Anda di halaman 1dari 190

Fluoride and the Oral Environment

Monographs in Oral Science

Vol. 22

Series Editors

A. Lussi Bern
M.C.D.N.J.M. Huysmans Nijmegen

H.-P. Weber Boston, Mass.

Fluoride and the
Oral Environment

Volume Editor

Marília Afonso Rabelo Buzalaf Bauru

50 figures, 31 in color, and 24 tables, 2011

Basel · Freiburg · Paris · London · New York · New Delhi · Bangkok ·

Beijing · Tokyo · Kuala Lumpur · Singapore · Sydney
‘This monograph is dedicated to Gary Milton Whitford, my master and friend,
who guided me through the avenues of Fluoride Research.’
Marília Afonso Rabelo Buzalaf
Marília Afonso Rabelo Buzalaf
Department of Biological Sciences
Bauru Dental School, University of São Paulo
Al. Octávio Pinheiro Brisolla, 9-75
Bauru-SP (Brazil)

This volume received generous financial support from Procter & Gamble

Library of Congress Cataloging-in-Publication Data

Fluoride and the oral environment / volume editor, Marília Afonso Rabelo
Buzalaf Bauru.
p. ; cm. -- (Monographs in oral science, ISSN 0077-0892 ; v. 22)
Includes bibliographical references and indexes.
ISBN 978-3-8055-9658-9 (hard cover : alk. paper) -- ISBN 978-3-8055-9659-6
1. Dental caries--Prevention. 2. Fluorides--Therapeutic use. I. Buzalaf,
Marília Afonso Rabelo. II. Series: Monographs in oral science ; v. 22.
[DNLM: 1. Fluorides, Topical--therapeutic use. 2. Dental
Caries--prevention & control. 3. Fluorosis, Dental. W1 MO568E v.22 2011 /
QV 282]
RK331.F558 2011

Bibliographic Indices. This publication is listed in bibliographic services, including MEDLINE/Pubmed.

Disclaimer. The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of
the publisher and the editor(s). The appearance of advertisements in the book is not a warranty, endorsement, or approval of the products or
services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or
property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.
Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord
with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations,
and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for
any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a
new and/or infrequently employed drug.
All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic
or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing
from the publisher.
© Copyright 2011 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland)
Printed in Switzerland on acid-free and non-aging paper (ISO 9706) by Reinhardt Druck, Basel
ISSN 0077–0892
ISBN 978–3–8055–9658–9
e-ISBN 978–3–8055–9659–6

IX List of Contributors
XI Foreword
Robinson, C. (Leeds)

Fluoride Intake, Metabolism and Toxicity

1 Fluoride Intake of Children: Considerations for Dental Caries and Dental Fluorosis
Buzalaf, M.A.R. (Bauru); Levy, S.M. (Iowa City, Iowa)
20 Fluoride Metabolism
Buzalaf, M.A.R. (Bauru); Whitford, G.M. (Augusta, Ga.)
37 Contemporary Biological Markers of Exposure to Fluoride
Rugg-Gunn, A.J. (Newcastle upon Tyne); Villa, A.E. (Santiago); Buzalaf, M.A.R. (Bauru)
52 Historical and Recent Biological Markers of Exposure to Fluoride
Pessan, J.P. (Araçatuba); Buzalaf, M.A.R. (Bauru)
66 Acute Toxicity of Ingested Fluoride
Whitford, G.M. (Augusta, Ga.)
81 Chronic Fluoride Toxicity: Dental Fluorosis
DenBesten, P.; Li, W. (San Francisco, Calif.)

Impact of Fluoride in the Prevention of Caries and Erosion

97 Mechanisms of Action of Fluoride for Caries Control
Buzalaf, M.A.R. (Bauru); Pessan, J.P. (Araçatuba); Honório, H.M. (Bauru);
ten Cate, J.M. (Amsterdam)
115 Topical Use of Fluorides for Caries Control
Pessan, J.P. (Araçatuba); Toumba, K.J. (Leeds); Buzalaf, M.A.R. (Bauru)
133 Systemic Fluoride
Sampaio, F.C. (João Pessoa); Levy, S.M. (Iowa City, Iowa)

146 Oral Fluoride Reservoirs and the Prevention of Dental Caries
Vogel, G.L. (Gaithersburg, Md.)
158 Fluoride in Dental Erosion
Magalhães, A.C. (Bauru); Wiegand, A. (Zurich); Rios, D.; Buzalaf, M.A.R. (Bauru); Lussi, A. (Bern)

171 Author Index

172 Subject Index

VIII Contents
List of Contributors

Marília Afonso Rabelo Buzalaf Adrian Lussi

Department of Biological Sciences Department of Preventive, Restorative and
Bauru Dental School, University of São Paulo Pediatric Dentistry, University of Bern
Al. Octávio Pinheiro Brisolla, 9-75 Freiburgstrasse 7
17012-901 Bauru-SP (Brazil) CH-3010 Bern (Switzerland)
Tel. +55 14 3235 8346, E-Mail Tel. +41 31 632 25 10
Pamela DenBesten
Department of Orofacial Sciences, School of Dentistry Ana Carolina Magalhães
University of California, San Francisco Department of Biological Sciences
513 Parnassus Avenue Bauru Dental School, University of São Paulo
San Francisco, CA 94143 (USA) Al. Dr. Octávio Pinheiro Brisolla 9-75
Tel. +1 415 502 7828 17012-190 Bauru-SP (Brazil)
E-Mail Tel. +55 14 3235 8247, E-Mail

