INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 36
FLUORINE AND FLUORIDES
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
World Health Orgnization
Geneva, 1984
The International Programme on Chemical Safety (IPCS) is a
joint venture of the United Nations Environment Programme, the
International Labour Organisation, and the World Health
Organization. The main objective of the IPCS is to carry out and
disseminate evaluations of the effects of chemicals on human health
and the quality of the environment. Supporting activities include
the development of epidemiological, experimental laboratory, and
risk-assessment methods that could produce internationally
comparable results, and the development of manpower in the field of
toxicology. Other activities carried out by the IPCS include the
development of know-how for coping with chemical accidents,
coordination of laboratory testing and epidemiological studies, and
promotion of research on the mechanisms of the biological action of
chemicals.
ISBN 92 4 154096 6
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 1984
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 FLUORINE AND FLUORIDES
PREFACE
1. SUMMARY AND RECOMMENDATIONS FOR FURTHER RESEARCH
1.1. Summary
1.1.1. Analytical methods
1.1.2. Sources and magnitude of exposure
1.1.3. Chemobiokinetics and metabolism
1.1.4. Effect of fluoride on plants and animals
1.1.5. Beneficial effects on human beings
1.1.6. Toxic effects on human beings
1.2. Recommendations for further research
2. PROPERTIES AND ANALYTICAL METHODS
2.1. Chemical and physical properties of fluorine
and its compounds
2.1.1. Fluorine
2.1.2. Hydrogen fluoride
2.1.3. Sodium fluoride and other alkali fluorides
2.1.4. Fluorspar, cryolite, and fluorapatite
2.1.5. Silicon tetrafluoride, fluorosilicic
acid, and fluorosilicates
2.1.6. Sodium monoflurophosphate
2.1.7. Organic fluorides
2.2. Determination of fluorine
2.2.1. Sampling and sample preparation
2.2.1.1 Air
2.2.1.2 Soil and rocks
2.2.1.3 Water
2.2.1.4 Animal tissues
2.2.1.5 Plants
2.2.2. Separation and determination of fluoride
2.2.2.1 Colorimetric methods
2.2.2.2 The fluoride selective electrode
2.2.2.3 Other methods
3. FLUORIDE IN THE HUMAN ENVIRONMENT
3.1. Fluoride in rocks and soil
3.2. Fluoride in water
3.3. Airborne fluoride
3.4. Fluoride in food and beverages
3.5. Total human intake of fluoride
4. CHEMOBIOKINETICS AND METABOLISM
4.1. Absorption
4.2. Retention and distribution
4.2.1. The fluoride balance
4.2.2. Blood
4.2.3. Bone
4.2.4. Teeth
4.2.5. Soft tissues
4.3. Excretion
4.3.1. Urine
4.3.2. Faeces
4.3.3. Sweat
4.3.4. Saliva
4.3.5. Milk
4.3.6. Transplacental transfer
4.4. Indicator media
5. EFFECTS ON PLANTS AND ANIMALS
5.1. Plants
5.2. Insects
5.3. Aquatic animals
5.4. Birds
5.4.1. Acute effects
5.4.2. Chronic effects
5.5. Mammals
5.5.1. Acute effects
5.5.1.1 Exposure to sodium fluoride
5.5.1.2 Exposure to fluorine, hydrogen
fluoride, or silicon tetrafluoride
5.5.2. Chronic effects on small laboratory animals
5.5.3. Chronic effects on livestock
5.6. Genotoxicity and carcinogenicity
5.6.1. Genetic effects and other related end
points in short-term tests
5.6.2. Carcinogenicity in experimental
animals
5.7. Experimental caries
5.8. Possible essential functions of fluorides
6. BENEFICIAL EFFECTS ON HUMAN BEINGS
6.1. Effects of fluoride in drinking-water
6.2. Cariostatic mechanisms
6.3. Fluoride in caries prevention
6.3.1. Fluoridated salt (NaCl)
6.3.2. Fluoridated milk
6.3.3. Fluoride tablets
6.3.4. Topical application of fluorides
6.4. Treatment of osteoporosis
7. TOXIC EFFECTS ON HUMAN BEINGS
7.1. Acute toxic effects of fluoride salts
7.2. Caustic effects of fluorine and hydrogen fluoride
7.3. Chronic toxicity
7.3.1. Occupational skeletal fluorosis
7.3.2. Endemic skeletal fluorosis
7.3.3. Dental fluorosis
7.3.4. Effects on kidneys
7.4. Carcinogenicity
7.5. Teratogenicity
7.6. Effects on mortality patterns
7.7. Allergy, hypersensitivity, and dermatological reactions
7.8. Biochemical effects
8. EVALUATION OF SIGNIFICANCE OF FLUORIDES IN THE ENVIRONMENT
8.1. Relative contribution from air, food, and
water to total human intake
8.2. Doses necessary for beneficial effects in man
8.3. Toxic effects in man in relation to exposure
8.3.1. Dental fluorosis
8.3.2. Skeletal fluorosis
8.3.3. Other effects
8.4. Effects on plants and animals
8.4.1. Plants
8.4.2. Animals
REFERENCES
IPCS TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR FLUORINE
AND FLUORIDES
Members
Dr F. Berglund, Department of Medical Research, KabiVitrum
Pharmaceuticals, Stockholm, Sweden
Dr A.W. Davison, Department of Plant Biology, The University,
Newcastle-Upon-Tyne, United Kingdom (Rapporteur)
Professor C.O. Enwonwu, National Institute for Medical
Research, Yaba, Lagos, Nigeria
Professor W. Künzel, Stomatology Section, Erfurt Academy of
Medicine, Democratic Republic of Germany
Mr F. Murray, Department of Biological Sciences, University of
Newcastle, New South Wales, Australia
Professor M.H. Noweir, Occupational Health Department, High
Institute of Public Health, Alexandria, Egypt (Chairman)
Dr P. Phantumvanit, Faculty of Dentistry, Chulalongkorn
University, Bangkok, Thailand
Dr R.G. Schamschula, WHO Collaborative Unit for Caries and
Periodontal Disease Research, Institute of Dental
Research, Sydney, Australia
Professor Dr Ch. Schlatter, Swiss Federal Institute of
Technology and University of Zurich, Institute of
Toxicology, Schwerzenbach, Switzerland
Dr D.R. Taves, Department of Radiation Biology and
Biophysics, University of Rochester School of Medicine,
Rochester, New York, USAa
Representatives of Non-Governmental Organization
Professor M. Lob, Permanent Commission and International
Organization on Occupational Health (PCIAOH)
Secretariat
Professor P. Grandjean, Department of Environmental Medicine,
Odense University, Odense, Denmark (Temporary Adviser)
Dr D.E. Barmes, Oral Health, World Health Organization,
Geneva, Switzerland
---------------------------------------------------------------------------
a Invited, but could not attend.
Secretariat (contd.)
Professor M. Guillemin, Institut de Médecine du Travail et
d'Hygiène industrielle, University of Lausanne, Le
Mont-sur-Lausanne, Switzerland
Professor F. Valic, International Programme on Chemical
Safety, World, Health Organization, Geneva, Switzerland
(Secretary)
NOTE TO READERS OF THE CRITERIA DOCUMENTS
While every effort has been made to present information in the
criteria documents as accurately as possible without unduly
delaying their publication, mistakes might have occurred and are
likely to occur in the future. In the interest of all users of the
environmental health criteria documents, readers are kindly
requested to communicate any errors found to the Manager of the
International Programme on Chemical Safety, World Health
Organization, Geneva, Switzerland, in order that they may be
included in corrigenda, which will appear in subsequent volumes.
In addition, experts in any particular field dealt with in the
criteria documents are kindly requested to make available to the
WHO Secretariat any important published information that may have
inadvertently been omitted and which may change the evaluation of
health risks from exposure to the environmental agent under
examination, so that the information may be considered in the event
of updating and re-evaluation of the conclusions contained in the
criteria documents.
* * *
A detailed data profile and a legal file can be obtained from
the International Register of Potentially Toxic Chemicals, Palais
des nations, 1211 Geneva 10, Switzerland (Telephone no. 988400 -
985850).
ENVIRONMENTAL HEALTH CRITERIA FOR FLUORINE AND FLUORIDES
Further to the recommendations of the Stockholm United Nations
Conference on the Human Environment in 1972, and in response to a
number of World Health Assembly resolutions (WHA 23.60, WHA 24.47,
WHA 25.58, WHA 26.68) and the recommendation of the Governing
Council of the United Nations Environment Programme (UNEP/GC/10,
July 3 1973), a programme on the integrated assessment of the
health effects of environmental pollution was initiated in 1973.
The programme, known as the WHO Environmental Health Criteria
Programme, has been implemented with the support of the Environment
Fund of the United Nations Environment Programme. In 1980, the
Environmental Health Criteria Programme was incorporated into the
International Programme on Chemical Safety (IPCS). The result of
the Environmental Health Criteria Programme is a series of criteria
documents.
The first, second, and final drafts of the Environmental Health
Criteria Document on Fluorine and Fluorides were prepared by Dr
B.D. Dinman of the USA, Dr P. Torell of Sweden, and Professor R.
Lauwerys of Belgium.
The Task Group for the Environmental Health Criteria for
Fluorine and Fluorides met in Geneva from 28 February to 5 March,
1984. The meeting was opened by Dr M. Mercier, Manager,
International Programme on Chemical Safety, who welcomed the
participants on behalf of the three co-sponsoring organizations of
the IPCS (UNEP/ILO/WHO). The Task Group reviewed and revised the
final draft criteria document and made an evaluation of the health
risks of exposure to fluorine and fluorides.
The efforts of all who helped in the preparation and the
finalization of the document are gratefully acknowledged.
* * *
Partial financial support for the publication of this criteria
document was kindly provided by the United States Department of
Health and Human Services, through a contract from the National
Institute of Environmental Health Sciences, Research Triangle Park,
North Carolina, USA - a WHO Collaborating Centre for Environmental
Health Effects.
PREFACE
In contrast with most compounds treated earlier in this series,
fluorine and fluorides encompass both beneficial and toxic effects,
each of which have extremely important public health implications.
Fluoride illustrates strikingly the classical medical concept
that the effect of a substance depends on the dose. As Paracelsus
said, "All substances are poisons; there is none which is not a
poison. The right dose differentiates a poison and remedy". While
a continuous daily intake of milligrams per day of fluoride has
been found to be beneficial in the prevention of caries, long-term
exposure to higher quantities may have deleterious effects on
enamel and bone, and single, gram doses cause acute toxic effects
or may even be lethal.
1. SUMMARY AND RECOMMENDATIONS FOR FURTHER RESEARCH
1.1. Summary
1.1.1. Analytical methods
The measurement of fluoride in inorganic and organic materials
includes sample collection, preparation, and determination.
Preparation usually involves one or more of the following stages:
washing, drying, ashing, fusion, acid extraction, distillation, or
diffusion. Ashing and fusion may be necessary to oxidize organic
matrices and to release fluoride from refractory compounds,
respectively. Separation is used to avoid interference or as a
means of concentration. Many methods are available for the
determination of fluoride in suitably prepared samples. The most
widely used involve colorimetry or the fluoride specific ion
electrode; several methods have specialized application. The ion
electrode method is more popular than other methods because it
offers speed and relative freedom from interference; in some
circumstances, separation may not be necessary. Variation in
reported concentrations of fluoride in the same media, and the
results of inter-laboratory collaborative trials involving all
methods of determination indicate that frequently accuracy and
precision are limited by poor quality control rather than by the
method.
1.1.2. Sources and magnitude of exposure
Because it is so reactive, fluorine rarely, if ever, occurs
naturally in the elementary state, existing instead in the ionic
form or as a variety of inorganic and organic fluorides. Rocks,
soil, water, air, plants, and animals all contain fluoride in
widely-varying concentrations. As a result of this variation, the
sources and their relative importance for human beings also vary.
Fluoride enters the body by ingestion and inhalation, and, in
extreme cases of acute exposure, through the skin. Not all of the
fluoride that is ingested or inhaled is absorbed, and a proportion
is excreted by various means. Intake is lowest in rural
communities in which there are no fluoride-rich soils or waters,
and no exposure to industrial, agricultural, dental, or medical
sources. Fluoridation of water for the prevention of caries may
result in this being the largest source, if there is no exposure to
other man-made sources, such as industrial emissions. Consumption
of high-fluoride foods such as tea or some fish dishes may increase
intake significantly. The use of fluorides or fluoride-containing
materials in industry leads not only to an increase in occupational
exposure but also, in some cases, to increased general population
exposure. Significant occupational exposure occurs where control
technology is old or outmoded. However, more significant than the
preceding sources are deposits of high-fluoride rocks that in some
areas cause a large increase in the fluoride content of water or
food. There are many parts of the world where this exposure to
fluoride is sufficiently high to cause endemic fluorosis.
