg/m3 for 17 h; light spots Nivard (1993)
assay of at least 500 eyes scored in adult f
(control 250 eyes scored); survival not
reduced
Chromosomal
abnormalities
Rat/Wistar/m cytogenetic 24 m/group exposed to 3900 mg/m3 for + Anderson &
assay a) 5 d (6 h/d) or b) 3 mo (6 h/d, 5 d/wk); Richardson
bone marrow sampled 24 h after exposure (1981)
period; increased number of cells with any
abnormality, significant in a)
Hamster/Chinese/ cytogenetic 2 m + 2 f per group exposed to 2.5% for 6, + Basler &
f+ m assay 12, or 24 h; 5 m + 5 f exposed to 5% VC Röhrborn
for 24 h; bone marrow samples prepared 26 h (1980)
after start of exposure; control 7 m + 7 f;
increased aberrations at > 6 h (gaps
excluded), effect dose related
Table 35. (cont'd)
Species/Strain/Sex Test type Test conditions; comments Results References
Mouse/CBA/m micronucleus 3 m exposed to 0 or 5% VC for 4 h; bone + Jenssen &
assay marrow examined 30 h after exposure Ramel (1980)
Mouse micronucleus 0, 260, 860 or 2600 mg/m3 for 2 × 4 h + Rodics et al.
CFLP assay (1981)
Mouse/ micronucleus 10 f and 10 m exposed to 0 or 130 000 mg/m3 + Richardson
C57BI/6J/f + m assay for 6 h; bone marrow examined 24 or 48 h et al. (1983)
after exposure
D. melanogaster/ sex m exposed to 0 or 78 000 mg/m3 for 2 d and - Verburgt &
Berlin K/m chromosome chromosomes analysed in progeny (at least Vogel (1977)
loss 6725 m + f per group)
D. melanogaster/ sex m exposed to 0 or 126 000 mg/m3 for 48 h; + Ballering et
ring-X/m chromosome chromosome loss determined in F1 (at al. (1996)
loss least n = 428; 3 broods)
Other effects
Rat/Wistar/m host mediated 20-30 rats exposed to 0 or 1% VC for 24 + Eckardt et al.
gene conversion h, starting 1 h after yeast cell (1981)
assay (S. cerevisiae D7RAD) injection (i.v.)
Mouse/NMRI/f alkaline elution 3-5 f per group; exposure to 1300 mg/m3 + Walles &
assay in liver for 39, 60, 117, 234 h (6 h/d, 5 d/ wk) Holmberg
DNA and sacrificed a) 2 h or b) 18 h (exposed (1984)
for 36, 114, 231 h) after exposure period;
concurrent control sacrificed after 36 or
231 h; positive in a) at 39 h, in b) at 114 h
Table 35. (cont'd)
Species/Strain/Sex Test type Test conditions; comments Results References
Hamster/Chinese SCE assay 2 m + 2 f per group exposed to 1.25 or 2.5% + Basler &
f + m VC for 6, 12, or 24 h; bone marrow samples Röhrborn
prepared 26 h after start of exposure; (1980)
control 4 m + 4 f; dose- and time-dependent
effect
a d = day; f = females; m = males; mo = months; n.d. = no data; SCE = sister-chromatid exchange;
SLRL = sex-linked recessive lethals; wk = weeks; + = positive; - = negative
CAA was 450 times less mutagenic than CEO in the Ames test but
more active than the concurrent positive control ethylene oxide
(Rannug et al., 1976). Positive results were reported with CAA in gene
mutation assays in bacteria (Malaveille et al., 1975; Bartsch et al.,
1975; McCann et al., 1975; Hussain & Osterman-Golkar, 1976; Elmore et
al., 1976; Laumbach et al., 1977; Perrard, 1985), yeast cells
(Loprieno et al., 1977), mammalian cells (Huberman et al., 1975), in a
human lymphoblast cell line (Sanchez & Recio, 1991) and in human cells
using shuttle vectors (Matsuda et al., 1995).
With chloroacetic acid no enhancement of the mutation frequency
could be detected in bacteria (Bartsch et al., 1975; Malaveille et
al., 1975; Rannug et al., 1976) or mammalian cells (Huberman et al.,
1975).