Heitor Marques Honório Juliano Pelim Pessan

Department of Pediatric Dentistry Department of Pediatric Dentistry and Public Health
Orthodontics and Public Health Araçatuba Dental School, São Paulo State University
Bauru Dental School, University of São Paulo Rua José Bonifácio, 1193
Al. Octávio Pinheiro Brisolla, 9-75 16015-050 Araçatuba - SP (Brazil)
17012-101 Bauru-SP (Brazil) Tel. +55 18 3636 3274, E-Mail
Tel. +55 14 3235 8256, E-Mail
Daniela Rios
Steven Marc Levy Department of Pediatric Dentistry
Departments of Preventive & Community Dentistry Orthodontics and Public Health
and Epidemiology, University of Iowa Bauru Dental School, University of São Paulo
College of Dentistry Al. Dr. Octávio Pinheiro Brisolla 9-75
Iowa City, IA 52242-1010 (USA) 17012-190 Bauru-SP (Brazil)
Tel. +1 319 335 7185, E-Mail Tel. +55 14 32358218, E-Mail

Wu Li Colin Robinson
Department of Orofacial Sciences, School of Dentistry Division of Oral Biology
University of California, San Francisco Leeds Dental Institute, University of Leeds
513 Parnassus Avenue Clarendon Way
San Francisco, CA 94143 (USA) LS2 9LU, Leeds (UK)
Tel. +1 415 476 1037, E-Mail Tel. +44 113 343 6159, E-Mail

Andrew John Rugg-Gunn Alberto Enrique Villa
Morven Institute of Nutrition and Food Technology
Boughmore Road INTA, University of Chile
Sidmouth Macul
Devon EX10 8SH (UK) Santiago 783–0480 (Chile)
Tel. +44 1395 578746 Tel. +56 2 2212249
E-Mail E-Mail

Fábio Correia Sampaio Gerald Lee Vogel

Federal University of Paraiba American Dental Association Foundation
Health Science Centre Paffenbarger Research Center
Department of Clinical and Social Dentistry 100 Bureau Drive Stop 8546
58051-900 João Pessoa (Brazil) National Institute of Standards and Technology
Tel. +55 83 3216 7795 Gaithersburg MD 20899-8546 (USA)
E-Mail Tel. +1 301 975 6821, E-Mail

Jacob Martien ten Cate Gary Milton Whitford

Department of Cariology Department of Oral Biology
Endodontology and Pedodontology School of Dentistry
Academic Center for Dentistry Amsterdam (ACTA) Medical College of Georgia
Gustav Mahlerlaan 3004 1120 15th Street
NL-1081 LA Amsterdam (The Netherlands) Augusta, GA 30912 (USA)
Tel. +31 20 518 8440, E-Mail Tel. +1 706 721 0388, E-Mail

Kyriacos Jack Toumba Annette Wiegand

Division of Child Dental Health Department of Preventive Dentistry
Leeds Dental Institute, University of Leeds Periodontology and Cardiology
Clarendon Way University of Zurich
LS2 9LU Leeds (UK) Plattenstrasse 11
Tel. +44 113 343 6141, E-Mail CH-8032 Zurich (Switzerland)
Tel. +41 44 6343412

X List of Contributors

Working and publishing in the field of skeletal fluoride-induced changes to the tooth tissues dur-
and dental tissues for the past 40 years, in par- ing their development. This concept, however, was
ticular on the biology of dental enamel, it became later challenged as topical exposure to fluoride in
apparent to me at a very early stage that fluoride, a the oral environment was shown to be extremely
minor tissue constituent, was an inextricably im- effective in reducing dental caries. The role of de-
portant aspect of this area of study. Indeed the velopmentally acquired fluoride in this respect re-
effects of fluoride seemed at odds with the ex- mains intriguingly open to question.
tremely small amounts present. Also, unlike oth- Such an important advantageous clinical ef-
er important minor components of the skeletal fect, together with the concomitant possibility of
and dental tissue mineral, such as carbonate and pathological change, led to a wide range of intense
magnesium, fluoride concentrations vary widely investigations. These centered on how fluoride is
and depend to a great extent on exposure to ex- obtained from the diet, how it is dealt with after
ternal sources. absorption and also its interaction with the calci-
Fluoride came to prominence by virtue of its um phosphate/apatite phase of dental tissues.
effect on skeletal tissue development, particular- The chemistry of biological calcium phosphates
ly in relation to environmental exposure. In cases is, however, very complex. As a result of this, the
of exposure to relatively high concentrations, its deposition and behavior of the highly substituted
presence during formation – as well as its direct and defect calcium hydroxyapatite crystals of the
incorporation into the skeletal mineral phase – skeletal and dental tissues has received an enor-
led to pathological changes in both skeletal and mous amount of attention. The interaction of flu-
dental tissues. Tooth enamel was found to be par- oride with this system added further to this com-
ticularly sensitive in this respect. The effects can plexity, and as a consequence studies of fluoride
be profound since pathology related to high levels and skeletal and dental mineral have generated a
of fluoride exposure involves changes in both tis- vast literature.
sue structure as well as chemistry. With the obvious potential for improving the
However, a paradox emerged from this field protective effect against dental caries, attention
of study in which it became clear that exposure was focused upon the effect of fluoride on the de-
to smaller amounts of fluoride, while often lead- veloping tooth. Focus then moved towards studies
ing to changes in the dental tissues, conferred of the role of fluoride in the complex interactions
considerable protection against the most widely between the tooth tissues and their environment
spread and costly of diseases – dental caries. The of plaque biofilm, saliva and pellicle. It is from
protection was dramatic and was first ascribed to these studies that many of the specific benefits