1.1.3. Chemobiokinetics and metabolism
A large proportion of the ingested and inhaled fluorides is
rapidly absorbed through the gastrointestinal tract and through the
lungs, respectively. Absorbed fluoride is carried by the blood and
is excreted via the renal system or taken up by the calcified
tissues. Most of the fluoride bound in the skeleton and teeth has
a biological half-life of several years. The concentration of
fluoride in the calcified tissues is a function of exposure and
age. No significant accumulation occurs in the soft tissues.
Renal excretion appears to be based on glomerular filtration
followed by a variable tubular reabsorption, which is higher at low
pH and low urinary flow rates. Fluoride passes through the
placenta and occurs in low concentrations in saliva, sweat, and
milk.
1.1.4. Effects of fluoride on plants and animals
(a) Plants
Uptake of fluoride in plants mainly occurs through the roots
from the soil, and through the leaves from the air. Fluoride may
induce changes in metabolism, decreased growth and yield, leaf
chlorosis or necrosis, and in extreme cases, plant death.
Considerable differences exist in plant sensitivity to atmospheric
fluoride, but little or no injury will occur when the most
sensitive species are exposed to about 0.2 µg/m3 air, and many
species tolerate concentrations many times higher than this.
(b) Animals
Plants are a source of dietary fluoride for animals and human
beings. Thus, elevation of plant fluoride many lead to a
significant increase in animal exposure. Chronic toxicity has been
studied in livestock, which usually develop skeletal and dental
fluorosis. Experimentally-induced chronic toxicity in rodents is
also associated with nephrotoxicity. Symptoms of acute toxicity
are generally non-specific. Fluoride does not appear to induce
direct mutagenic effects, but at high concentrations it may alter
the response to mutagens.
1.1.5. Beneficial effects on human beings
With exposure to optimal levels of fluoride in the drinking-
water (0.7 - 1.2 mg/litre, depending on climatic conditions), there
is a clearly demonstrated cariostatic effect. The extent of caries
reduction by various methods is influenced by the initial caries
prevalence and the standard of health care in the community.
Fluoride has been used in the treatment of osteoporosis for two
decades and, though beneficial effects have been reported, the
dose-response relationships and efficacy need further clarification.
1.1.6. Toxic effects on human beings
The most important toxic effect of fluoride on human beings is
skeletal fluorosis, which is endemic in areas with soils and water
containing high fluoride concentrations. The sources of fluoride
that contribute to the total human intake vary geographically
between endemic fluorosis areas, but the symptoms are generally
similar. They range from skeletal histological changes, through
increases in bone density, bone morphometric changes, and exostoses
to crippling skeletal fluorosis. This condition is usually
restricted to tropical and subtropical areas, and is frequently
complicated by factors such as calcium deficiency or malnutrition.
In non-endemic areas, skeletal fluorosis has occurred as a
result of industrial exposure. This condition, whether of endemic
or industrial origin, is normally reversible by reducing fluoride
intake.
In endemic fluorosis areas, developing teeth exhibit changes
ranging from superficial enamel mottling to severe hypoplasia of
the enamel and dentine.
Patients with kidney dysfunction may be particularly
susceptible to fluoride toxicity.
Acute toxicity usually occurs as a result of accidental or
suicidal ingestion of fluoride, and it results in gastrointestinal
effects, severe hypocalcaemia, nephrotoxicity, and shock.
Inhalation of high concentrations of fluorine, hydrogen fluoride,
and other gaseous fluorides may result in severe respiratory
irritation and delayed pulmonary oedema. Exposure of the skin to
gaseous fluorine results in thermal burns, while hydrogen fluoride
causes burns and deep necrosis.
A special case of acute toxicity is the reversible water-losing
nephritis caused by metabolic liberation of fluoride ions from
fluoride-containing anaesthetic gases.
1.2. Recommendations for Further Research
(a) Because of the large number of people affected and the
severity of the symptoms, the most important adverse effect of
fluoride on human beings is endemic skeletal fluorosis. The
problem needs a multi-disciplinary approach and good
communication among the scientists in the different areas at a
global level. The most important recommendation is that there
should be an assessment of the magnitude of the problem and
research carried out on the following:
(i) the sources of fluoride in the diet, especially
water, in different areas;
(ii) dose-response relations, and the influence of other
factors, notably malnutrition; and
(iii) the means of prevention and cure (e.g.,
defluorination).
Assessment of the problem could best be accomplished by means
of a workshop under the auspices of the WHO.
(b) Mapping of the fluoride concentrations in household water
should be carried out to determine, on the one hand, where
there might be unrecorded excessive exposure to fluoride
conducive of fluorosis, and on the other hand, where there
might be concentrations of fluoride sub-optimal for caries
prevention.
(c) Balance studies are required, especially in relation to
variation in the availability and retention of different
chemical forms of fluoride. Data are also needed on the rate
and control of release of fluoride from calcified tissues.
(d) The etiology and pathology of early skeletal fluorosis
should be further studied, particularly in relation to the
biochemistry of bone mineralization.
2. PROPERTIES AND ANALYTICAL METHODS
2.1. Chemical and Physical Properties of Fluorine and its Compounds
The terms "fluorine" and "fluoride" are used interchangeably in
the literature as generic terms. In this document, the terminology
suggested by US NAS (1971) is followed:
"This document, rather than following common usage, uses the
term "fluoride" as a general term everywhere, where exact
differentiation between ionic and molecular forms or between
gaseous and particulate forms is uncertain or unnecessary. The
term covers all combined forms of the element, regardless of
chemical form, unless there is a specific reason to stress the
gaseous elemental form F2, in which case the term "fluorine" is
used."
Fluorine and fluorides occur ubiquitously in the environment,
and because of their wide and growing use in industrial processes,
their environmental importance is increasing. The use of fluorides
in dental health care products is also growing.
Compounds dealt with in this document, besides fluorine,
include hydrogen fluoride, alkali fluorides, fluorspar, cryolite,
fluoroapatite, other inorganic fluoride compounds, and certain
organic fluorides that release fluoride when metabolized.
2.1.1. Fluorine
Fluorine has a relative atomic mass of 19; at room temperature
it is a pale, yellow-green gas. It is the most electronegative and
reactive of all elements and thus, in nature, is rarely found in
its elemental state. Fluorine combines directly at ordinary or
elevated temperatures with all elements other than oxygen and
nitrogen (Banks & Goldwhite, 1966) and therefore reacts vigorously
with most organic compounds. Fluoride ions have a strong tendency
to form complexes with heavy metal ions in aqueous solutions, e.g.,
FeF63-, AlF63-, MnF52-, MnF3-, ZrF62-, and ThF62-. The toxic
potential of inorganic fluorides is mainly associated with this
behaviour and the formation of insoluble fluorides.
Fluorine reacts with metallic elements to form compounds that
are usually ionic, both in the crystalline state and in solution.
Most of these fluorides are readily soluble in water; however,
lithium, aluminium, strontium, barium, lead, magnesium, calcium,
and manganese fluorides are insoluble or sparingly soluble. Some
high-melting fluorides such as aluminium fluoride (AlF3) are not
completely decomposed, even by boiling sulfuric acid.
Fluorine and hydrogen fluoride react with nonmetallic elements
to form covalent compounds, e.g., fluorine monoxide, silicon
tetrafluoride, sulfur hexafluoride, organic compounds containing
fluorine, and complex anionic forms. Covalent compounds of
fluorine tend to have low melting points and high volatility
(Durrant & Durrant, 1962).
2.1.2. Hydrogen fluoride
At room temperature, hydrogen fluoride is a colourless liquid
or gas with a pungent odour. Its freezing point is -83°C and its
boiling point is +19.5°C. It fumes in air; below +20°C, it is
completely soluble in water. Anhydrous hydrogen fluoride is one of
the most acidic substances known (Horton, 1962). It readily
protonates and dissolves even nonbasic compounds such as alcohols,
ketones, and mineral acids. It is a strong dehydrating agent; wood
and paper are charred on contact and aldehydes undergo condensation
by elimination of water (Gall, 1966).
In the industrial production of hydrogen fluoride, the mineral
fluorspar is treated with concentrated sulfuric acid. The volatile
hydrogen fluoride formed is then condensed and purified by
distillation. On the basis of the quantity produced, hydrogen
fluoride is the most important fluoride manufactured. About 292
000 tonnes were produced in the USA in 1977 (Chemical Marketing
Reporter, 1978b). Approximately 40% of this amount was used in the
manufacture of aluminium while 37% was converted into fluorocarbon
compounds and other products.
Because of its extensive industrial use, hydrogen fluoride is
probably the greatest single atmospheric fluoride contaminant.
However, owing to its great reactivity, it is unlikely to remain in
its original form for very long.
2.1.3. Sodium fluoride and other alkali fluorides
The alkali fluorides are typical salts. They have high melting
and boiling points and are fairly to highly soluble in water. All
alkali fluorides, with the exception of the lithium salt, absorb
hydrogen fluoride to form acid fluorides of the type MHF2, where M
is the alkali metal (Banks & Goldwhite, 1966).
Sodium fluoride is the most important of the alkali fluorides.
It is a white, free-flowing crystalline powder that is usually
prepared by neutralizing aqueous solutions of hydrofluoric acid
with sodium carbonate or sodium hydroxide. Sodium fluoride is
widely used in fluxes and has been proposed for the removal of
hydrogen fluoride from exhaust gases. Sodium fluoride was the
first fluoride compound used in the fluoridation of drinking-water
in the USA in 1950.
There are more reports of accidental intoxications caused by
sodium fluoride than by any other fluorine compound. This is
chiefly because of the confusion of edible materials with sodium
fluoride preparations domestically used for the extermination of
insects, fungi, rodents, etc.
2.1.4. Fluorspar, cryolite, and fluorapatite
From an industrial point of view, fluorspar (CaF2) is the
principal fluorine-containing mineral; the theoretical fluorine
content is 48.5%. It is mined in many countries. The world
production of fluorspar in 1979 was estimated to be 4 866 000
tonnes (US Bureau of Mines, 1980).
Cryolite (3NaF x AlF3) is a relatively rare mineral that is an
essential raw material in the aluminium industry; it has a
theoretical fluorine content of 545 g/kg. The formerly important
cryolite deposits of Greenland are now almost exhausted; today most
supplies have to be prepared synthetically (US NAS, 1971).
Fluorapatite [CaF2 x 3Ca3(PO4)3], a constituent of rock
phosphate, has a theoretical fluoride content of only 38 g/kg.
Thus, rock phosphate is unimportant as a commercial source of
fluorine. However, it is of great environmental significance as it
is the source of fluoride in some areas of endemic fluorosis and
because vast quantities are mined and consumed in the production of
elemental phosphorus, phosphoric acid, and phosphate fertilizers.
The fluorine content of the rock phosphate mined annually in the
USA has been estimated to be 729 000 tonnes (Wood, 1975) and the
total amount of fluoride emitted into the atmosphere from
industrial sources to be more than 16 300 tonnes in 1969 (US NAS,
1971).
2.1.5. Silicon tetrafluoride, fluorosilicic acid, and fluorosilicates
Silicon tetrafluoride (SiF4) is a colourless, very toxic gas
with a pungent odour. Its boiling point is -86°C and its melting
point is -90°C. When bubbled into water, hydrolysis results in
the formation of the hexafluorosilicate ion (SiF62-), which is very
toxic. Most of the fluorosilicates are soluble in water.
Silicon tetrafluoride is of important environmental
significance as it is formed in large quantities during the
combustion of coal and in the manufacture of superphosphate
fertilizers, elemental phosphorus, wet-process phosphoric acid,
aluminium, and brick and tile products. In plants, where off-gases
are scrubbed with water, most of the silicon tetrafluoride is
removed as fluorosilicic acid.
Fluorosilicic acid is a colourless flowing liquid that is
increasingly used to fluoridate drinking-water, as it is simple to
transport and store and easily provides the ideal fluoride level
for drinking-water. Sodium fluorosilicate is suitable for dry-
dosage fluoridation equipment as it is obtained as a dry, free-
flowing powder.
2.1.6. Sodium monofluorophosphate
Sodium monofluorophosphate (Na2FPO3) is another form of
synthetic fluoride that is widely used in the fluoride dentifrice
industry, since it is compatible with most abrasives used in
dentifrices.