Evidence for mutagenic activity of photoreaction products formed
from VC was presented by Victorin & Ståhlberg (1991). Mixtures of VC
(up to 260 mg/m3) and nitrogen dioxide (but not VC alone) were
mutagenic in S. typhimurium TA100 after 40 min UV irradiation of the
gas mixture before exposure of bacteria.
CEO but not CAA showed a similar toxicity/mutagenic profile to VC
in the hprt locus in a metabolically competent human
B-lymphoblastoid cell line (Chiang et al., 1997; see also section
8.4.2).
7.8.4 Other toxic effects of VC metabolites
In the above studies on the genotoxic effects of CAA (see section
7.8.3), this compound appeared to be highly cytotoxic in various
cellular systems. CAA has also a high acute toxicity in animals, with
a LD50 value of 0.15 mmol/kg body weight. Kandala et al. (1990)
showed in vitro the concentration-dependant reversible inhibition of
DNA synthesis by CAA in rat and mouse cells without a reduction in
thymidine uptake or formation of nucleotides. In isolated rat
hepatocytes, CAA stimulates lipid peroxidation (Sood & O'Brien, 1993).
7.8.5 Mutagenic and promutagenic properties of DNA adducts formed by
VC metabolites
The major DNA adduct of VC, 7-(2'-oxoethyl)guanine (7-OEG), lacks
miscoding properties (Barbin et al., 1985a). In contrast,
1, N6-ethenoadenine (Epsilon A), 3, N4-ethenocytosine (Epsilon C)
and N2,3-ethenoguanine (Epsilon G) showed miscoding properties
(Singer, 1996; see Table 37). The promutagenic properties of the
etheno adducts involve mainly base-pair substitution mutations
(Grollman & Shibutani, 1994). Site-specific mutagenesis studies in E.
coli and in mammalian cell lines have shown that both Epsilon G and
Epsilon C can induce G:C -> A:T transitions; Epsilon C can also lead
to C:G -> A:T transversions (Cheng et al., 1991; Moriya et al.,
1994). Epsilon A can induce misincorporation of G, C, or A during
replication, thus inducing the base-pair substitutions A:T -> C:G,
A:T -> G:C or A:T -> TA (Basu et al., 1993; Pandya & Moriya, 1996).
7.8.6 Mutations in VC-induced tumours
Barbin et al. (1997) examined the presence of p53 gene
mutations (the function of the p53 gene is described in
section 8.4.2) in ASL and HCC tumours induced by VC in Sprague-Dawley
rats. Mutations were found in 11/25 ASL and 1/8 HCC. A twelve-base
deletion was found in one tumour; all others were base-pair
substitutions. Nine of the point mutations were observed at A:T base
pairs and of three G:C -> A:T transitions (Table 43).
Mutations of the p53 gene were also found in tumours from vinyl
chloride-exposed autoclave workers with liver angiosarcoma (ASL) and
hepatocellular carcinoma (HCC) (Hollstein et al., 1994; Boivin et al.,
1997; see also section 8.4 and Table 43). To date (1998) 11 out of 15
(73%) ASL from VC-exposed workers have been shown
immunohistochemically to have mutant p53 protein. Furthermore, a
statistically significant trend for mutant p53 protein has been found
in the serum of VC-exposed workers (Smith et al., 1998). In contrary
to studies in humans, no mutations were found in codons 12, 13 and 61
of the Ki- ras gene in rat liver tumours induced by VC, but mutations
were found involving codon 61 of the Ha- ras proto-oncogene (Froment
et al., 1994; Boivin-Angčle et al., in press) (see Table 44 and
section 8.4.2).
Connexin genes have been shown to restore normal cell growth when
transfected into certain tumorigenic cells and thus are considered to
form a family of tumour suppressor genes. Mutations of the connexin
37 (Cx37) gene in rat liver tumours (22 hepaticangiosarcomas and 3
hepatocellular carcinomas) induced by VC were analysed by
PCR-single-strand conformation polymorphism analysis and DNA
sequencing. The results suggested that Cx37-mediated gap junctional
intercellular communication may be disturbed in most of these
angiosarcomas but mutation of the Cx37 gene is rare (Saito et al.,
1997).