of the role of fluoride in caries prevention have substantially to this area in recent years, has
emerged. brought together a number of internationally
While mechanisms behind fluoride-induced known authors with an impressive series of pub-
change to skeletal and dental tissues and the way lications across the width of the fluoride research
fluoride behaves in protecting against dental car- area.
ies are much clearer than they were 40 years ago, With regard to the structure of the monograph,
the area is still very complex and the plethora of the first section deals with the availability of flu-
literature is sometimes confusing. oride and how it is dealt with by the body from
This monograph has brought together current a physiological and metabolic standpoint. This
concepts relating to fluoride and its role in relation forms the basis for and introduction to fluoride
to the prevention of dental caries. Information toxicity and the subject of fluorosis and the im-
from a large and complex field has been assem- portance of monitoring intake. For the clinician,
bled in a clear sequence and presented in a very this highlights and clarifies the advantages of fluo-
lucid fashion. Of particular note are the diagrams, ride as well as possible hazards.
which are very clear and a great help in present- The second section focuses in more detail
ing highly complex data in an easily understood on modes of fluoride application and the way in
context. which fluoride has been and is used to effect the
With this in mind, the text will be valuable for dramatic reductions in dental caries with which it
research workers or postgraduate students begin- is associated. The complex mechanisms by which
ning a career in this or allied fields, and provide fluoride exerts its effects are described with clar-
a clear up-to-date summary of current thinking ity, and the entire text is accompanied by particu-
in this area. Established researchers and teach- larly useful illustrations.
ers, whether in clinical or basic sciences, will also Whether to those new to the field or to the es-
find the monograph a valuable addition to their tablished worker, this monograph will prove to be
libraries. a most valuable resource to the field of fluoride
The value of this text stems from the contri- research.
butions of distinguished researchers in this field. Colin Robinson, Leeds
The editor, whose own laboratory has contributed

XII Foreword
Fluoride Intake, Metabolism and Toxicity
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 1–19

Fluoride Intake of Children: Considerations

for Dental Caries and Dental Fluorosis
Marília Afonso Rabelo Buzalafa ⭈ Steven Marc Levyb
aDepartment of Biological Sciences, Bauru Dental School, University of São Paulo, Bauru, Brazil;
bDepartments of Preventive and Community Dentistry and Epidemiology, University of Iowa, Iowa City, Iowa, USA