2.1.7. Organic fluorides
Covalently-bound fluorine so closely resembles hydrogen that it
is possible, in principle, to synthesize fluoride analogues for
almost all of the hydrocarbons known at present and their
derivatives; already several thousand fluorine-containing compounds
have been prepared (Banks & Goldwhite, 1966).
(a) Fluorocarbons
Fluorination of organic compounds producing chiefly
fluorocarbons constituted the greatest single use of hydrogen
fluoride in the USA in 1977. About 108 000 tonnes were used
resulting in the production of 386 000 tonnes of fluorocarbons
(Chemical Marketing Reporter, 1978a,b). The fluorocarbons were
used in aerosol propellants (24%), refrigerants (39%), solvents
(11%), and blowing agents (12%). Owing to the banning of non-
essential uses of fluorocarbon propellants in 1978 by the US
Environmental Protection Agency, the propellant segment of the
fluorocarbon market in the USA, for example, shrank to about 2% of
its former value.
Although most of the saturated compounds of fluorine and carbon
are neither toxic nor narcotic, many of the higher unsaturated
compounds of carbon, hydrogen, fluorine, and other halogens are
very toxic (ACGIH, 1980).
(b) Methoxyflurane (2,2-dichloro-1,1-difluoro-1-methoxyethane) (Penthrane)
Methoxy-flurane (CH3-O-CF2-CCl2H), enflurane (CHF2-O-CF2-CClFH),
and isoflurane (CHF2-O-CHCl-CF3), are organofluorine anaesthetics
that release fluoride when metabolized in the body (Cousins &
Mazze, 1973; Cousins et al., 1974; Marier, 1982).
(c) Natural organic fluorides
Natural organic fluorides are rare. Only a few such compounds
have been described, the most well-known being fluoroacetic acid
and fluoro-oleic acid. These have been reported to occur in over
20 tropical or arid-zone plants; in some species to such an extent
that their leaves are poisonous to animals (Cameron, 1977;
Weinstein, 1977; King et al., 1979).
Conflicting evidence has been published concerning the possible
presence of fluoroacetate and fluorocitrate in common crop plants
exposed to fluorides. The very high concentrations reported in
some papers were found to be in error, because of contamination
with inorganic fluorides (Yu et al., 1971). From recent
literature, there does not seem to be any reason to modify the
opinion expressed by Hall (1974): "... it seems unlikely that the
levels of toxic compounds in pastures arising from the sources of
industrial contaminants mentioned earlier, if they are formed at
all, constitute a serious hazard." However, organic fluorides have
been identified in human plasma, but the significance of this
finding is still unknown (section 4.2.2).
2.2. Determination of Fluorine
The determination of elementary fluorine is difficult, chiefly
because of the great reactivity of this element. Many methods and
modifications have been proposed for the liberation of fluoride
from samples of various origins as well as for the subsequent
determination of the separated fluoride. This section deals with
the most commonly used methods. Other techniques are discussed by
Jacobson & Weinstein (1977).
2.2.1. Sampling and sample preparation
In investigations of fluoride-containing compounds present in
the environment, due care must be taken in collecting and handling
the samples in order to obtain a representative sample and to avoid
contamination with outside fluoride and loss of fluoride after the
sampling. It must be recognized that determination of fluoride is
one step in a series of operations, all of which may affect the
accuracy or validity of the final data. Errors introduced in
sampling or handling may be much greater than those due to lack of
accuracy or reliability in the analytical technique. The media to
be monitored are many and varied; each situation should be assessed
and a scheme devised for the collection, preparation, and analysis
of samples.
2.2.1.1. Air
Sampling for airborne fluorides is complicated by the very low
concentrations of these compounds generally found in the ambient
air and by the occurrence of both gaseous and particulate forms.
For instance, gaseous fluoride compounds such as hydrogen fluoride
and silicon tetrafluoride are more toxic to vegetation than most
particulate fluoride compounds (Less et al., 1975; Weinstein,
1977). Thus, it is important to use a method for collection in
which it is possible to separate the two forms, when potential
injury to vegetation is concerned (Jacobson & Weinstein, 1977).
Much equipment has been designed for the collection of air
samples. Particulate fluorides are usually collected on acid-
treated filters. Gaseous fluorides are trapped: (a) on sodium
bicarbonate in tubes or on beads; (b) in bubblers containing water
or a solution of sodium hydroxide, potassium hydroxide, or alkali
carbonates; or (c) on alkali-treated filters. Automated equipment
for determining gaseous fluorides in the air is available.
However, it is expensive, may need skilled technical attention, and
is of limited value for measuring low concentrations.
Once trapped, determination of fluorides does not present any
difficulties. Particulate fluorides, however, usually require
fusing with alkali to convert refractory fluorides into soluble
forms. Details of these methods are given in surveys by
Hendrickson (1968), MacDonald (1970), the American Industrial
Hygiene Association (1972), Israel (1974), Jacobson & Weinstein
(1977), and US EPA (1980).
An alternative or adjunct to the above method for the
determination of gaseous fluoride concentrations in air is to
measure flux to alkali impregnated surfaces (Davison & Blakemore,
1980; Alary et al., 1981).
2.2.1.2. Soil and rocks
Mixing soils of different types and horizons to provide a
single composite sample can mask important information and lead to
larger errors than separate analysis of each soil type and horizon
(Jacobson & Weinstein, 1977). Similar problems may arise in the
sampling of rocks, because wide variations in the fluoride contents
of rocks may occur. Soil samples are usually pulverized and
homogenized in ball or hammer mills. Organic matter present in the
samples is removed by ashing, generally with fluoride-free calcium
oxide as a fixative (Horton, 1962).
Determination of fluoride in soil is preceded by operations to
convert the fluoride compounds into readily soluble compounds.
Usually fusion is necessary as soils may often yield refractory
fluorine combinations of iron, aluminium, and silicon.
In many circumstances, the labile (Larsen & Widdowson, 1971)
fraction is of greater significance than total soil fluoride, as
this fraction is more available for plant and animal uptake
(Murray, 1982). A review of the methods is given in Davison
(1984).
2.2.1.3. Water
Samples from reservoirs, lakes, rivers, and seas must be
representative, and repeated collection of samples from several
sampling stations is often necessary. Sampling from different
depths is sometimes advisable, especially when studying industrial
fluoride discharged into a water recipient.
Usually, the fluoride content of the water in deep wells is
fairly constant, while water from shallow wells can present
fluctuating values. For example, values may be high during dry
periods or periods when the ground is frozen compared with
those obtained during rainy periods. The fluoride levels in
relatively shallow wells, therefore, should be assessed by repeated
determinations.
The fluoride content of samples stored in polyethylene
containers does not change significantly (Sholtes et al., 1973).
Although fluoride usually occurs in water in the ionic form,
direct quantitative determination of fluoride ions is possible only
in samples of fresh water with a low mineral content. In other
cases, a preceding step of acid distillation is recommended if a
colorimetric method is to be used, otherwise polyvalent cations
such as A13+, Fe3+, Si4+, as well as anions such as C1-, SO42-, and
PO43- may interfere. When using the ion selective electrode, no
distillation is normally required, provided the fluoride is in a
free ionic form. This may be achieved by using a buffer to
maintain a suitable total ionic strength, pH, and to avoid complex
ion formation (section 2.2.2.2).
2.2.1.4. Animal tissues
Preparation of samples usually includes mineralization before
separation of the fluoride ion. Fusing to decompose refractory
fluorides is seldom necessary as animal tissues contain little
silicon or aluminium.
Bone samples are usually collected and prepared for subsequent
analysis using the fluoride selective electrode. The samples can
be prepared in different ways. For instance, samples freed of
flesh are simply dissolved in acids (chiefly perchloric acid
HC104); ashed and dissolved in HC104 or hydrochloric acid (HC1); or
defatted, ashed, and dissolved in HC104 or HC1. Charen et al.
(1979) compared results obtained using the fluoride selective
electrode, and different methods of preparation of bone samples.
They did not find any significant systematic differences in the
results obtained between methods with or without ashing or, with or
without defatting.
Before the fluoride selective electrode became available, body
fluids (e.g., blood, serum, saliva, and urine) were generally
gently evaporated to dryness and ashed before the separation of
fluoride (US NAS, 1971). Since ashing frequently results in
refractory fluorides, the solubility of the residue is ensured by
fusing with alkali carbonate or hydroxide. Following fusion, the
melt is dissolved and the fluoride separated for determination.
2.2.1.5. Plants
Representative sampling requires thorough planning as the
fluoride content of plants varies with time, the type of soil,
meteorological conditions, the physiological condition and age of
the plant, and the nature of the fluoride emission (Jacobson &
Weinstein, 1977; US EPA, 1980). The fluoride contents of different
parts of the canopy and different organs also vary considerably.
Whether or not fluoride compounds superficially adhering to
leaves and stems of plants should be removed by washing depends on
the use of the analytical results. Washing is used when an
indication of the internal fluoride content of the plant is
required. It is not used when the purpose is to estimate the
fluoride intake of cattle from grass, forage, etc.
2.2.2. Separation and determination of fluoride
It may be necessary to separate fluoride from other
constituents of the sample to be analysed. Most frequently, the
separation is obtained by distillation or diffusion. Ion exchange
has also been proposed. Pyrohydrolysis and precipitation
techniques are used in more specialized cases.
Distillation was formerly the separation technique most
commonly used in fluorine determination. The basic Willard &
Winter (1933) method included the volatilizaton of
hexafluorosilicic acid with vapour from perchloric or sulfuric acid
at 135°C, in the presence of glass beads or glass powder. It is
still widely used and is the method used for comparing other
methods for fluoride determination.
Diffusion methods for the separation of fluoride have been
widely used for determinations in microsamples and have found many
applications in clinical and biochemical work. With the diffusion
method developed by Singer & Armstrong (1954, 1959, 1965), hydrogen
fluoride liberated by perchloric or sulfuric acid is trapped by an
alkali layer in a closed vessel.
2.2.2.1. Colorimetric methods
With certain multivalent ions, fluoride ions form stable
colourless complexes such as (A1F6)3-, (FeF6)3-, (ZrF6)2-, and
(ThF6)2-. Most of the colorimetric methods for the indirect
determination of fluoride ions are based on such complex formation,
i.e., on the bleaching resulting from reactions of fluoride ions
with coloured complexes of these metals and organic dyes. The
degree of colour change can be assessed by comparison with a
standard either by visual titration or, as in most cases, by
spectrophotometry. Commonly used reagents are zirconyl-alizarin
for visual titration and zirconyl-SPADNS or zirconium-eriochrome
cyanine R for spectrophotometry.
The alizarin fluoride blue method (Belcher et al., 1959) has
been widely applied for direct spectrophotometric determination of
fluoride. The basic reaction is that a red cerium complex with
alizarin complexone turns blue on the addition of fluoride ions.
Colorimetric methods for fluoride determination have been used
for plant and animal tissues and fluids, water, soils, foods and
beverages, and air (Jacobson & Weinstein, 1977). A semi-automated
analyser using colorimetric technique is commercially available.
2.2.2.2. The fluoride selective electrode
The fluoride selective electrode was introduced by Frant & Ross
(1966). Because of its excellent performance, speed, and general
convenience, it has become an important method for determining
fluoride in a wide variety of environmental and industrial samples
(Jacobson & Weinstein, 1977; US EPA, 1980; Victorian Committee,
1980).
The selectivity of the electrode is based on the properties of
a membrane of sparingly soluble single crystals of lanthanum,
praseodynium or neodynium fluoride. It gives an electrochemical
response that is proportional to the fluoride ion activity in the
sample.
The fluoride selective electrode is used for the determination
of fluorides in drinking-water, in industrial effluents, sea water,
air, and aerosols, flue gases, soils and minerals, urine, serum,
plasma, plants, and other biological materials (Jacobson &
Weinstein, 1977). Micromethods have been developed in which
determinations can be made in volumes as small as 10 µlitres or
less (Ritief et al., 1977). Instruments are available for the
automated monitoring of fluoride levels using the fluoride
selective electrode.
The precision and accuracy of the electrode method equal or
even exceed those of the colorimetric techniques for most samples.
2.2.2.3. Other methods
Ion chromatography has recently been introduced for the
determination of fluoride in a variety of media.
The remaining methods for determining fluoride are mostly
specialized procedures that are appropriate for selected samples or
that involve specialized facilities; they seem unlikely to find
widespread, general application (US EPA, 1980).