7.9 Factors modifying toxicity
Concurrent administration of ethanol (5% in drinking-water until
termination at month 30) and VC (1560 mg/m3, inhalation 4 h/day,
5 days/week for 1 year) to male Sprague-Dawley rats (80 rats per group
for histopathological evaluation) resulted in increased incidences of
ASL from 23% after exposure to VC alone to 50% in rats exposed to VC
and ethanol versus 0% in ethanol-treated rats and 0% in concurrent
controls. Ethanol had an additive effect on the incidence of
hepatocellular carcinoma (VC 43%, VC-ethanol 60%, ethanol 10%, control
1.3%) and lymphosarcoma (VC 7.5%, VC-ethanol 14%, ethanol 5%, control
2.5%) (Radike et al., 1981). This effect may be due to the interaction
of ethanol with VC metabolism.
Induction of certain enzymes of the mixed-function oxidase system
by pretreatment with phenobarbital (Jaeger et al., 1974, 1977;
Reynolds et al., 1975a,b,c) or the mixture of polychlorinated
Table 36. Summary of genotoxic effects induced by exposure to vinyl chloride in vitro and in vivoa
In vitro In vivo
Bacteria Yeast Plants Mammalian Human Insects Mammalia Humansc
GM DD GM GC GM GM DD CT GM SCE GM AN MR GM CA MN DD SCE GM CA MN SCE
+ -b + + + + + + + + + - + - + + + + + + + +
a AN = aneuploidy; CA = chromosomal aberration; CT = cell transformation; GM = gene mutation; GC = gene conversion;
DD = DNA damage; MN = micronuclei; MR = mitotic recombination; SCE = sister-chromatid exchange
b Single study, only tested without metabolic activation
c These results are described in Table 42 (section 8.4.1)
Table 37. Evidence for base-pair substitutions caused by etheno-DNA adducts
Ethenobase Base Base-pair System used Reference
incorporated substitution for study
opposite
adducta
1,N6-ethenoadenine C AT -> GC in vivo bacteriophage M13-Nhei in E. coli Basu et al.
(epsilon-A) transition (1993)
A AT -> TA in vivo single-strand vector shuttle in Pandya & Moriya
C AT -> CG E. coli or simian kidney (COS) cells; (1996)
transversions
3,N4-ethenocytosine A CG -> TA in vivo single-strand vector shuttle in Moriya et al.
(epsilon-C) transition E. coli or simian kidney (1994)
T CG -> AT in vitro (COS) cells; Zhang et al.
transversion E. coli DNA polymerase (1995)
I system
in vivo M13AB28 in SOS-(UV)-induced Jacobsen &
E. coli Humayun (1990)
in vivo M13 glyU phage transfection Borys et al.
of E. coli tester strain (1994)
in vitro E. coli DNA polymerase I system Simha et al.
(1991);
Palejwala et al.
(1991)
N2,3-ethenoguanine T GC -> AT in vivo E. coli DNA synthesis by M13G*1 Cheng et al. (1991)
(epsilon-G) transition assay
in vitro E. coli DNA polymerase I (Klenow Singer et al. (1991)
fragment); exonuclease-free Klenow;
Drosophila melanogaster polymerase
alpha-primer compex;
human immunodeficient virus-I
reverse transcriptase (HIV-RT)
in vitro E. coli DNA-dependent RNA Mroczkowska &
polymerase Kusmierek (1991)
Table 37. (cont'd)
Ethenobase Base Base-pair System used Reference
incorporated substitution for study
opposite
adducta
1,N2-ethenoguanine A GC -> TA in vitro E. coli DNA polymerase I Langouet et al.
G GC -> CG (exonuclease-free Klenow) (1997)
transversions
a A = adenine; C = cytosine; G = guanine; T = thymine
biphenyls (Arochlor 1254) (Reynolds et al., 1975a,b; Conolly et al.,
1978) enhanced acute hepatotoxicity in rats as measured by increased
activity of hepatic enzymes and/or focal hepatic necrosis.
Administration of SKF-525A, an inhibitor of the mixed-function oxidase
system, 30 min prior to VC exposure in phenobarbital-pretreated rats,
inhibited the enhancing effect of phenobarbital on VC hepatotoxicity
(Jaeger et al., 1977). Application of cysteine, a rate-limiting
precursor in hepatic glutathione (detoxification of reactive
chemicals), via drinking-water prior to VC exposure protected Arochlor
1254-pretreated rats against acute VC hepatotoxicity (Conolly &
Jaeger, 1979).