Abstract and severity of caries among children. In the

Caries incidence and prevalence have decreased signifi- 1980s, a paradigm shift regarding the cariostatic
cantly over the last few decades due to the widespread mechanisms of fluorides was proposed [1]. This
use of fluoride. However, an increase in the prevalence considered that the predominant, if not entire,
of dental fluorosis has been reported simultaneously in explanation of how fluorides control caries de-
both fluoridated and non-fluoridated communities. Den- velopment is their topical effect on the de- and
tal fluorosis occurs due to excessive fluoride intake during re-mineralization processes that occur at the in-
the critical period of tooth development. For the perma- terface between the tooth surface and the adja-
nent maxillary central incisors, the window of maximum cent dental biofilm. This concept became wide-
susceptibility to the occurrence of fluorosis is the first 3 ly accepted [2–6], and made it possible to obtain
years of life. Thus, during this time, a close monitoring of very substantial caries protection without signifi-
fluoride intake must be accomplished in order to avoid cant ingestion of fluorides. With this in mind and
dental fluorosis. This review describes the main sources being aware of the increase in the prevalence of
of fluoride intake that have been identified: fluoridated dental fluorosis in both fluoridated and in non-
drinking water, fluoride toothpaste, dietary fluoride sup- fluoridated areas [7–9], researchers around the
plements and infant formulas. Recommendations on how world turned their attention toward controlling
to avoid excessive fluoride intake from these sources are the amount of fluoride intake.
also given. Copyright © 2011 S. Karger AG, Basel The most important risk factor for fluorosis is
the total amount of fluoride consumed from all
Fluorides play a key role in the prevention and sources during the critical period of tooth devel-
control of dental caries. In the middle of the pre- opment. Thus, it is important not only to know
vious century, it was generally believed that flu- the main sources of fluoride intake, but also the
oride had to be incorporated into dental enamel critical periods of formation in which the teeth
during development to exert its maximum pro- are more susceptible to the effects of fluoride and
tective effect. It was then considered unavoidable the levels of fluoride intake above which dental
to have a certain prevalence and severity of fluo- fluorosis is expected to occur. The purpose of this
rosis in a population to minimize the prevalence review is to discuss the levels of fluoride intake
that have been accepted as ‘optimal’ and the win- Data from a recent cohort study (Iowa Fluoride
dow of maximum susceptibility to the occurrence Study) on longitudinal fluoride intake for children
of dental fluorosis (focusing on the permanent free of fluorosis in the early-erupting permanent
maxillary central incisors), as well as to summa- dentition and free of dental caries in both the pri-
rize the recent literature on risk factors for den- mary and early-erupting permanent teeth were
tal fluorosis, and describe the multiple sources of compiled in an attempt to add scientific evidence
fluoride intake identified thus far and measures to the ‘optimal fluoride intake’ [32]. The estimated
that should be adopted to reduce fluoride intake mean daily fluoride intake for those children with
from these sources. All this information is of fun- no caries history and no fluorosis at age 9 years
damental interest to clinicians who deal with chil- was at or below 0.05 mg/kg during different pe-
dren, in order that adequate counseling regarding riods of the first 48 months of life, and this level
fluoride intake can be provided to their parents. declined thereafter. Children with caries generally
had slightly lower intakes, whereas those with flu-
orosis had slightly higher intakes. Despite this be-
‘Optimal’ Fluoride Intake ing the only recent outcome-based assessment of
‘optimal’ fluoride intake, the overlap among car-
The widely accepted ‘optimal’ intake of fluoride (be- ies/fluorosis groups in mean fluoride intake and
tween 0.05 and 0.07 mg/kg) has been empirically the high variability in individual fluoride intakes
established [10]. Its origin is attributed to McClure for those with no fluoride or caries history dis-
[11], who in the 1940s estimated that the ‘average courage the strict recommendation of an ‘optimal’
daily diet’ contained 1.0–1.5 mg fluoride, which fluoride intake. When it is necessary to employ
would provide about 0.05 mg/kg for children aged parameters of ‘optimal’ fluoride intake, the range
1–12 years. Later on this information was interpret- of 0.05–0.07 mg/kg should still be used.
ed as a recommendation when Farkas and Farkas
[12] cited various sources that suggested 0.06 mg/
kg fluoride was ‘generally regarded as optimum’. In Window of Maximum Susceptibility to the
the 1980s, this range of estimates started being used Development of Fluorosis
as a recommendation for ‘optimal’ fluoride intake
[13]. However, it is not clear if this level of intake is Considering that fluorotic changes in teeth cannot
‘optimal’ for caries prevention, for fluorosis preven- be reversed but may easily be prevented by con-
tion or a combination of both. It should also be not- trolling fluoride intake during the critical period
ed that some authors regard 0.1 mg/kg per day to of tooth formation, the identification of periods
be the exposure level above which fluorosis occurs during which fluoride intake most strongly results
[14], although others have found dental fluorosis in enamel fluorosis assumes great importance.
with a daily fluoride intake of less than 0.03 mg/ For the whole permanent dentition (excluding
kg per day [15]. It is worth mentioning that other the third molars), the age for possible fluorosis de-
factors may increase the susceptibility of individu- velopment has been considered to be the first 6–8
als to dental fluorosis, including residence at high years of life [33, 34]. However, most of the studies
altitude [16–24], renal insufficiency [25–28], mal- concerning the window of maximum susceptibili-
nutrition [22, 29] and genetics [22, 30, 31]. Some ty to dental fluorosis development have focused on
of these factors can produce enamel changes that the permanent maxillary central incisors, which
resemble dental fluorosis in the absence of signifi- are of the greatest cosmetic importance. While
cant exposure to fluoride (for details, see Buzalaf there is general consensus that the early matura-
and Whitford, this vol., pp. 20–36). tion stage of enamel development is more critical

2 Buzalaf · Levy
Table 1. Window of maximum susceptibility to the development of dental fluorosis in the permanent maxillary
central incisors

Type of study n Window of maximum susceptibility Fluoride source References

1 86 6–23 months toothpaste, Holm and Andersson [40], 1982


2 16 35–42 months water Ishii and Suckling [51], 1986

1 139 first 2 years toothpaste Osuji et al. [41], 1988

2 1,062 22–26 months water Evans and Stamm [38], 1991

2 1,085 15–24 months (males) water Evans and Darvel [50], 1995
21–30 months (females)

1 113 first 2 years toothpaste Lalumandier and Rozier [42], 1995

1 48 first year water Ismail and Messer [43], 1996

1 383 0–20 months toothpaste, Wang et al. [44], 1997


1 66 first 2 years water, Bårdsen and Bjorvatn [45], 1998


1 and 2a n.a. first 2 years (but duration of variable Bårdsen [52], 1999
exposure more important)

2 1,896 first 3 years water Burt et al. [48], 2000

Burt et al. [49], 2003

1b 579 first 2 years total intake Hong et al. [46], 2006

1b 628 first 3 years total intake Hong et al. [47], 2006

Study type 1 = Individuals introduced to fluoride at different ages; study type 2 = populations exposed from birth
that experienced an abrupt reduction in intake.
Longitudinal design.

for fluorosis than the secretory stage [15, 35–39], been exposed from birth and then had an abrupt
the evidence is not completely conclusive regard- reduction in daily fluoride intake [38, 48–51].
ing the age at which maxillary central incisors are Most of these were cross-sectional, retrospective
most susceptible to dental fluorosis. The results of and focused on just one or two sources of fluoride
studies focused on this topic are summarized in intake. Only one more recent study used longitu-
table 1. They can be divided into two categories: dinal data on individual fluoride intake [46, 47].
studies involving subjects whose exposure to fluo- While one study reported that the first year of life
ride started at different ages during tooth forma- was the most critical period for developing fluo-
tion [40–47] and those involving subjects that had rosis in the permanent central maxillary incisors