3. FLUORIDE IN THE HUMAN ENVIRONMENT
Fluorine ranks 13th among the elements in the order of
abundance in the Earth's crust. However, despite the prevalence of
the fluoride ion, gaseous fluorine rarely, if ever, occurs
naturally.
3.1. Fluoride in Rocks and Soil
The mean fluoride content of rocks lies between 0.1 and 1.0 g/kg.
The main primary fluoride-containing minerals are fluorspar (CaF2),
cryolite (3NaF x AlF3), and apatite (3Ca3(PO4)2 x Ca(F,OH,Cl)2),
but in most soils it is associated with micas and other clay
minerals (Davison, in press). Sodium fluoride and magnesium
fluoride are also found as natural minerals.
The mean fluoride content of mineral soils is 0.2 - 0.3 g/kg
(US NAS, 1971), whereas that of organic soils is usually lower.
However, in soils which have developed from fluoride-containing
minerals it may range from 7 (Smith & Hodge, 1979) to 38 g/kg
(Vinogradov, 1937; Danilova, 1944).
The fluoride content of top soil may be increased by the
addition of fluoride-containing phosphate fertilizers, pesticides,
irrigation water, or by deposition of gaseous and particulate
emissions. In a recent review, Davison (in press) calculated that
phosphate fertilizers typically add between 0.005 and 0.028 mg F/kg
per year to soil. A concentration of 1 µg F/m3 in air similarly
adds about 0.004 - 0.018 g/kg per year. Soils have a capacity to
fix fluoride, so depletion by leaching and removal by crops is very
slow. In the USA, one estimate of the annual loss was 0.0025 g/kg
per year (Omueti & Jones, 1977). Much research by MacIntire and
his colleagues showed that addition of fluoride did not
significantly increase uptake by plants, though there was evidence
that this might be the case in saline soils (Davison, 1984).
3.2. Fluoride in Water
Some fluoride compounds in the Earth's upper crust are fairly
soluble in water. Thus, fluoride is present in both surface- and
ground-water. The natural concentration of fluoride in ground-
water depends on such factors as the geological, chemical, and
physical characteristics of the water-supplying area, the
consistency of the soil, the porosity of rocks, the pH and
temperature, the complexing action of other elements, and the depth
of wells (Livingstone, 1963; Worl et al., 1973). Owing to these
factors, fluoride concentrations in ground-water fluctuate within
wide limits, e.g., from < 1 to 25 mg or more per litre. In some
areas of the world, e.g., India, Kenya, and South Africa, levels
can be much higher than 25 mg/litre (WHO, l970). In surface fresh
waters, less influenced by fluoride-containing rocks, the fluoride
content is usually low, 0.01 - 0.3 mg/litre (Gabovic, 1957).
Fluoride concentrations are higher in sea than in fresh water,
averaging 1.3 mg/litre (Mason, 1974). Most of the fluoride in sea
water has come from rivers. At the present rate of delivery of
fluoride from rivers to seas, it would take about one million years
to double the average concentration in sea water. However, it
appears that a steady-state equilibrium has almost been reached as
the seas lose fluoride in the form of aerosols to the atmosphere,
by precipitation as insoluble fluorides, and by incorporation in
the carbonate- or phosphate-containing tissues of living organisms
(Carpenter, 1969).
Data on the fluoride contents of natural waters and drinking-
water are available from many parts of the world (WHO, 1970).
However, sufficiently detailed information is still lacking.
Supplementation of drinking-water with fluoride has been carried
out since 1945. Today more than 260 million individuals receive
fluoridated drinking-water throughout the world. In addition,
about as many individuals are supplied with drinking-water with a
natural fluoride content of 1 mg/litre or more. A procedure such
as adding fluoride to drinking-water occasionally carries with it a
risk of overexposure. A few instances of control system breakdown
have resulted in acute intoxications of population subgroups but
the effects lasted only a short time (Waldbott, 1981).
About 50% of sewage fluoride is removed by biological treatment
(Masuda, 1964), and considerable amounts of fluoride will be
precipitated by aluminium, iron, or calcium salts during chemical
treatment. Thus, effluents from areas with fluoridation will have
a limited influence on the final fluoride level in the fresh-water
recipient (Singer & Armstrong, 1977).
3.3. Airborne Fluoride
Traces of fluoride in the air of rural communities and cities,
arise from both natural sources and human activities. The natural
dispersal of fluoride into the air has long been recognized in
regions of volcanic activity. The contribution of this source to
the Earth's atmosphere is 1 - 7 x 106 tonnes per year (US EPA,
1980). Other natural sources of fluoride in the air are the dust
from soils, and sea-water droplets, carried up into the atmosphere
by winds. However, most of the airborne fluoride found in the
vicinity of urbanized areas is generated through human activities.
It has been estimated that, in 1968, more than 155 000 tonnes of
fluoride were discharged into the atmosphere from power production
and major industrial sources in the USA (Smith & Hodge, 1979). The
aluminium industry was responsible for about 10% of this fluoride
emission. Other industrial sources include steel production
plants, superphosphate plants, and ceramic factories, coal-burning
power plants, brickworks, glassworks, and oil refineries. In many
of these industries, occupational exposures of the order of
magnitude of 1 mg/m3 may occur.
The amount of airborne fluoride increases with increasing
urbanization, because of the burning of fluoride-containing fuels
(coal, wood, oil, and peat) and because of pollution from
industrial sources. Individual types of coal contain fluoride
levels ranging from 4 to 30 g/kg (MacDonald & Berkeley, 1969;
Robinson et al., 1972). Increased burning of fuel during the
winter months results in increased concentrations of airborne
fluoride. However, in densely-populated areas, the fluoride
concentrations will only occasionally reach a level of 2 µg/m3.
In a 3-year study, Thompson et al. (1971) found that in only
0.2% of urban samples did the fluoride concentration exceed 1
µg/m3. The maximum value was 1.89 µg/m3. A survey of fluoride
in the atmosphere of some communities in the USA showed that
concentrations varied between 0.02 and 2.0 µg/m3 (US EPA, 1980).
The American data are in accordance with the European findings of
Lee et al. (1974). Near heavily industrialized Duisburg in the
Federal Republic of Germany, Schneider (1968) found a mean
concentration of 1.3 µg/m3, the 90% range being 0.5 - 3.8 µg/m3.
In the immediate vicinity of factories producing fluorides or
processing fluoride-containing raw materials, the amount of
fluoride in the ambient air may be much higher for short periods.
More recently reported values for fluoride concentrations in
ambient air near fluoride-emitting factories are usually lower
than the older values, because of improved control technology.
Fluorides emitted into the air exist in both gaseous and
particulate forms. Particulate fluorides in the air around
aluminium smelters vary in size from 0.1 mm to around 10 mm (Less
et al., 1975; Davison, in press).
3.4. Fluoride in Food and Beverages
Several thorough reviews concerning the fluoride content of
foods have been presented, e.g., McClure (1949), Truhaut (1955),
Kumpulainen & Koivistonen (1977), and Becker & Bruce (1981).
Comprehensive determinations of fluoride in foods have been
reported from Finland (Koivistonen, 1980), the Federal Republic of
Germany (Oelschläger, 1970) and Hungary (Toth & Sugar, 1978; Toth
et al., 1978). Becker & Bruce (1981) compiled data from studies
before 1956 and data from the two last decades (Table 1). With the
exception of values for fish, more recent data tend to be slightly
lower; values for meat and grain products are sometimes
considerably lower. There are other notable differences between
values given in some of the papers.
Various values for fluoride concentrations in vegetables have
been reported. Occasional values in the range of 1 - 7 mg/kg fresh
weight have been reported for spinach, cabbage, lettuce, and
parsley, while values for other vegetables have seldom exceeded 0.2
- 0.3 mg/kg. Probably, in some cases, the high fluoride values
have been caused by contamination from air, soil, pesticides, etc.
It also seems probable that some kind of contamination is
responsible for the very high values of 10.7 mg/kg and 11 mg/kg,
respectively, for polished rice given by Oelschläger (1970) and
Ohno et al. (1973) since more recent confirmation of the results
is lacking.
Table 1. Fluoride content of foods according to different
investigationsa
----------------------------------------------------------------
Food Before Oelschlager Toth & Sugar Koivistoinen
1956b (1970) Toth et al. (1980)
(1978)
(mg/kg fresh weight)
----------------------------------------------------------------
Egg products 0.3 - 1.4 - - 0.3 - 1.7
Wheat, whole 0.1 - 3.1 0.1 - 0.2 0.1 - 0.4 0.2 - 1.4
Wheat, white 0.2 - 0.9 - - 0.1 - 0.9
Other cereal 0.1 - 4.7 (rice 0.2 - - 0.1 - 2.5
products 10.7)
Pulses 0.1 - 1.3 0.1 - 14.1c 0.1 - 0.2 0.1 - 1.3
Roots 0.1 - 1.2 0.1 - 0.2 0.1 - 0.5 0.1 - 0.2
Leafy 0.1 - 2.0 0.1 - 1.1 0.1 - 1.0 0.1 - 0.8
vegetables
Other 0.1 - 0.6 0.1 - 0.3 0.1 - 0.4 0.1 - 0.3
vegetables
Fruit 0.1 - 1.3 0.1 - 0.7 0.1 - 0.4 0.1 - 0.5
Margarine 0.1 - - -
Milk 0.1 - 0.1 < 0.1 0.1 0.1
Butter 1.5 - - -
Cheese 0.1 - 1.3 0.3 - 0.3 - 0.9
Pork, fresh 0.2 - 1.2 0.3 0.2 - 0.3 0.1 - 0.3
Pork, salted 1.1 - 3.3 - 0.1 - 0.2 -
Beef 0.2 - 2.0 0.2 0.2 - 0.3 0.1 - 0.3
Other meats 0.1 - 1.2 - 0.2 - 0.7 0.1 - 0.2
Offal 0.1 - 2.6 0.3 - 0.5 0.2 - 0.6 0.1 - 0.3
Blood < 0.1 - - < 0.2
Sausages 1.7 0.3 0.1 - 0.6 0.1 - 0.4
Fish fillets 0.2 - 1.5 1.3 - 5.2 1.3 - 2.5 0.2 - 3.0
----------------------------------------------------------------
Table 1. (contd.)
-------------------------------------------------------------------
Food Before Oelschlager Toth & Sugar Koivistoinen
1956b (1970) Toth et al. (1980)
(1978)
mg/kg fresh weight)
-------------------------------------------------------------------
Fish, canned 4.0 - 16.1 - 3.8 - 9.4 0.9 - 8.0
Shellfish 0.9 - 2.0 - - 0.3 - 1.5
Eggs 0.1 - 1.2 < 0.1 0.1 - 0.2 0.3
Tea, leaves 3.2 - 178.8 100.8 - 143.6 - -
Tea, beverage 1.2 1.6 - 1.8 - 0.5
-------------------------------------------------------------------
a From: Becker & Bruce (1981).
b Danielsen & Gaarder (1955); Nömmik (1953); Truhaut (1955); von
Fellenberg (1948).
c Product dried.
According to McClure (1949), the fluoride contents of fresh
pork and fresh beef varied within the range of 0.2 - 2 mg/kg and
the range for salted beef was 1.3 - 3.3 mg/kg wet weight. For
healthy animals, none of the more recent studies have reported
values higher than 0.6 mg/kg wet weight. However, Szulc et al.
(1974) found 0.9 mg of fluoride/kg wet weight in beef from cattle
with symptoms of fluorosis. Incomplete deboning could have
contributed to certain high values reported for pork, beef, and
chicken. Kruggel & Fiels (1977) and Dolan et al. (1978) have shown
that bone fragments left in meat can increase considerably the
fluoride content of, e.g., frankfurters. Bone contains high
amounts of fluoride; 376 - 540 mg fluoride/kg bone meal was
reported by Manson & Rahemtulla (1978) and 260 - 920 mg/kg by Capar
& Gould (1979). However, the availability of fluoride in ingested
bone fragments is lower than in meat.
Fluoride values given for fish fillets vary appreciably, from
0.1 - 5 mg/kg wet weight. However, as fishbone contains
considerable amounts of fluoride, incomplete gutting could have
contributed to the high fluoride values reported. It is most
likely that bone fluoride contributed to the high fluoride values
for fish protein concentrate, e.g., 21 - 761 mg/kg dry weight,
reported by Ke et al. (1970). Canned fish contains fairly large
amounts of fluoride, mainly originating from the skeleton. In
studies by Koivistonen (1980), there were no major differences
between the amounts of fluoride in the fillets of fish from fresh
water and those of fish from marine water.