7.10 Mechanisms of toxicity - mode of action
7.10.1 Mechanisms of VC disease
Based on evidence of immunological abnormalities, such as
hyperimmunoglobinaemia and circulating immune complexes in workers
with "vinyl chloride disease" (section 8.3), Ward et al. (1976) have
proposed a possible mechanism for the vascular changes associated with
this disease. Reactive VC metabolite(s) bind(s) to a protein,
resulting in a structurally abnormal protein. This protein would react
as an antigen and initiate an immune response with B-cell
proliferation and hyperimmunoglobinaemia. Circulating immune complexes
formed by interaction of this antigen and antibodies would precipitate
in the extremities of exposed humans in response to the cold and
activate the complement sequence. The cryoprecipitates and reactions
secondary to the complement activation are proposed to produce
vascular occlusion and fibrinogen/fibrin conversion. The mechanism is
supported by findings of IgG deposition with associated complement C3
and fibrin in histological lesions (Grainger et al., 1980). The
resulting vascular insufficiency would explain the observed clinical,
radiological and histological findings in skin, skeletal and soft
tissues, and lungs (section 8.3) in workers occupationally exposed to
high concentrations of VC (Ward et al., 1976). No further studies were
identified that would directly confirm this mechanism, and the degree
to which it has been accepted is not clear. The available evidence
does not seem sufficient to suggest an autoimmune disease as a
pathogenetic mechanism. Further studies will be needed to establish
the true significance of the possibly transient immunological
abnormalities in VC-induced disorders.
7.10.2 Mechanism of carcinogenesis
There is a large body of data showing that VC acts as a genotoxic
carcinogen. After metabolic activation to CEO by CYP2E1, VC exerts
various genotoxic effects (including gene mutations and chromosomal
aberrations) in different organisms, including bacteria, yeasts,
mammalian cells in culture, Drosophila, rodents and humans (Table 36).
Among the mutagenic events induced by VC, base-pair substitutions
appear, so far, to be the most frequent. VC in the presence of an
activation system has a transforming activity on mammalian (rodent)
cells in culture (see Table 34).
Studies in vitro have demonstrated that metabolically activated
VC and its electrophilic metabolites CEO and CAA can alkylate nucleic
acid bases. 7-OEG, the major DNA adduct formed by VC and CEO does not
exhibit promutagenic properties. In contrast, four minor adducts,
Epsilon A, Epsilon C, N2,3-Epsilon G and 1,N2-Epsilon G, show
promutagenic properties, inducing mainly base-pair substitution
mutations and a low level of frameshift mutations.
7-OEG and three etheno adducts (Epsilon A, Epsilon C,
N2,3-Epsilon G) have been detected in DNA from rats and mice exposed
to VC. Highly variable background levels of Epsilon A and Epsilon C
were found in all the tissues examined. Following exposure of rats to
VC, significantly elevated levels of Epsilon A and Epsilon C were
measured in most tissues, except the brain; there was also no
significant increase of Epsilon A levels in the kidney and spleen.
The liver is one of the primary targets for VC-induced
carcinogenesis in rats and humans. It is also, by far, the major
tissue involved in VC activation in rats. Following exposure of rats
to VC, the distribution of etheno adduct levels (induced by the
exposure) is rather homogeneous within the organism. Epsilon C was
shown to accumulate as a function of length of exposure in at least
three organs (liver, kidney and lung). In contrast, Epsilon A
accumulated in the liver but not in the kidney and lung. In addition,
adduct levels (Epsilon A, Epsilon C) did not decrease in the liver,
for at least two months following the end of exposure. Etheno adducts
are formed as endogenous background levels in various tissues in
humans; no data are available on etheno adduct levels in humans
exposed to VC.