Fluoride Intake of Children 3

[43], three studies found the first 3 years critical sources, as well as the fluoride intake from all
[47–49] and another recognized a later period sources – especially in children [53–82]. Case-
(between 35 and 42 months) [51] – most of the control studies, cohort studies and randomized
studies agreed that the first 2 years of life are the clinical trials whose results were compiled in sys-
most important [40–42, 44, 45] which was also tematic reviews with or without meta-analysis
the conclusion of a meta-analysis [52]. However, led to the identification of 4 major risk factors
this meta-analysis acknowledged that the dura- for dental fluorosis: fluoridated drinking wa-
tion of fluoride exposure during amelogenesis, ter [83–85], fluoride supplements [86], fluoride
rather than specific risk periods, would seem to toothpaste [87] and infant formulas [84]. Some
explain the development of dental fluorosis in the manufactured infant foods and drinks may also
maxillary permanent central incisors, i.e. long pe- be important contributors to the total daily fluo-
riods of fluoride exposure (>2 out of the first 4 ride intake [72–75, 78, 88]. These major sources
years) led to an odds ratio (OR) of 5.8 (95% CI of fluoride intake, as well as recommendations on
2.8–11.9) versus shorter periods of exposure (<2 how to reduce fluoride intake from them, will be
out of the first 4 years of life). This is in line with discussed in detail below.
data from a more recent longitudinal study which
concluded that: (1) although the first 2 years of life Fluoridated Drinking Water
were generally found to be more important com- Fluoridation of community drinking water is rec-
pared with later years, fluoride intake during each ognized among the top 10 greatest public health
individual year (until the fourth year of life) was achievements in the world in the last century [89].
associated with fluorosis; (2) subjects with higher Although other fluoride-containing products are
levels of fluoride intake (estimated mean daily in- available, water fluoridation remains the most
gestion of 0.059 mg/kg) during the whole first 3 equitable and cost-effective method of delivering
years of life had the highest risk of fluorosis [46]. fluoride to all members of most communities, re-
Thus, the development of fluorosis appears to be gardless of age, income level or educational attain-
related not only to the timing of fluoride intake ment. Additionally, there is some evidence that
relative to the periods of tooth formation, but also water fluoridation may reduce the oral health gap
to the cumulative duration of fluoride exposure between social classes [85]. The mean estimated
[46, 52]. costs for water fluoridation are only about USD
From the available evidence, it seems rational 0.72/year per person in the USA [90].
to monitor fluoride intake of children in the first 3 Early in the 1940s, it was known that about
years of life in order to minimize the risk of devel- 10% of children in areas naturally fluoridated at
oping dental fluorosis of the permanent maxillary optimum levels (1.0 ppm) were affected by mild
central incisors, which are the most relevant teeth or very mild fluorosis of the permanent teeth,
from an aesthetic point of view [46, 47, 52]. and this rate was less than 1% in low-fluoride ar-
eas [91]. These levels of prevalence were record-
ed when fluoridated drinking water was the only
Sources of Fluoride Intake significant source of fluoride intake, before the
widespread distribution of packaged beverages
Concern with the increase in the prevalence of or the availability of fluoridated dental prod-
mostly mild but also some moderate-to-severe ucts. Studies conducted in the 1980s and 1990s
dental fluorosis and its potential impact on qual- reported that the prevalence of dental fluorosis
ity of life has led investigators all over the world in areas where the water fluoride concentration
to estimate the fluoride concentration of potential was 0.8 ppm was 4 times as high as that found

4 Buzalaf · Levy
in non-fluoridated communities [92–94]. In a the only significant source of fluoride was the wa-
systematic review of 214 studies on water fluo- ter supply.
ridation published in 2000, 88 studies on dental The increased prevalence of dental fluorosis
fluorosis were included [83]. The authors found found more recently indicates that some young
a significant dose-response association between children are ingesting fluoride from sources oth-
the fluoride concentration in the drinking water er than drinking water. One study estimated that
and the prevalence of dental fluorosis. It was es- approximately 2% of US schoolchildren would
timated that, at a fluoride level of 1 ppm in the experience perceived aesthetic problems which
drinking water, the prevalence of any dental fluo- could be attributed to the currently recommend-
rosis was 48%, and 12.5% of exposed people had ed levels of fluoride in drinking water [96]. The
dental fluorosis that they would find of aesthet- US Department of Health and Human Services
ic concern (moderate to severe). This is much has recently proposed a new recommendation on
higher than that reported by Dean et al. [91] in water fluoride levels that is 0.7 ppm fluoride for
1942, who found virtually no cases of moderate the entire nation [97] and replaces the 1962 US
or severe fluorosis, although the results are not Public Health Service Drinking Water Standards
directly comparable since different case defini- which were based on ambient air temperature of
tions were used. geographic areas and ranged from 0.7 to 1.2 ppm
The studies that took advantage of breaks in fluoride. This guidance is based on several con-
water fluoridation to assess dental fluorosis of dif- siderations that include: (1) scientific evidence re-
ferent birth cohorts are of special interest when lated to effectiveness of water fluoridation on car-
analyzing the impact of fluoridated water on the ies prevention and control across all age groups;
prevalence of dental fluorosis in a community. In (2) fluoride in drinking water as one of several
this way, Burt et al. [48] evaluated the impact of available fluoride sources; (3) trends in the preva-
an unplanned break of 11 months in water flu- lence and severity of dental fluorosis; (4) current
oridation, and concluded that the prevalence of evidence that fluid intake in children does not
dental fluorosis is affected by changes in fluo- increase with increases in ambient air tempera-
ride exposure from drinking water. However, in ture due to augmented use of air conditioning and
a subsequent study [49], the prevalence of den- more sedentary lifestyles [98].
tal fluorosis remained stable, in spite of an ex- A recent study estimated the total daily fluo-
pected increase in the next cohort due to the re- ride intake from different constituents of the diet
establishment of water fluoridation. Buzalaf et al. and from dentifrice by 1- to 3-year-old children
[95] analyzed the effect of a 7-year interruption living in an optimally fluoridated area. Standard
in water fluoridation on the prevalence of den- fluoride concentration dentifrice alone was re-
tal fluorosis in a Brazilian city. The authors found sponsible for, on average, 81.5% of the daily fluo-
a lower prevalence of dental fluorosis in the per- ride intake, while among the constituents of the
manent maxillary central incisors of children who diet, water and reconstituted milk were the most
were 36, 27 and 18 months old when water flu- important contributors and were responsible for
oridation ceased when compared with children about 60% of the total contribution of the diet
who were born 18 months after fluoridation was [68]. For 4- to 6-year-old children living in the
interrupted. When analyzed together, the results same community, however, the impact of fluoride
of these studies conducted in the 2000s suggest ingested from dentifrice was less, and water alone
that the relative importance of fluoridated water provided a mean of 34% of the estimated daily flu-
on the prevalence of dental fluorosis in current oride from the diet, which corresponds to about
populations might not be as great as it was when 0.014 mg/kg [69, 70]. Thus, since fluoride present