The fluoride content of water used in industrial food
production and home cooking affects the fluoride content of ready-
to-eat products. Some examples are presented in Table 2. Martin
(1951) observed that the uptake by vegetables of fluoride from
cooking water was proportional to the fluoride content of the water
over a concentration range of 1 - 5 mg/litre. The fluoride content
of vegetables cooked in fluoridated water was about 0.7 mg/kg
higher than the content of vegetables cooked in water containing a
negligible amount of fluoride (Martin, 1951). In general, the
fluoride content of processed foods and beverages prepared with
water containing a fluoride level of 1 mg/litre will contain
about 0.5 mg/kg more fluoride than those prepared with non-
fluoridated water (Marier & Rose, 1966; Auermann, 1973; Becker &
Bruce, 1981). Thus, foodstuffs processed with fluoridated water
may contain a fluoride concentration of 0.6 - 1.0 mg/kg rather than
the normal 0.2 - 0.3 mg/kg (US EPA, 1980).
Table 2. Influence of the fluoride content of process water on
fluoride levels in processed food
------------------------------------------------------------------
Content in process water
0 - 0.2 mg/litre 1.0 mg/litre
Food (mg/kg fresh weight) Reference
------------------------------------------------------------------
Bakery products 0.3 - 0.6 0.8 - 1.7 Auermann (1973)
Margarine 0.4 1.0 - 1.2
Sausage 0.4 - 1.8 0.7 - 3.3
Beer 0.3 0.7 Marier & Rose
(1966)
Vegetables 0.3 (0.1 - 0.4) 0.8
(canned) (0.6 - 1.1)
Beans and pork 0.3 0.8
(canned)
Cheese 0.2 - 0.3 1.3 - 2.2 Elgersma & Klomp
(1975)
------------------------------------------------------------------
Generally, substitutes for human milk have a relatively high
fluoride content compared with that of human milk. Infant
formulae, infant gruel, syrups, and juices prepared with
fluoridated water contain 0.9 - 1.3 mg fluoride/litre compared with
0.2 - 0.5 mg/litre if prepared with low fluoride water, i.e., water
containing < 0.2 mg/litre (Becker & Bruce, 1981). Similar results
were obtained by Singer & Ophaug (1979) who also compared fluoride
levels in fruit juices made from concentrates by the addition of
fluoridated or non-fluoridated water.
Tea leaves are usually very rich in fluoride, and levels
ranging from 3.2 - 400 mg/kg dry weight have been reported
(Canadian Public Health Association, 1979). About 40 - 90% of the
fluoride in tea leaves is eluted by brewing. The mean fluoride
concentration of tea brewed with water containing fluoride at 0.1
mg/litre was found to be 0.85 mg/litre, the upper level being 3.4
mg/litre (Anderberg & Magnusson, 1977). Duckworth & Duckworth
(1978) reported that the fluoride concentrations in tea infusions,
prepared from 12 different brands of tea, varied from 0.4 to 2.8
mg/litre. The authors estimated that the ingestion of fluoride by
tea drinkers of all ages in the United Kingdom ranged from 0.04 to
2.7 mg per day.
Other beverages are usually low in fluoride. However, mineral
waters may contain fluorine levels higher than 1 mg/litre. It is
desirable that the fluorine concentration of mineral waters be
declared on the container.
3.5. Total Human Intake of Fluoride
The fluoride contents of air, water, and food determine the
human intake of fluoride. As discussed above, there are
considerable variations in fluoride levels, and a significant
variability in human fluoride intake would therefore be expected.
The average respiration rate in an adult person is about 20 m3
per day. Thus, even if the fluoride concentration in urban air
occasionally rose to 2 µg/m3, the amount of fluoride inhaled would
only be 0.04 mg/day. Martin & Jones (1971) estimated that a person
living in central London inhaled 0.001 - 0.004 mg of fluoride per
day. They stated that this amount might be increased by a factor
of five or ten on an exceptionally foggy day. In heavily
industrialized English cities, the authors considered that the
maximal amount of fluoride inhaled daily would be of the order of
0.01 - 0.04 mg. In the close vicinity of an aluminium plant in the
Federal Republic of Germany, fluoride intake by inhalation was
calculated to be 0.025 mg/day (Erdmann & Kettner, 1975). Biersteker
et al. (1977) estimated that persons living near two industrial
sources of fluoride could inhale 0.06 mg fluoride during a day of
maximal pollution. Similar values have been reported from
fluoride-emitting industries in Sweden (SOU, 1981). As only a
proportion of inhaled fluoride is retained, actual uptake will be
less than the above estimate.
Occupational exposure may add considerably to the total intake
of fluoride. Such exposures occur in the mining and processing of
fluorspar, cryolite, and apatite (in sedimentary phosphate rock).
According to NIOSH (1977), fluorides are used in industry as a flux
in metal smelting; catalysts for organic reactions; fermentation
inhibitors; wood preservatives; fluoridating agents for drinking-
water; bleaching agents; anaesthetics; and in pesticides,
dentifrices, and other materials. They are also used or released
in the manufacture of steel, iron, glass, ceramics, pottery, and
enamels; in the coagulation of latex; in the coating of welding
rods; and in the cleaning of graphite, metals, windows, and
glassware. Assuming a total respiration rate of 10 m3 during a
working day, the daily amount of fluoride inhaled could be as high
as 10 - 25 mg, when the air concentration is at the most frequent
exposure limits of 1 -2.5 mg/m3 (ILO, 1980). Depending on hygiene
conditions, dust contamination in the industrial setting could also
add to the oral intake of fluoride.
Water requirements increase in hot climates. Based on the mean
maximum temperature, Galagan & Vermillion (1957) presented the
following widely-used formula for the calculation of the "optimum"
fluoride concentration in drinking-water in different climatic
regions: "optimum" level of fluoride in mg/litre = 0.34/(-0.038 +
0.0062 times the mean maximum temperature in degrees Fahrenheit).
In temperate areas, the "optimum" level has been established to be
about 1 mg/litre (section 6.1).
When estimating the fluoride intake during the first 6 months
of life, whether the infant is bottle- or breast-fed should be
taken into account because of the very low fluoride concentration
in breast milk. Different methods of preparing substitutes for
breast milk will result in different fluoride concentrations in the
formulae. In the USA, the mean daily fluoride intake of bottle-fed
infants during the first 6 months of life has been estimated to be
0.09 - 0.13 mg/kg body weight in fluoridated areas and a minimum of
0.01 - 0.02 mg/kg in areas without water fluoridation (Singer &
Ophaug, 1979). The corresponding estimate for areas with optimal
fluoride content in the drinking-water in Sweden is 0.13 - 0.20
mg/kg body weight and 0.05 - 0.06 mg/kg in low-fluoride areas
(Becker & Bruce, 1981). In contrast, the breast-fed infant will
only receive 0.003 - 0.004 mg fluoride/kg body weight, assuming a
fluoride level of 0.025 mg/litre in human milk (Ericsson, 1969).
The fluoride content of human milk is practically the same in low-
fluoride and fluoridated areas (Backer Dirks et al., 1974).
Between 6 and 12 months of age, the fluoride intake will be
determined mainly by the proportion of tap-water used for the
preparation of infant food. Between 1 and 12 years of age, about
half of the necessary quantity of fluids may be ingested in the
form of cow's milk with a fluoride concentration of 0.10 mg/litre
(Backer Dirks et al., 1974) or slightly more.
The intake of drinking-water in a temperate climate by direct
consumption and by addition to food, has been estimated to be 0.5 -
1.1 litre per day for children aged 1 - 12 years (McClure, 1953).
McPhail & Zacherl (1965) calculated the total amount of water
necessary for children aged 1 - 10 years to be 0.7 - 1.1 litre per
day.
The fluoride intake of adults from food and drinking-water has
been estimated in several studies. Table 3 includes data from low-
fluoride areas with drinking-water containing < 0.4 mg of fluoride
per litre. These data indicate that the daily fluoride intake does
not exceed 1.0 mg. However, certain national consumption habits,
for instance, the ingestion of tea in Asia, and seafood in some
other parts of the world, can be of significance. The various
estimates have differed significantly, possibly as a consequence of
the analytical method used. Differences can also be related to the
calculations of weight or the contribution from different
components in a characteristic diet. The total diet in communities
where the water is fluoridated may contain a mean of 2.7 mg
fluoride/day, compared with 0.9 mg/day where the water is not
fluoridated (Kumpulainen & Koivistoinen, 1977). Estimates of the
daily fluoride intake in fluoridated areas in several studies have
ranged from 1.0 to 5.4 mg (Table 4). These figures correspond to
data given in a number of papers from the USSR (Gabovich &
Ovrutskiy, 1969). With the different fluoride levels in various
food items, considerable variations in individual fluoride intake
may occur. Thus, subgroups with very low or very high fluoride
exposures through the diet may exist.
Close to a fluoride-emitting industry, limited contamination of
leafy vegetables may increase the total fluoride intake of local
residents by about 1.7% or 1.0% in non-fluoridated and fluoridated
localities, respectively (Jones et al., 1971). The fluoride intake
from animal products is practically unaffected by industrial air
pollution (US NAS, 1971; US EPA, 1980). Thus, no increase in
fluoride concentrations in soft tissues could be found in cattle
with a high fluoride intake, severe dental fluorosis, and a very
high level of bone fluoride (US EPA, 1980). Backer Dirks et al.
(1974) reported that the normal fluoride concentration in cow's
milk was 0.10 mg/litre compared with 0.28 mg/litre in milk from
cows feeding close to an aluminium plant. Poultry eggs were found
not to be affected by industrial fluoride pollution (Balazowa &
Hluchan, 1969; Rippel, 1972).
Table 3. Daily fluoride intake of adults in areas with a low
fluoride content in the drinking-water (< 0.4 mg/litre)a
--------------------------------------------------------------------------
Reference Fluoride Fluoride Total Comments
in food in liquid intake
(mg/day)
--------------------------------------------------------------------------
Armstrong & 0.27-0.32 - - Analyses of 3 meals for
Knowlton (1942) hospital staff, no
water
Machle et al. 0.16 0.30b 0.46 Analyses of one persons
(1942) 0.54 max 0.75 daily intake during 40
weeks
McClure et al. 0.3-0.5 Analyses of normal
(1944) diets for young men
Ham & Smith 0.43-0.76 0.0-0.03 0.43-0.79 Analyses of diets of 3
(1954b) young women avoiding
high fluoride foods
(tea, fish)
Danielsen & 0.56-0.57 - - Calculated intake of
Gaarder (1955) persons > 14 years
of age
Cholak (1960) 0.3-0.8 - - Excluding fluoride from
drinking-water
Kramer et al. 0.8-1.0b - - Analyses of general hos-
(1974) pital diets in 4 cities,
3 meals, no food/drink
between meals
Osis et al. 0.7-0.9b - - See Kramer et al. (1974)
(1974)
Singer et al. 0.37 0.54 0.91
(1980)
Becker & Bruce 0.41 0.20 0.61 Calculated from anal-
(1981) yses of market basket
samples and from food
consumption data
--------------------------------------------------------------------------
a From: Becker & Bruce (1981).
b Includes tea/coffee.
Table 4. Daily fluoride intakes of adults in areas with fluoridated
drinking-water (ca 1 mg/litre)
-------------------------------------------------------------------------
Reference Fluoride Fluoride Total Comments
in food in liquid intake
(mg/day)
-------------------------------------------------------------------------
San Filippo & 0.78-0.90 1.3-1.5 2.1-2.4 Analyses of 4 market
Battistone basket samples
(1971)
Marier & Rose 1.0-2.1 1.0-3.2 1.9-5.0 Calculated from the
(1966) diets of 7 laboratory
workers
Spencer et al. 1.2-2.7 1.6-3.2 3.6-5.4 Analyses of diets de-
(1969) signed for low calcium
content for 9 patients
Kramer et al. 1.7-3.4a Analyses of hospital
(1974) diets in 12 cities, 3
meals per day, no food/
drink between meals
Osis et al. 2.0a Analyses of hospital
(1974) diets in 4 cities, 3
meals, no food/drink
between meals
Osis et al. 1.6-1.8a Analyses of a metabolic
(1974) diet, 3 meals, no food/
drink between meals
Singer et al. 0.33-0.59 0.61-1.1 0.99-1.7 Calculated from analyses
(1980) of market basket samples
and from food
consumption data
Koivistoinen 0.56b
(1980)
Becker & Bruce 0.41 1.6-1.9 2.0-2.3 Calculated from
(1981) analysies and food
consumption data
-------------------------------------------------------------------------
a Includes tea/coffee.
b Includes liquid except drinking-water.