Mutations have been found in liver tumours associated with
exposure to VC. In human ASL, the Ki-ras gene is activated through a
GC -> AT mutation at base 2 of codon 13. Mutations, all AT -> TA
transversions, have been described in the p53 gene in three human
ASL. The Ki-ras gene activation is not found in rat ASL. However, 44%
of rat ASL were found to contain a mutated p53 gene: most mutations
were base-pair substitutions, involving mainly A:T base pairs. The
data suggest the existence of hot spots for mutations in the p53
gene, and one mutation found in two rat ASL was equivalent to the
same mutation characterized in one human ASL associated with VC
exposure. The Ha-ras gene is activated in rat HCC induced by VC,
through an AT -> TA transversion in codon 61.
The mutation spectra observed in liver tumours (ASL and/or HCC)
associated with VC exposure in humans and rats are clearly distinct
from those observed in sporadic liver tumours or in hepatic tumours
associated with other exposures. In rats, the substitution mutations
found at A:T base pairs in the ras and p53 genes are consistent
with the promutagenic properties of Epsilon A and with the
accumulation and persistence of this lesion in hepatic DNA.
Altogether, available data suggest that etheno adducts could be
involved in the initiation of hepatocarcinogenesis by VC. However,
they cannot explain the observed tissue- and cell-specificity.
More studies on the formation and repair of etheno adducts at the
cellular level (cell specificity), as well as at the molecular (gene
and DNA sequence) level, are warranted. Carcinogenesis is a multi-step
process and, obviously, there is a need for quantitative evaluation of
other critical biological end-points, such as effects of VC on
apoptosis, cell proliferation or intercellular communication in
vivo.
8. EFFECTS ON HUMANS
Only reports on effects for humans of exposure to VC have been
considered here and not reports where exposure has been to a number of
chemicals, e.g., at landfills, in factories manufacturing a number of
chemicals or in the PVC processing industry.
8.1 General population exposure
After an accident in Schönebeck, Germany in June 1996, involving
the derailment of a train carrying VC and subsequent fire, 325 persons
were documented as having acute symptoms but these correlated with
exposure to the pyrolytic products (e.g., HCl) and not to VC itself.
But a study on 29 persons exposed as a result of this accident showed
a significant increase in chromosomal aberrations compared to an
unexposed control group (Hüttner & Nikolova, 1998; see also Table 42
and section 3.2.3).
A case has been reported of epithelioid haemangioendothelioma
involving liver, bone and lungs in a man living for over 8 years
several hundred metres from a toxic waste dump next to a chemical
plant producing VC (Shin et al., 1991).
There have been several reports on the possible increased
prevalence of congenital malformations in populations exposed to
emissions from polymerization facilities (Edmonds et al., 1975, 1978;
Infante, 1976; Thériault et al., 1983; Rosenman et al., 1989) but none
of these studies showed a statistically significant correlation
between developmental toxicity and proximity to the facility
(Clemmesen, 1982; Hemminki & Vineis, 1985). A number of studies (e.g.,
Goldberg et al., 1995; Dolk et al., 1998) examined risk for cancer and
for adverse reproductive outcome in relation to proximity to
landfills. Although VC is one of the potential emissions from the
landfills, these studies do not directly address the population
exposure to VC and were not further considered.
In England and Wales, from 1975-1987 data, there were no
confirmed non-occupationally exposed cases of ASL (Elliott &
Kleinschmidt, 1997). In the USA, five non-occupational cases were
reported living within 1.6 km of a VC plant (Brady et al., 1977).
8.2 Controlled human studies
Three men and three women were exposed twice daily with a 6-h
interval for three successive days to 0, 0.4, 0.8, 1.2, 1.6 or 2.0%
VC. The NOEL was between 0.8 and 1.2%. Above this, dizziness, nausea,
dulling of vision and auditory cues were reported (Lester et al.,
1963).
Thirteen male volunteers were exposed to 130, 650 and 1300 mg/m3
for 7.5 h and subjective and neurological responses were measured
before the subject entered the chamber, 15 min after entrance and at
1-h intervals thereafter; 24-h post-exposure urine and blood samples
were taken and tested. No significant adverse effects were noted with
the exception of some dryness of eyes and nose at 1300 mg/m3. The
exposure had no noticeable effect on neurological responses nor did it
produce significant changes in the results of mental, coordination or
manual dexterity tests conducted during the exposure period. All
clinical laboratory studies performed in the post-exposure period were
normal and not significantly different from pre-exposure values
(Baretta et al., 1969).