Fluoride Intake of Children 5

in water contributes only a small portion of intake Fluoride Toothpaste
from the dietary constituents, fluoridated water For several decades, fluoridated water was recog-
probably has its greatest impact on dental fluo- nized as the major risk factor for dental fluoro-
rosis prevalence indirectly, through being used in sis as a result of the classic studies by Dean et al.
the reconstitution of infant formulas and in the [91]. Observations that the prevalence of dental
processing of other children’s foods and beverages fluorosis had increased more in non-fluoridated
[10]. Taking into account the low risks and great than in fluoridated areas [109] resulted in efforts
benefits of public water fluoridation, as well as the to better understand the relative impact of oth-
levels of prevalence and especially the severity of er potential sources of fluoride ingestion on the
dental fluorosis found today, this measure must prevalence of dental fluorosis. Among them, flu-
be maintained in the areas where it already exists oride toothpastes were identified as a potential
and extended to the areas where it is feasible to risk factor for dental fluorosis, since an inverse
implement water fluoridation. relationship can be observed between the age of
In order to minimize the possible impact of the child and the mean percentage ingestion of
water fluoridation on dental fluorosis, some mea- toothpaste [110]. A recent review compiled data
sures should be taken. One of them is external for the estimated total fluoride intake of children
monitoring of water fluoridation by an indepen- living in different locations [111]. It was noted
dent assessor. This measure has been shown to be that toothpaste was usually the main contributor
successful in improving the consistency of fluo- for young children. Thus, toothpaste is an impor-
ridation [99] and ideally should be implement- tant source of fluoride during the critical period
ed wherever there is adjusted fluoridation, but of tooth development.
at least in the communities where fluctuations in Table 2 summarizes the main findings of
water fluoride levels commonly occur [100]. cross-sectional, case-control, cohort studies
It is also important to advise that, for infants and randomized clinical trials conducted in dif-
and small children receiving large quantities of ferent countries, both in fluoridated and non-
reconstituted infant formula, water containing fluoridated communities, which investigated the
<0.5 ppm fluoride should be used. A recent meta- association between the use of fluoride tooth-
analysis found that a 1.0-ppm increase in the flu- paste and the prevalence or severity of dental flu-
oride level in the water supply is associated with orosis. A positive association was found in most
a 67% increase in the OR for dental fluorosis as- of these studies, mainly related to the early use of
sociated with infant formula [84]. Thus, bottled fluoride toothpaste (before age 24 months), re-
water with relatively low fluoride content can be gardless of the community fluoridation status.
used instead of fluoridated water from the pub- This issue was addressed in a recent systematic
lic supply [73, 77, 101]. Many brands of bottled review and meta-analysis [87] compiling the re-
water commercially available have low fluoride sults of 25 studies published between 1988 and
content and should be adequate for this purpose 2006 that investigated the relationship between
[53, 102–108]. However, one difficulty is that in the use of fluoride toothpastes and dental fluoro-
many cases fluoride concentrations are not stat- sis. Among these, two RCTs [112, 113], one co-
ed or are stated inaccurately on the labels, and hort study [114], six case-control studies [36, 41,
unexpectedly high fluoride concentrations can 115–118] and sixteen cross-sectional surveys [44,
be found [102, 106]. This reinforces the need for 92, 93, 119–131] were included. Among the 25
global labeling of fluoride levels in bottled water studies included, only one RCT was considered
and rigorous surveillance by the competent pub- at low risk of bias [112]. The main findings of this
lic health authorities. systematic review with meta-analysis were: (1) a

6 Buzalaf · Levy
Table 2. Studies assessing the association between the use of fluoride toothpaste and dental fluorosis

Study n Age, Country Fluoride Other risk factors Main outcome related to References
design years water or fluoride dentifrice

Case- 633 8–10 Canada yes infant formula OR = 11.0 (brushing before 25 Osuji et al.
control months) [41], 1988

Cross- 556 6–12 USA varied water, rinses no association Szpunar and
sectional Burt [92],

Case- 850 11–14 USA no supplements, family OR = 2.9 Pendrys and

control income, infant formula Katz [36],

Cross- 350 7.5 Australia varied water, weaning age OR = 2.6 Riordan [93],
sectional 1993