Health hazards have been associated with fluoride pollution
near industrial sources. Neighbourhood fluorosis in cattle has
been described since 1912. Results of case-finding studies in the
vicinity of facilities producing fluoride-pollution in the German
Democratic Republic revealed several cases of human skeletal
fluorosis. The total number of cases mentioned was about 50,
mostly slight cases of osteosclerosis and periosteal thickening,
but detailed clinical examination was only carried out on a few
patients (Schmidt, 1976a,b; Franke et al., 1978). Most of the
German patients had resided within 2 km of the source for at least
20 years. A few additional cases were reviewed by Smith & Hodge
(1979). Moller & Poulsen (1975) identified dust pollution from a
phosphate mine as the cause of extensive dental fluorosis in
several hundred children living within 1 - 1.5 km of the mine.
Thus, several cases of skeletal abnormalities have been identified
in a few case-finding studies in the vicinity of fluoride-emitting
production facilities. In all these cases, the emission control
technology was old or outmoded.
The human intake of fluoride may also include iatrogenic
sources. A frequent assumption is that the use of fluoridated
dentifrices and mouthrinses results in a daily fluoride uptake of
about 0.25 mg (Ericsson & Forsman, 1969), although individual
fluoride intake could conceivably be higher. Accidental intake of
sodium fluoride tablets has only occasionally lead to intoxication
in children (Spoerke et al., 1980; Duxbury et al., 1982). Adverse
effects have been attributed to daily ingestion of considerable
amounts of fluoride as a remedy for osteoporosis (Grennan et al.,
1978). Several anaesthetic gases contain fluoride. After
inhalation of these compounds, fluoride ions may be released,
resulting in considerable internal exposure to fluoride (Marier,
1982).
4. CHEMOBIOKINETICS AND METABOLISM
4.1. Absorption
Absorption of fluoride entering the gastrointestinal tract is
affected by a number of factors such as the chemical and physical
nature of the ingested fluoride and the characteristics and amount
of other components of the ingesta (US NAS, 1971). Solutions of
fluoride salts are rapidly and almost completely absorbed from the
gastrointestinal tract, probably by simple diffusion (Carlson et
al., 1960a). Fluoride from insoluble or sparingly soluble
substances, such as calcium fluoride and cryolite, is less
efficiently absorbed. However, some fluorides may be more easily
dissolved in the stomach because of the low pH, and hydrogen
fluoride will then be formed. This compound may easily penetrate
biological membranes, and its chemical reactivity is the probable
cause of the resulting gastrointestinal symptoms when large amounts
have been ingested. Recent balance studies have shown that less
than 10% of the ingested fluoride is excreted in the faeces, but
the proportion varies with circumstances (US EPA, 1980) (section
4.3.2). The simultaneous presence of strongly fluoride-binding
ions, especially calcium ions, will reduce the absorption of
fluoride (Ekstrand & Ehrnebo, 1979). In comparison with calcium,
phosphate, and magnesium, aluminium is much more effective in
reducing fluoride absorption. Thus, in patients ingesting
aluminium-containing antacids, fluoride absorption decreased to
about 40%, and the retention decreased to nil (Spencer et al.,
1980).
In the industrial environment, the respiratory tract is the
major route of absorption of both gaseous and particulate fluoride.
Hydrogen fluoride being highly soluble in water is rapidly taken up
in the upper respiratory tract (Dinman et al., 1976a). Depending
on their aerodynamic characteristics, fluoride-containing particles
will be deposited in the nasopharynx, the tracheo-bronchial tree
and the alveoli (Task Group on Lung Dynamics, 1966).
Dermal absorption of fluoride has only been reported in the
case of burns resulting from exposure to hydrofluoric acid (Burke
et al., 1973).
4.2. Retention and Distribution
4.2.1. The fluoride balance
The fluoride absorbed by the human body will circulate in the
body and then be retained in the tissues, predominantly the
skeleton, or excreted, mainly in the urine. Both uptake in
calcified tissues and urinary excretion appear to be rapid
processes (Charkes et al., 1978). The previously retained fluoride
may be slowly released from the skeleton, and this fluoride may add
to the levels in blood and urine. If this factor is taken into
account, the results of recent balance studies (Maheswari et al.,
1981; Spencer et al., 1981) in a number of subjects over several
weeks of observation suggest that retention may be 35 - 48%. Thus,
these results have, in general, confirmed the early findings of
Largent & Heyroth (1949) that daily retention of increased amounts
of fluoride intake approximates 50%. Additional metabolic studies
have been conducted using radioactive fluoride (F-18) in healthy
subjects and in patients (Charkes et al., 1978). Using published
data, these authors conducted a computer simulation of a
compartmental model for fluoride kinetics. The results suggested
that bone retains about 60% of intravenously-injected fluoride and
that the half-time for this uptake is only about 13 min; both blood
and extracellular fluid levels therefore decrease rapidly. After
ingestion of sodium fluoride, plasma fluoride levels show a much
slower change with a half-life of about 3 h (Ekstrand et al.,
1977a). This protracted course may be caused by a longer
absorption time. Approximately 99% of the fluoride in the body is
localized in the skeleton. The rest is distributed between the
blood and soft tissues.
4.2.2. Blood
The blood acts as a transport medium for fluoride. About 75%
of the blood fluoride is present in the plasma; the rest is mainly
in or on the red blood cells (Carlson et al., 1960b; Hosking &
Chamberlain, 1977). The levels of total plasma fluoride reported
in the literature before 1965 differ by several orders of magnitude
from more recently reported levels. Differences in analytical
performance may explain these discrepancies. It is now generally
accepted that fluoride in human serum exists in both ionic and
nonionic forms. This conclusion was orginally derived from the
observation by Taves (1968a) that the total fluoride content of
serum determined with the fluoride-ion selective electrode after
ashing was greater than the values obtained with procedures that
measure ionic fluoride and do not involve ashing of the specimen.
The nonionic fraction of serum fluorine was found by Taves
(1968a,b) to be nonexchangeable with radioactive fluoride, and not
ultrafilterable from human serum. Electrophoresis of human plasma
at pH 9.0 resulted in a clear separation of inorganic fluoride from
the nonionic fluorine which migrated with albumin (Taves, 1968c).
Guy et al. (1976) isolated and characterized the compounds that
comprise the major portion of the nonionic fluorine fraction of
human serum and found them to be predominantly perfluoro-fatty acid
derivatives containing six to eight carbons. They indicated that
human serum also contains much smaller quantities of other
uncharacterized organic fluorocarbons. In human serum, the
nonionic fluorine normally constitutes at least 50% of the total
fluorine. However, when fluoride intake is high, the ionic form
may predominate (Guy et al., 1976). In a group of rural Chinese,
organic fluoride constituted about 17% of the serum fluoride
(Belisle, 1981). The origin of nonionic fluorine in the serum is
still unknown (Singer & Ophaug, 1982).
For the general population under steady-state conditions of
exposure, the concentration of fluoride ions in plasma is directly
related to the fluoride content of the drinking-water. This close
relationship has been clearly demonstrated by several authors (Guy
et al., 1976; Ekstrand et al., 1978; Singer & Ophaug, 1979). The
half-time of fluoride in plasma has been found to increase with
dose, ranging from 2 to 9 h (Ekstrand, 1977; Ekstrand et al.,
1977b), perhaps related to a delayed uptake of higher doses. For
the same intake, the plasma fluoride ion concentration increases
significantly with age (Carlson et al., 1960a; Ekstrand, 1977;
Singer & Ophaug, 1979). A possible explanation of this phenomenon
in children is that uptake is faster in young bone, which is less
saturated with fluoride (Wheatherell, 1966). In addition, because
of the accumulation of fluoride in the skeleton, increased amounts
may be released from bone remodelling processes to the plasma in
older individuals.
Several studies on plasma or serum fluoride levels have been
performed, and a few should be mentioned to illustrate the
magnitude of fluoride concentrations. In 16 non-fasting young
adults from an area in which the water was fluoridated, Taves
(1966) found an average serum fluoride concentration of 13
mg/litre. In 20 adults from an area with a fluoride content of
0.18 mg/ litre in the drinking-water, Fuchs et al. (1975) found a
mean plasma fluoride ion concentration of 10.4 µg/litre. Schiffl &
Binswanger (1980) found a mean serum fluoride ion concentration of
9.8 µg/litre in 8 healthy persons living in an area with a fluoride
level of 0.06 mg/litre drinking-water. Five subjects living in an
area with a fluoride level in the drinking-water of 0.15 mg/litre
had plasma fluoride ion concentrations ranging from 27 to 99
µg/litre, whereas the plasma fluoride level in 7 subjects living in
an area where the fluoride in drinking-water might reach a value of
3.8 mg/litre ranged from 57 to 277 µg/litre (Jardillier & Desmet,
1973) . Ekstrand (1977) measured plasma fluoride concentrations in
13 fluoride-exposed workers. The concentrations were elevated
compared with a normal range of 10 - 15 µg/litre and exceeded 50
µg/litre in several workers. The maximum concentration was 91
µg/litre, 2 h after the end of exposure. These marked variations
found in different studies stress the importance of future
investigations on blood levels of fluoride, and inter-laboratory
analytical comparison programmes.
4.2.3. Bone
Fluoride ions are taken up rapidly by bone by replacing
hydroxyl ions in bone apatite. It has been suggested that fluoride
in extracellular fluid enters the apatite crystal by a three-stage
ion exchange process: the hydroxyapatite of bone mineral exists as
extremely small crystals surrounded by a hydration shell; fluoride
first enters the hydration shell, in which the ions are in
equilibrium with those of the surrounding tissue fluids and those
of the apatite crystal surface; the second stage reaction
constitutes an exchange between the fluoride of the hydration shell
and the hydroxyl group at the crystal surface; once it has entered
the surface of the crystal, fluoride is more firmly bound; in the
third stage, some of the fluoride may migrate deeper into the
crystal as a result of recrystallization. The consensus is that
absorbed fluoride is incorporated into the hard tissues largely by
a process of exchange and by incorporation into the apatite lattice
during mineralization (Neuman & Neuman, 1958; US NAS, 1971).
The amount of fluoride present in bone depends on a number of
factors including fluoride intake, age, sex, bone type, and the
specific part of the bone. About half of the absorbed fluoride is
deposited in the skeleton (section 4.2.1) where it accumulates
because of the long biological half-life of fluoride in bone.
Young animals store more of the daily intake than older ones, this
is perhaps related to the skeletal growth; this observation may
partly explain the faster removal of fluoride ion from the plasma
of young individuals and their lower fluoride ion concentrations in
plasma. The concentration of fluoride in bone increases with age
(Smith et al., 1953; Jackson & Weidmann, 1958). For example, in
cortical bone from midshaft diaphysis of human femora from areas
supplied with drinking-water containing less than 0.5 mg/litre,
Weatherell (1966) found fluoride concentrations ranging from 200 to
800 mg/kg (ash) in the age group 20 - 30 years and from 1000 to
2500 mg/kg (ash) in the age group 70 - 80 years, respectively.
Trabecular bone contains more fluoride than compact bone, and the
biologically active surfaces of bone take up fluoride more readily
than the interior (Armstrong et al., 1970). Fluoride can be
released from bone, as is evidenced by its continuous appearance in
the urine in increased amounts after exposure has ceased. Hodge &
Smith (1970) have suggested, on the basis of published data, that
such removal takes place in two phases: a rapid process taking
weeks and probably involving an ionic exchange in the hydration
shell, and a slower phase with an average half-life of about 8
years owing to osteoclastic resorption of bone. Human data have
suggested that 2 - 8% of fluoride retained is excreted during 18
days following the initial retention (Spencer et al., 1975, 1981).
Because of slower remodelling process, fluoride would be released
even more slowly from compact than from trabecular bone. Limited
information on 43 cases of skeletal fluorosis suggests that the
fluoride content from iliac crest biopsies may be reduced by one-
half, 20 years after cessation of exposure (Baud et al., 1978).
4.2.4. Teeth
The factors controlling the incorporation of fluoride into
dental structures have been reviewed by Weidemann & Weatherell
(1970); they are essentially the same as those pertaining to bone.
Cementum is more akin to bone than to enamel and dentin, but
its fluoride concentration has been found to be higher than that of
bone (Singer & Armstrong, 1962). Cementum exposed to oral fluids
by recession of the gingiva may accumulate considerable amounts of
fluoride.