8.3 Occupational exposure
8.3.1 Overview
VC was first produced commercially in the late 1920s. Various
effects caused by exposure to VC were reported: cardiac arrhythmia in
experimental animals (Oster et al., 1947); hepatic abnormalities in VC
workers (Tribukh et al., 1949; Filatova et al., 1958), acroosteolysis,
Raynaud-type-phenomenon and sclerodermoid skin lesions (Lelbach &
Marsteller, 1981). But it was not until it was found that VC could
cause cancer in animals (Viola et al., 1971; Maltoni et al., 1974) and
humans (Creech & Johnson, 1974) that levels of VC in the workplace
were drastically reduced. Some VC workers, in particular autoclave
cleaners, were estimated to have been exposed to as much as
2600 mg/m3 (1000 ppm) in the 1950s or earlier, reducing to a tenth of
this level by the mid-1970s (Table 20). After 1975 levels were usually
2.6-13 mg/m3 (1-5 ppm) in many countries, but in some countries where
production plants were not modernized, workers were or are exposed to
high levels of VC (Fucic et al., 1990a; Gįlikovį et al., 1994; Hozo et
al., 1996, 1997; see also Table 21).
It should be noted that the non-neoplastic and neoplastic effects
described in the following sections are due in most cases to the high
exposure of workers to VC before 1974. No cases of ASL (section 8.3.1)
have been reported to the International Register of Cases among (West)
European workers first exposed after 1972 (Storm & Rozman, 1997). This
is not the case for Eastern European countries where the old
regulation of 194 mg/m3 (75 ppm) in the working environment was valid
until recently (Hozo et al., 1997).
8.3.2 Non-neoplastic effects
8.3.2.1 Acute toxicity
In acute VC intoxication, the symptoms described include vertigo,
nausea and headache. At higher concentrations, VC exerts narcotic
effects and at one time was considered as a possible anaesthetic
(Patty et al., 1930; Peoples & Leake, 1933; Oster et al., 1947).
VC caused almost immediate death of VC polymerization workers in
two incidents of accidental poisoning. No values were given but the
strong smell and the narcotic effects of VC were reported (Danziger,
1960).
Concentrations of VC of the order of 26 000 mg/m3 (1%) in the
air induce unconsciousness and cardiac arrhythmia. Exposure of two
workers to 2.5% VC for 3 min caused dizziness, disorientation and a
burning sensation in the soles of the feet. There was complete
recovery except for a slight headache lasting 30 min (Danziger, 1960).
A man whose hands were accidentally sprayed with VC developed
erythema and some second-degree burns which healed without
complication (Harris, 1953). A patient complaining of eye burns from
VC recovered 48 h after the eye was rinsed for 15 min with saline
(McLaughlin, 1946).
8.3.2.2 Effects of short- and long-term exposure
Concentrations of VC in the region of 2590 mg/m3 (1000 ppm),
which were not unusual prior to 1974, over periods ranging from
1 month to several years, have been reported to cause a specific
pathological syndrome found in VC workers called the "vinyl chloride
illness". Symptoms described were earache and headache, dizziness,
unclear vision, fatigue and lack of appetite, nausea, sleeplessness,
breathlessness, stomachache, pain in the liver/spleen area, pain and
tingling sensation in the arms/legs, cold sensation at the
extremities, loss of libido and weight loss (Thiess & Versen, 1974).
Clinical findings included scleroderma-like changes in the fingers
with subsequent bony changes in the tips of the fingers described as
acroosteolysis, peripheral circulatory changes similar to Raynaud's
disease, and enlargement of the liver and spleen with a specific
histological appearance, and respiratory manifestations (Lange et al.,
1974; Suciu et al., 1975; Veltmann et al., 1975; Lelbach & Marsteller,
1981).
8.3.2.3 Organ effects
a) Skin and skeletal tissues
Occupational acroosteolysis has largely affected the most highly
exposed workers involved in scraping the insides of autoclaves in the
PVC production process (Harris & Adams, 1967; section 3). It is a rare
bone disease resulting in de-calcification of the terminal phalanges
of the hands and other extremities (Cordier et al., 1966; Wilson et
al., 1967; Markowitz et al., 1972). In these cases, acroosteolysis was
often preceded by soreness and tenderness, numbness, pallor and
cyanosis of the extremities especially the hands, a Raynaud-type
phenomenon caused by reversible constriction of the arterioles.