Case- 401 12–16 USA yes supplements, infant OR = 2.80 Pendrys et al.
control formula (frequent brushing) [116], 1994

RCT 1,523 9–10 Norway varied supplements TF lower for children using Holt et al.
550-ppm fluoride dentifrice [113], 1994

Case- 157 8–17 USA varied water higher risk of fluorosis in Skotowski et
control children who used larger al. [118], 1995
amounts of dentifrice

Case- 708 5–19 USA varied supplements OR = 3.0 Lalumandier

control (age when started brushing) and Rozier
[42], 1995

Case- 460 10–13 USA no supplements OR = 2.5 (early dentifrice use) Pendrys et al.
control [117], 1996

Cross- 383 8 Norway no supplements use of dentifrice before 14 Wang et al.

sectional months increased prevalence [44], 1997
of fluorosis

Cross- 325 8–9 UK yes not evaluated fluoride ingestion from Rock and
sectional dentifrice associated with Sabieha [128],
fluorosis 1997

Cross- 197 1–7 USA yes supplement use from no association Morgan et al.
sectional ages 0 to 3 years [131], 1998

Cross- 1,189 12 India no not evaluated OR = 1.83 (use of fluoride Mascarenhas

sectional dentifrice before age 6 years); and Burt
beginning brushing before [124], 1998
age 2 years increased severity
of fluorosis

Case- 233 10–14 USA yes supplements (OR = 6.0 OR = 6.4 and 8.4 for early-and Pendrys and
control and 10.8 for early- and later-forming enamel surfaces, Katz [115],
later-forming enamel respectively (early use of 1998
surfaces, respectively); fluoride dentifrice)
powdered formula (OR =
10.7 for later-forming
enamel surfaces)

Fluoride Intake of Children 7

Table 2. Continued

Study n Age, Country Fluoride Other risk factors Main outcome related to References
design years water or fluoride dentifrice
Cross- 752 7–8 Canada no formula feeding, no association Brothwell and
sectional supplements Limeback
[121], 1999

Cross- 3,500 7–14 USA varied supplements, brushing before age 2 years Kumar and
sectional continuous exposure to increased risk of fluorosis Swango
fluoride water [166], 1999

Cross- 314 11–12 Brazil no no association OR = 4.4 (brushing before age Pereira et al.
sectional 3 years) [126], 2000

Cross- 763 10–14 USA varied non-fluoridated area: early toothbrushing behaviors Pendrys
sectional supplements; fluoridated regardless of water fluoride [145], 2000
area: supplements and levels
powdered infant formula

Cross- 867 8–9 UK varied water use of adult dentifrice Tabari et al.
sectional [130], 2000

Cross- 582 10 Australia varied fluorosis prevalence fluorosis prevalence declined Riordan [127],
sectional declined after reduction after use of low-fluoride 2002
in use of supplements dentifrices increased

Cross- 8,277 not Canada varied supplements; high beginning brushing between Maupomé et
sectional informed parental educational 1st and 2nd birthdays al. [125], 2003
level increased fluorosis (vs.
between 2nd and 3rd

Cross- 4,128 11 Belgium no supplements: ever vs. toothbrushing frequency: ≥2/ Bottenberg et
sectional never (OR = 1.31), taken day vs. <2/day (OR = 1.4) al. [120], 2004
not in milk vs. in milk (OR
= 1.69);
water fluoride

RCT 703 8–9 UK no not evaluated all subjects identified with TF = Tavener et al.
3 were found in the 1,450-ppm [167], 2004
fluoride dentifrice group

Cross- 320 6–9 Mexico yes main source of fluoride effect of supplementary Beltran-
sectional exposure: professionally sources different between Valladares et
vs. self-applied (OR = children brushing before 2 al. [119], 2005
2.13) years (OR = 6.15) and after (OR
= 2.14)

Cross- 548 7–9 Sweden no no association no association Conway et al.

sectional [122], 2005

8 Buzalaf · Levy
Table 2. Continued

Study n Age, Country Fluoride Other risk factors Main outcome related to References
design years water or fluoride dentifrice
Cohort 343 7–11 USA varied ingestion from significant association Franzman et
beverages, between fluorosis and al. [114], 2006
selected foods and toothpaste ingestion at age 24
fluoride supplements at months
ages 16 months, 36
and AUC ages 16–36

RCT 1,268 8–10 UK no prevalence of fluorosis prevalence of TF ≥2 and ≥3 Tavener et al.

significantly higher in significantly higher for groups [112], 2006
less-deprived districts receiving 1,450-ppm fluoride
dentifrice vs. 440 ppm

Cross- 1,373 6–12 Mexico yes salt fluoridation toothbrushing frequency Vallejos-
sectional associated with fluorosis (OR = Sánchez et al.
1.63) [168], 2006

Cross- 699 12 Ireland yes no association no association Sagheri et al.

sectional and [129], 2007

Cross- 677 9–13 Australia varied fluoridated water use of standard-concentration Do and
sectional fluoridated dentifrice; eating/ Spencer
licking toothpaste were risk [123], 2007
factors for fluorosis

Case- 2,106 13 Norway no supplements: regular use no children who had Pendrys et al.
control and mild-to-moderate exclusively used only a pea- [169], 2010
fluorosis (OR = 6.5) sized amount of toothpaste
(1,000 ppm fluoride) had mild-
to-moderate fluorosis

TF index = Thylstrup and Fejerskov index.