Once formed, enamel and dentin differ from bone in that they do
not undergo continuous remodelling. The fluoride content of enamel
is acquired partly during development and partly from the oral
environment after eruption. While the concentration of enamel
fluoride decreases exponentially with distance from the surface,
the actual values also vary with site, age, surface attrition, and
increases with systemic and topical exposure to fluoride
(Weatherell et al., 1977; Schamschula et al., 1982). In adult
teeth, the fluoride content of the surface layer of enamel
(thickness 10 mm) is reported to be 900-1 000 mg/kg in areas with
low fluoride levels in the water, about 1 500 mg/kg in fluoridated
areas, and about 2 700 mg/kg in areas with fluoride concentration in
the drinking-water of 3 mg/litre (Berndt & Sterns, 1979). High
fluoride content of enamel is associated with decreased solubility
(Isaac et al., 1958) and probably with increased resistance to
caries (Schamschula et al., 1979).
The average concentration of fluoride in dentine is 2 - 3 times
that in enamel and is affected by growth and mineralization. As
with bone and enamel, dentin fluoride levels are higher in the
surface (circumpulpal) regions than in the interior (US NAS, 1971).
4.2.5. Soft tissues
The concentrations of fluoride in human soft tissues reported
by different authors vary greatly. It is generally agreed,
however, that the normal soft tissue fluoride concentration in
human beings is low, usually less than 1 mg/kg wet weight (US EPA,
1980). Fluoride has a relatively short biological half-life in
these organs, and the soft tissue fluoride concentration is
therefore practically in equilibrium with that in the plasma.
Unlike fluoride in bone, the concentration does not increase with
age or duration of exposure (Underwood, 1971). However, ectopic
calcification loci may accumulate fluoride in certain tissues, e.g.
aorta, tendons, cartilage, and placenta (Hodge & Smith, 1970).
4.3. Excretion
The principal route of fluoride excretion is via the urine.
Some excretion takes place through sweat and faeces, and fluoride
also appears in saliva. Fluoride crosses the placenta; it rarely
seems to be excreted in milk to any significant extent.
4.3.1. Urine
In adults, approximately half of the absorbed fluoride is
excreted via the urine (section 4.2.1). Renal fluoride ion
excretion involves glomerular filtration followed by pH-dependent
tubular reabsorption. Fluoride clearance is less than that of
creatinine (typically about 0.15 l/h per kg body weight, according
to Ekstrand et al. (1977b)). Fluoride appears rapidly in the urine
after absorption. Following a single oral dose of a soluble
fluoride compound, the maximal rate of excretion is observed 2 - 4 h
after fluoride intake; the half-time for the fast compartment after
gastrointestinal absorption averages about 3 h (Ekstrand et al.,
1977b), but injected fluoride is excreted even more rapidly
(Charkes et al., 1978). Several factors may influence the urinary
excretion of fluoride, such as total current intake, previous
exposure to fluoride, age, urinary flow, urine pH, and kidney
status (Whitford et al., 1976; Ekstrand et al., 1978, 1982; Schiffl
& Binswanger, 1980). In urine, fluoride exists both as the ion (F-)
and to a small extent as HF. The equilibrium between F- and HF is
pH-dependent. The tubular reabsorption of fluoride occurs mainly
as HF and is therefore greater in acid urine (Whitford et al.,
1976). Fluoride excretion can therefore be increased by
maintaining alkalosis in a poisoned patient. In a study where
alkaline urine was produced by a vegetarian diet and acid urine by
a protein-rich diet, renal fluoride clearance was significantly
related to urinary pH and also to urinary flow (Ekstrand et al.,
1982). In practice, exposure is the most important factor and
urinary fluoride concentration is recognized as one of the best
indices of fluoride intake.
On a group basis, the correlation between the fluoride
concentration in urine and that in drinking-water is excellent.
This finding implies that, during periods of relatively constant
fluoride supply, there exists an almost steady-state relationship
between absorbed fluoride and fluoride excreted in the urine.
However, some of the fluoride excreted originates from fluoride
release during bone remodelling. Thus, excretion rates may
increase slightly with age, but no sex difference in fluoride
excretion has been found (Vandeputte et al., 1977; Toth & Sugar,
1976). In patients with skeletal fluorosis from an area where this
disease occurs endemically, the urinary excretion of fluoride was
related to the severity of the disease and, to some degree, to the
length of exposure (Rao et al., 1979). Excess excretion rates may
continue for years after the cessation of high exposure (Linkins et
al., 1962; Grandjean & Thomsen, 1983).
Younger persons who are actively forming bone minerals excrete
less fluoride, i.e., a lower proportion of the absorbed dose, than
adults. Zipkin et al. (1956) examined the urinary fluoride
concentrations of children and adults before and after the start of
fluoridation of the drinking-water supply. Already after one week,
the urinary fluoride levels of adults had reached 1 mg/litre. In
contrast, it took several years for the urinary fluoride of the
children to reach the same concentration. In chronic renal
failure, the urinary excretion of fluoride is diminished when
creatinine clearance values drop below 25 ml/min (Schiffl &
Binswager, 1980). In such cases, the impairment of urinary
fluoride excretion is also reflected by an increase in the fluoride
content of bone (Parsons et al., 1975). The health significance of
fluoride in dialysis fluids is discussed in section 7.3.4. In
situations with extremely high plasma levels of fluoride, e.g.,
following anaesthesia with methoxyflurane, acute kidney dysfunction
may ensue with decreased clearance of fluoride.
4.3.2. Faeces
The proportion of ingested fluoride that is eliminated in the
faeces varies depending on circumstances (US EPA, 1980; Maheshwari
et al., 1981; Spencer et al., 1981). Fluoride present in faeces
results from two sources: the ingested fluoride that is not
absorbed and the absorbed fluoride that is reexcreted into the
gastrointestinal tract. In persons not occupationally exposed to
fluoride and not using fluoridated water, the faecal elimination of
fluoride is usually less than 0.2 mg/day (US NAS, 1971).
4.3.3. Sweat
Usually, only a few percent of the fluoride intake is excreted
in the sweat. However, under excessive sweating as much as 50% of
the total fluoride excreted may be lost via perspiration (Crosby &
Shepherd 1957).
4.3.4. Saliva
Less than 1% of absorbed fluoride is reported to appear in the
saliva (Carlson et al., 1960a; Ericsson, 1969). Saliva fluoride
levels were found to be about 65% of plasma levels (Ekstrand et
al., 1977a). In fact, saliva does not represent true excretion,
because most of the fluoride will be recycled in the body.
However, the fluoride content of the saliva is of major importance
for maintaining a fluoride level in the oral cavity.
4.3.5. Milk
The concentration of fluoride in human milk is quite similar to
that in plasma (Ekstrand et al., 1981b), and significant exposure
to fluoride through human milk is therefore very unlikely. In
fact, fluoride levels in human milk are lower than those in milk
substitutes (Backer Dirks et al., 1974).
4.3.6. Transplacental transfer
Fluoride crosses the placenta. A study by Armstrong et al.
(1970) measured fluoride from maternal uterine vessels and the
umbilical vein and artery at caesarean section in human patients
and did not find any significant gradient between maternal and
fetal blood levels. At higher fluoride levels, a partial barrier
may exist (Gedalia, 1970). The fluoride content of the fetal
skeleton and teeth increases with the age of the fetus and with the
fluoride concentration of the drinking-water used by the mother
(Gedalia, 1970).
4.4. Indicator Media
Under steady-state conditions of exposure, the plasma fluoride
concentration is a reflection of the balance between fluoride
absorption, excretion, and transfer to, and release from, storage
depots. Several authors have found a relationship between fluoride
ion levels in plasma and fluoride intake (sections 4.2.1 and
4.2.2). Previous methods for fluoride determination needed an
intravenous blood sample, but micro-methods using the fluoride ion
selective electrode have now made capillary blood sampling
feasible, if contamination from the skin surface can be excluded.
Thus, plasma (or serum) may become a useful indicator medium in the
future.
Urinary fluoride has usually been used to estimate the absorbed
amount (Kaltreider et al., 1972; Pantchek, 1975; Dinman et al.,
1976a,b). In persons not occupationally exposed to fluoride, the
fluoride level in urine is almost the same as the fluoride
concentration in the drinking-water. In occupational fluoride
exposure, the results of a retrospective study by Dinman et al.
(1976a) suggest that group post-shift urinary fluoride
concentrations averaging less than 8 mg/litre over a long period
were not associated with enhanced risk of skeletal fluorosis and
the same result appears to apply if preshift urinary fluoride
concentrations are less than 4 mg/litre. However, the presence of
skeletal fluorosis in 43 aluminium potroom workers, of whom 37 had
a urinary fluoride excretion below 4 mg/24 h during an exposure-
free period (Boillat et al., 1979), may cast some doubt on the
validity of this limit, since the exposure causing the disease was
probably much higher several years previously. With exposure
mainly by the respiratory route, an average urinary fluoride
concentration in postshift samples of 8 mg/litre in aluminium
workers was found to correspond to an exposure of 2 mg/m3 (Dinman
et al., 1976b). However, because of the rapid excretion process,
the timing of urine sampling is crucial. Since it is usual for
only spot urine samples to be available, correction of the widely
varying urine volumes per time unit is advisable. Correction to a
standard density, to a defined amount of creatinine or to
osmolality is used. Furthermore, a postshift level below a certain
limit on one day does not exclude that this limit may be exceeded
on other days, if exposure conditions are somewhat variable. Also,
since a number of factors, including urinary flow and pH, may
influence the fluoride concentration in the urine, it is not
possible to make an accurate assessment of individual fluoride
status on the basis of the urinary fluoride level in a single
sample.
In addition, nails and hair may be useful indicators of long-
term fluoride exposures, under conditions where external
contamination can be excluded.
5. EFFECTS ON PLANTS AND ANIMALS
5.1. Plants
Plants are exposed to fluoride in the soil, and in the air as a
result of volcanic activity, natural fires, wind-blown dusts,
pesticides, or as emissions from processes in which fluorine-
containing materials are burned, manufactured, handled, or used (US
NAS, 1971). The main route of entry of fluoride into animals is by
ingestion, so plants are important vectors of the element in all
ecosystems.
Fluoride is taken up from the soil by passive diffusion, then
it is carried to the shoot by transpiration. In temperate
climates, and in most soils, the amount accumulated in this way is
small so the average content of leaves in a non-polluted atmosphere
is usually less than 10 mg F/kg dry weight. Where soils are saline
or enriched by fluoride-containing minerals or the atmosphere
contains elevated fluoride concentrations, the concentration may be
much higher. In such areas, there may be sufficient plant uptake
of fluoride to contribute significantly to the human or animal
diet. This factor should be considered in areas with endemic
fluorosis. A number of species accumulate high concentrations,
even when grown on low-fluoride soils, perhaps as a result of
complex formation with aluminium (Davison, 1984). The tea family,
Theaceae, is the best known of these accumulators, but there are
several others that warrant further investigations (Davison, 1984).
Gaseous and particulate fluorides in the air are deposited on
exposed plant surfaces, whilst gaseous fluoride enters leaves
through stomatal pores. Fluoride is also constantly lost from
plants by a variety of little-understood processes (Davison, 1982,
1984). Superficial deposits may be tenaciously held and may
account for over 60% of the total fluoride content of the leaf.
Though such deposits are of negligible toxicity to the plant, they
may present a hazard for grazing animals. Fluoride that penetrates
the internal tissue of leaves or that is deposited on active
surfaces such as stigmata may affect a variety of metabolic
processes and result in effects on appearance, growth, or
reproduction. Recent reviews of the metabolic effects of fluoride
have been reported by Bonte (1982) and by Weinstein & Alscher-
Herman (1982).
The visible effects of toxic concentrations of fluoride on
plants are well documented (Jacobson & Hill, 1970; Weinstein,
1977). They may include chlorosis, peripheral necrosis, leaf
distortion, and malformation or abnormal fruit development. None
of these symptoms are specific to fluoride, and the effects of many
other stresses may appear very similar. The diagnosis of fluoride
injury normally involves both visual and chemical evidence, and
comparison of a number of species of known tolerance growing around
the source. Factors relating to the frequency of exposure have
also to be taken into account.
The susceptibility of different plant species to excessive
atmospheric fluoride varies considerably (Jacobson & Hill, 1970; US
NAS, 1971). Many conifers are very susceptible during the short
period of needle growth, but they then become much more resistant.