Sclerodermoid-like changes appear in the skin of the hands and
forearms and osteolytic and sclerotic lesions of the bones,
particularly the extremities and sacroiliac joints (Lange et al.,
1974). Most of the abnormalities disappeared and bones showed signs of
healing a year or two after the men had stopped work (Harris & Adams,
1967). A chronothermodynamic study of Raynaud's phenomenon secondary
to past exposure to VC showed that, although reduced, the symptoms
were still present after 8 years (Fontana et al., 1996).
A review of five studies from four countries involving 725
workers at risk from VC exposure showed that 3% developed
acroosteolysis; 10% Raynaud-type phenomenon and 6% sclerodermoid skin
lesions (Lelbach & Marsteller, 1981). Genetic susceptibility has been
suggested as a possible reason that not all workers who had been in
contact with VC developed symptoms (Black et al., 1983, 1986). The
scleroderma-like syndrome induced by VC exposure appears to be
different from that of other (systemic) scleroderma (Ostlere et al.,
1992). It has a shorter incubation period (1 month to 3 years compared
to 4 to 44 years) (Ishikawa et al., 1995) and there are immunological
differences (section 8.3.2.3f).
b) Hepatic effects
Exposure to VC is associated with hepatomegaly and/or
splenomegaly and with various histological lesions in the liver (Lange
et al., 1974; Ho et al., 1991). In one carefully described study,
advanced portal hypertension with histological findings of
non-cirrhotic fibrosis was diagnosed in 17 of 180 VC polymerization
workers (Lelbach & Marsteller, 1981).
Focal hepatocellular hyperplasia and focal mixed hyperplasia
(hyperplasia of sinusoidal cells along with hyperplasia of
hepatocytes) are early histological alterations indicative of VC
exposure (Tamburro et al., 1984). The precursor stage of ASL is
characterized by subcapsular fibrosis, progressive portal fibrosis and
a borderline increase of intralobular connective tissue, all
associated with focal stimulation and proliferation of sinusoidal
lining cells and hepatocytes. Transition to angiosarcoma is preceded
by focal dilatation of sinusoids with enlarged dedifferentiated lining
cells often containing peg-like hyperchromatic polymorphic nuclei
(Popper & Thomas, 1975; Gedigk et al., 1975).
There is a great similarity between the histological sequences in
the liver of rodents exposed by inhalation to VC (chapter 7; Spit et
al., 1981) and the lesions observed in VC-exposed workers (Popper et
al., 1981).
Twenty of the 39 "liver disease deaths" reported in an Italian
study of highly exposed PVC workers were due to liver cirrhosis, the
remainder being due to miscellaneous liver disorders. A statistical
evaluation was not possible (Pirastu et al., 1990). In another study,
two of 21 heavily exposed VC workers died from sequelae to
noncirrhotic portal fibrosis and portal hypertension (Lelbach, 1996).
In the USA and European cohort studies of VC-exposed workers (study
descriptions are given in section 8.3.3), deaths from chronic
non-malignant diseases of the liver had a low SMRs of 62 and 88; in
the European study this was significantly less than expected (Wong et
al., 1991; Simonato et al., 1991).
c) Cardiovascular effects
Some old studies reported statistically non-significant increased
mortality due to cardiovascular diseases among workers exposed to VC
(Ebihara, 1982; Greiser et al., 1982).
Laplanche et al. (1992) report an elevated incidence of
circulatory system diseases other than Raynaud's disease (RR 1.4, 95%
CI 1.0-1.8) in a 7-year follow-up of a cohort of 1100 VC workers. The
increased risk was mainly due to hypertension and "other circulatory
disorders". Likewise, another study with a 5-year follow-up on the
incidence of arterial hypertension (AH) and coronary heart disease
(CHD) in 105 VC and PVC workers exposed to VC at between 4 and
1036 mg/m3 showed that exposed workers had significantly higher blood
pressure than controls. The estimated relative risk (RR) for AH in
exposed workers was twice as high as in the controls, while there was
no significant difference regarding CHD. There was an
exposure-response relationship between the intensity of exposure and
the incidence of AH (Kotseva, 1996).
In contrast, both the large cohort studies reported a
statistically significant deficit in the mortality from cardiovascular
diseases, with SMRs of 87 and 81. In the USA study, the mortality was
even lower for the subcategory of arteriosclerotic heart disease (SMR
74.5, 95% CI 81.6-89.2 (Simonato et al., 1991; Wong et al., 1991). In
the Canadian study (Thériault & Allard, 1981), there was a
statistically non-significant 20% deficit in cardiovascular mortality,
based on 25 exposed cases.
d) Respiratory effects
Adverse respiratory effects reported in older case studies
included increased incidence of emphysema (Suciu et al., 1975),
decreased respiratory volume and vital capacity, respiratory
insufficiency (Suciu et al., 1975), decreased respiratory oxygen and
carbon dioxide transfer (Lloyd et al., 1984), pulmonary fibrosis of
the linear type (Suciu et al., 1975), abnormal chest X-rays (Lilis et
al., 1975) and dyspnoea (Walker, 1976). This is probably due in part
to confounding by smoking and presence of PVC-resin dust, which is
known to cause respiratory lesions (Mastrangelo et al., 1979; Lilis,
1981).
Both large cohorts (Wong et al., 1991; Simonato et al., 1991)
found a deficit in the mortality from non-malignant respiratory
diseases (SMR 81.6, and 77, respectively). Despite overall deficit in
mortality from respiratory diseases, Wong et al. (1991) found an
excess of emphysema/chronic obstructive pulmonary disease (COPD)
mortality, which, however, was highest among workers with a duration
of exposure less than 10 years.
e) Neurotoxicity
In patients with chronic occupational exposure, neurological
disturbances include sensory-motor polyneuropathy (Perticoni et al.,
1986; Podoll et al., 1990), trigeminal sensory neuropathy, slight
pyramidal signs and cerebellar and extrapyramidal motor disorders
(Langauer-Lewowicka et al., 1983). Psychiatric disturbances included
neurasthenic or depressive syndromes (Penin et al., 1975).
Sleeplessness (Gnesina et al., 1978; Gnesina & Pshenitsina, 1980;
Langauer-Lewowicka et al., 1983; Gnesina & Teklina, 1984) and loss of
sexual functions (see below) were frequently encountered. Pathological
EEG alterations were found in a high proportion of patients (Penin et
al., 1975; Stblovį et al., 1981).
f) Immunotoxicity
The major immunological abnormalities reported in VC disease
patients include hyperimmunoglobulinaemia with a polyclonal increase
in IgG, cryoglobulinaemia, cryofibrinogenaemia, and in vivo
activation of complement (Ward et al., 1976). Immunofluorescent
examinations of skin and lung biopsies have demonstrated the
deposition of IgG with associated complements C3 and fibrin in
relation to the histological lesions described in small blood vessels
(Grainger et al., 1980). A statistically significant increase in
circulating immune complexes was observed in workers exposed to VC,
compared to unexposed workers (Ward et al, 1976) . The increase in
circulating immune complexes was greatest in women and in those with
duties involving exposure to relatively higher levels of VC
(Bogdanikowa & Zawilska, 1984).
Bencko et al. (1988) found significantly elevated IgA, IgG and
IgM levels in the serum of workers exposed to low levels of VC
(< 10 mg/m3), but found that in workers with excessive exposures to
VC (> 10 mg/m3) there was a significant drop in IgG level.
The antinuclear antibody (ANA) test is negative or at low titre
in VC disease (Ward et al., 1976), whereas it is often positive in
other scleroderma-like disorders.
There is much heterogeneity within VC disease, in terms of skin
involvement, severity, and organ involvement, and it is clear that
this could have an immunological background. Systemic sclerosis may be
a genetically linked autoimmune disease. Many autoimmune diseases show
statistically significant associations with certain
human-leukocyte-associated antigen (HLA) alleles. Black et al. (1983,
1986) compared the HLA frequencies and autoantibodies in workers with
VC disease ( n=44), asymptomatic workers ( n=30), systemic sclerosis
( n=50) and normal (blood donor) controls ( n=200). The HLA-DR5