Some monocotyledons, such as gladiolus and tulip, are similarly
susceptible, though there is great varietal variation. In some
species, there is a great difference in susceptibility between
leaves and fruits. For example, peach fruit are extremely
sensitive to very low concentrations of fluoride, but leaves are at
least an order of magnitude more resistant.
Available evidence (Weinstein, 1977; Davison, 1982) indicates
that visible injury and effects on growth or yield are to a large
extent independent. Many cases have been reported where there was
visible foliar injury but no associated growth effects. Equally,
there have been instances where visible symptoms were combined with
stimulation of some parameters of growth, probably by alteration of
resource allocation. Most significant, however, are reports that
there may be economically significant reductions in yield with no
visible symptoms on the leaves (MacLean & Schneider, 1971; Pack &
Sulzbach, 1976; Unsworth & Ormrod, 1982). This last aspect of
effects on plants needs clarification.
Attempts have been made to devise air quality criteria for the
protection of plants, notably by McCune (1969), who produced dose-
response curves for a number of species. Generally, there is a
non-linear, negative relationship between concentration and the
length of exposure necessary to cause an effect, so air quality
criteria must be stated in terms of time-related concentration.
Tissue concentrations are a useful adjunct to diagnosis and to
quality criteria, but superficial fluoride deposits and
compartmentation within the leaf make interpretation difficult. A
useful summary of air quality criteria adopted by different
organisations for plant protection is given in IPAI (1981). Such
criteria can be adjusted from place to place and from time to time
to take account into: (a) the dissimilarity of vegetation and its
consequent sensitivity in different areas; (b) changes in
susceptibility of vegetation to fluoride during the year; and (c)
the intended use of the vegetation (MacLean, 1982).
Generally, little or no injury will occur when the most
sensitive species are exposed to a fluoride level of about 0.2
mg/m3. Most species tolerate concentrations many times higher than
this. It is difficult to define the minimum tissue fluoride
concentration associated with injury; however, there are reports of
some species showing effects with concentrations as low as 20 mg/kg
dry weight (Weinstein, 1977).
Fluoride taken up by plants from soil or air is transferred to
animals by ingestion of plant cellular fluids, nectar, pollen,
tissues, or whole organs. Because the concentration of fluoride
varies greatly in different parts of plants, the amount ingested by
an animal depends on its feeding strategy. For example, animals
that consume whole shoots will ingest much greater quantities of
fluoride than phloem-sucking invertebrates. Food preparation
reduces the amount of fluoride ingested from contaminated
vegetables by human beings, because the outer leaves are removed
and the material is washed before eating.
Because of the potential economic importance of fluoride
accumulation in livestock and the role of plants in fluoride
transfer to animals, air quality criteria designed to protect
livestock from fluoride injury are usually based on the fluoride
content of forage, although the role of fluoride in dietary
supplements must also be considered (section 5.5.3).
5.2. Insects
Both inorganic and organic fluoride compounds have been used as
insecticides for many years. In sub-lethal doses, the former have
been shown to reduce growth and reproduction in many species of
invertebrates (US EPA, 1980). It has been suggested that fluoride-
insect interactions have been responsible for extensive insect
damage to forests around aluminium smelters, although the mechanism
of this interaction is not clear (Weinstein, 1977; Alcan
Surveillance Committee, 1979).
Honey bees are known to be susceptible to fluoride, and
apiarists have suffered significant economic damage in areas around
some sources of fluoride emission.
Insects collected from fluoride-polluted areas show higher
concentrations of this element than those from non-polluted areas,
and it has been suggested that this is partly due to food chain
accumulation (US EPA, 1980). However, firm evidence concerning
biomagnification is lacking.
Genotoxic effects are discussed in section 5.6.
5.3. Aquatic Animals
Reactions to fluoride have been examined in several studies on
aquatic animals , chiefly on fish, to provide a basis for
regulations on the permissible amount of fluoride in waste water
discharged into the sea or fresh water recipients.
Fish exposed to poisonous amounts of sodium fluoride (Tables 5,
6) become apathetic, lose weight, have periods of violent movement,
and wander aimlessly. Finally, there is a loss of equilibrium
accompanied by tetany and death. Mucous secretion increases,
accompanied by proliferation of mucous-producing cells in the
respiratory and integumentary epithelium (Neuhold & Sigler, 1960).
Studies on the effects of fluoride on aquatic animals (some
results are given in Tables 5 & 6) show that sensitivity and lethal
doses are influenced by many factors, e.g., size of fish, density
of fish per m3 of aquarium, water temperature, calcium and chloride
concentrations in the water, and proper maintenance of streaming
water. Crustaceans may be more tolerant to fluorides than fish (US
EPA, 1980). However, the studies give only scattered information
concerning the effects of fluoride on the fish under various living
conditions. New and more systematic and detailed studies
concerning the long-term influence of fluorides on aquatic animals
are therefore necessary.
5.4. Birds
The bones of birds collected near emission sources show
elevated fluoride levels, but there are no reports of any other
effects. Most reports on the effects on birds pertain to domestic
species such as chickens, turkeys, quails, etc. The paucity of
reports on wild birds may be the consequence of their lower
economic value. In addition, the mobility of birds makes it
difficult to define the exposure to fluoride. High fluoride
ingestion by birds can result in reduced growth rate, leg weakness,
and bone lesions. Tolerance to fluorides varies among bird species
and among individuals of the same species (US NAS, 1971, 1974; US
EPA, 1980).
5.4.1. Acute effects
Typical symptoms of acute toxicity are reduction or loss of
appetite, local or general congestion, and sub-mucosal haemorrhages
of the gastrointestinal tract (Cass, 1961; US EPA, 1980). Such
acute responses were recognized when chickens were fed for 10 days
on a diet containing 6786 mg F- per kg (as sodium fluoride).
Roosters receiving sodium fluoride at 200 mg/kg body weight, twice
in 24 h, developed gastro-enteritis with oedema of the mucosa of
the stomach and upper bowels, subcutaneous oedema, hepatomegaly,
and atrophy of the pancreas.
5.4.2. Chronic effects
Chronic fluorosis in birds can be difficult to diagnose, partly
because birds do not have teeth, which are important aids to
diagnosis in other animals (US NAS, 1974). It is necessary to
establish the presence of characteristic lesions, a history of
exposure at the proper time to levels of fluoride known to be
toxic, and analytical evidence that bone contains fluoride
concentrations consistently associated with the lesions of chronic
fluorosis, before a definite diagnosis can be made (Cass, 1961).
In birds, chronic fluoride toxicosis develops slowly and
primarily involves gross and microscopic changes in bone. If
elevated fluoride intake persists, the general health of the
animals deteriorates progressively. Growth rate decreases,
lameness may develop, and usually there is loss of appetite (US
EPA, 1980). Levels of fluoride in the ration that can be tolerated
have been given as 300 mg/kg for growing chicks and 400 mg/kg for
laying hens and turkeys (US NAS, 1974).
The body weight, tibia weight in Japanese quail, and the bone
ash, and eggshell thickness were not affected by a sodium fluoride
concentration in the drinking-water of 50 mg/litre (Vohra, 1973).
Table 5. Effect of excessive fluoride on fresh water fish
---------------------------------------------------------------------------------
Species Fluoride Exposure Effect Reference
(mg/litre) time
---------------------------------------------------------------------------------
Goldfish 1000 60 h No survival Ellis (1937)
Carp 75 - 91 480 h 50% survival Neuhold & Sigler (1960)
Red-eye fry < 25 5 - 6 days None Vallin (1968)
Red-eye roe < 25 7 days None Vallin (1968)
Juvenile salmon 100 5 days Survival Vallin (1968)
Juvenile trout 200 5 days Survival Vallin (1968)
(brackish
water)
Brown trout fry 15 240 h 50% survival Wright (1977)
Brown trout fry 2.0 240 days uncertain Wright (1977)
Brown trout fry 0.9 240 days None Wright (1977)
Rainbow trout 2 - 4 10 days uncertain Angelovic et al. (1961)
Rainbow trout 5.9 - 7.5 10 days 50% survival Angelovic et al. (1961)
Rainbow trout 8.5 504 h 95% survival Herbert & Shurben (1964)
Rainbow trout 4.0 504 h 50% survival Herbert & Shurben (1964)
Rainbow trout egg 222 - 273 424 h 50% survival Neuhold & Sigler (1960)
Rainbow trout fry 61 - 85 825 h 50% survival Neuhold & Sigler (1960)
---------------------------------------------------------------------------------
Table 6. Effect of excessive fluoride on marine animals
-----------------------------------------------------------------------------------
Species Fluoride Exposure Effect Reference
(mg/litre) time
-----------------------------------------------------------------------------------
Mugil cephalus 100 96 h None Hemens & Warwick (1972)
(mullet)
Mugil cephalus 5.5 113 days None Hemens et al. (1975)
Mugil cephalus 52 72 days Increased Hemens & Warwick (1972)
mortality
Ambassis safgha 100 96 h None Hemens & Warwick (1972)
(small fish)
Therapon jarbua 100 96 h None Hemens & Warwick (1972)
(small fish)
Penaeus indicus 5.5 113 days None Hemens et al. (1975)
(prawn)
Penaeus indicus 100 96 h None Hemens & Warwick (1972)
Penaeus monodon 100 96 h None Hemens & Warwick (1972)
Tylodiplax bleph- 52 72 days Increased Hemens & Warwick (1972)
ariskios (crab) mortality
Tylodiplax bleph- 100 96 h None Hemens et al. (1975)
ariskios
Palaemon pacificus 52 72 days Affected Hemens & Warwick (1972)
reproducibility
Perna perna 7.2 5 days Evidence of Hemens & Warwick (1972)
(brown mussel) toxic effects
-----------------------------------------------------------------------------------
5.5. Mammals
The toxicity of various fluorides has been studied mainly in
two categories of animals, i.e., laboratory animals (rats, mice,
guinea-pigs, rabbits, dogs, and cats) and live-stock. The acute
and chronic effects have usually been examined in studies on
laboratory animals, especially rats. The chronic effects have been
extensively studied in large and long-term studies on domestic
mammals.
5.5.1. Acute effects
5.5.1.1. Exposure to sodium fluoride
For laboratory animals, the single lethal dose of F-, when
administered orally as easily soluble fluorides, is in the range of
20 - 100 mg/kg body weight (Davis 1961; Eagers, 1969). The lethal
dose for intravenous, intraperitoneal, and subcutaneous injection
of sodium fluoride is half of the oral lethal dose (Muehlburger,
1930). Fatal acute intoxication may occur in laboratory animals
following repeated oral administration of sublethal doses of
soluble fluorides.
Signs of acute systemic fluoride intoxication are increased
salivation, lacrimation, vomiting, diarrhoea, muscular
fibrillation, and respiratory, cardiac, and general depression.
The rapidity of onset and the progression of the intoxication
varies directly with the magnitude of the initial dose (Davis,
1961). The anatomical lesions of fatal acute intoxication are non-
specific, but the gastro-enteric irritation is more general and
more intense than that usually found in most other forms of gastro-
enteritis.
Several authors have determined the levels of ionic fluoride in
the plasma of laboratory animals, that are sufficiently high to
result in acute fluoride poisoning and ultimately death. De Lopez
et al. (1976) determined the LD50 for female rats, weighing 80,
150, or 200 g, when given sodium fluoride by stomach intubation.
The LD50 for 80 g rats and 150 g rats was practically the same (54
and 52 mg/kg body weight, respectively), but this value was
decidedly higher than that for 200 g rats (31 mg/kg body weight).
The low LD50 observed in the oldest (heaviest) rats was ascribed to
a higher degree of fluoride saturation in their skeletons. The
plasma ionic fluoride concentration associated with the LD50s
ranged from 8 - 10 mg/litre and spontaneous death occurred in all
three groups at these levels. The maximum fluoride levels were
reached within 15 min of administration, and levels of at least 4
mg/litre persisted for 4 h or more.
Singer et al. (1978) studied the ionic fluoride levels in
plasma following intraperitoneal administration of 15, 20, or 25 mg
of fluoride per kg body weight to 200 g rats. In animals given 25
mg/kg, the mean ionic fluoride level in plasma was 38 mg/litre
after 10 min and the animals invariably died within 1 h. All
animals receiving 15 or 20 mg/kg survived, despite mean ionic
fluoride levels in plasma of 22.9 and 29.2 mg/litre, respectively.
These levels are considerably higher than the levels that resulted
in death in the previously mentioned study by De Lopez et al.
(1976). Singer & Ophaug (1982) explained this seeming disagreement
in the following way: "Administration of the fluori