significant reduction in the risk of dental fluorosis was found between frequency of toothbrushing
was found if toothbrushing with fluoride tooth- or amount of toothpaste used and fluorosis (data
paste did not start until the age of 12 months, but from cross-sectional surveys); (3) using tooth-
the evidence for starting toothbrushing with flu- paste with a higher concentration of fluoride in-
oride toothpaste before the age of 24 months was creased the risk of dental fluorosis (data from two
inconsistent (data from case-control and cross- RCTs; evidence from cross-sectional studies was
sectional studies); (2) no significant association inconsistent). From the available evidence, the

Fluoride Intake of Children 9

authors concluded that decisions involving the it seems reasonable to recommend low-fluoride
use of topical fluorides (including toothpastes) (500 ppm) toothpastes for young children who
should balance their benefits in caries preven- are at risk of developing dental fluorosis in the
tion and the risk of causing dental fluorosis. They permanent maxillary central incisors (<3 years of
noted: ‘if the risk of fluorosis is of concern, the age) but have low caries risk, especially if they live
fluoride level of toothpaste for young children is in a fluoridated area. In all other cases, toothpastes
recommended to be lower than 1,000 ppm’ [87]. containing at least 1,000 ppm fluoride should be
Risk-benefit considerations are critical. A recent used. Although to date there is not unequivocal
systematic review and meta-analysis of 83 inde- evidence supporting the association between the
pendent trials concluded that only toothpastes amount of toothpaste used and dental fluorosis
containing ≥1,000 ppm fluoride have been prov- [87], it seems rational to recommend the use of
en to be beneficial for preventing caries in chil- a small amount of toothpaste by young children,
dren and adolescents [132]. However, for the which can be easily achieved using the ‘transverse’
deciduous dentition (age related with the devel- [136] or ‘drop’ [137] techniques. It is equally im-
opment of dental fluorosis), uncertainty regard- portant that young children brush under adult
ing the effectiveness of low-fluoride toothpastes supervision and be instructed to expectorate the
for preventing caries was reported due to the lack foam after toothbrushing as much as possible.
of trials [132]. An alternative to improve the anti-
caries effectiveness of low-fluoride dentifrices Dietary Fluoride Supplements
might be pH reduction, since lowering the pH Table 2, which describes the studies that investi-
enhances the tendency for calcium fluoride for- gated the association between the use of fluoride
mation on enamel [133]. A recent RCT evaluated toothpaste and the occurrence of dental fluorosis,
the caries increment in high caries risk 4-year-old also shows that the most cited risk factor for den-
children living in a fluoridated area with the use tal fluorosis besides fluoride toothpaste is fluoride
of a low-fluoride (550 ppm) acidic (pH 4.5) liq- supplements.
uid dentifrice. It was observed that the caries pro- Dietary fluoride supplements were originally
gression rate was similar to that found with the designed to help prevent dental caries in children
use of a conventional 1,100-ppm fluoride tooth- living in fluoride-deficient areas. The recom-
paste [134]. Also, the long-term use of this acid- mended daily dose was based on the age of the
ic dentifrice was shown to result in lower finger- child and fluoride concentration in the drink-
nail fluoride concentrations of the children using ing water. In 1999, a systematic review of stud-
this product compared to the control toothpaste ies evaluating the association between the use of
[135], which supports lower fluoride intake. The fluoride supplements by children living in non-
tested formulation could be an alternative to stan- fluoridated areas and dental fluorosis was car-
dard fluoride concentration toothpaste in order ried out [138]. By conducting a Medline search
to avoid dental fluorosis in young children, but between 1966 and 1997, the authors were able to
additional clinical trials are necessary to provide perform a qualitative review of 10 cross-sectional/
unequivocal evidence on this matter. Also further case-control studies [36, 42, 44, 94, 117, 139–143]
work should be done in attempt to enhance the and found a strong consistent association be-
anti-caries efficacy of low-fluoride toothpastes in tween the use of fluoride supplements and dental
order to further maximize benefits and minimize fluorosis. The meta-analysis of these studies esti-
risk of accidental ingestion. mated that the OR of dental fluorosis in children
In conclusion, based on the available evidence living in non-fluoridated areas who had regularly
regarding the risks of caries and dental fluorosis, used supplements during the first 6 years of life

10 Buzalaf · Levy
when compared with non-users was about 2.5 the American Dental Association recommends:
[138]. Recently, the same group updated the for- (1) no supplements from birth to 6 months or
mer systematic review by including an addition- for residents of areas containing more than 0.6
al 4 studies in the meta-analysis [115, 120, 144, ppm fluoride in the drinking water; (2) 0.25 mg
145]. This inclusion confirmed the positive asso- fluoride/day from 6 months to 3 years for chil-
ciation between the use of supplements and the dren living in areas containing less than 0.3 ppm
occurrence of dental fluorosis. The OR for dental fluoride in the drinking water; (3) 0.50 and 0.25
fluorosis increased by 84% for each year of fluo- mg/day for children aged 3–6 years, living in ar-
ride supplement use between the ages of younger eas with less than 0.3 and 0.3–0.6 ppm fluoride
than 6 months and 7 years, but the first 3 years of in the drinking water, respectively, while double
life were considered more important [86]. It must the dose is recommended from 6 to 16 years. It
be highlighted, however, that most cases of dental should be emphasized that the American Dental
fluorosis attributable to the use of fluoride sup- Association recommends dietary fluoride supple-
plements are graded as mild, with little likelihood ments should only be used for children who are
of causing social impact, despite there being rela- at high risk of developing dental caries [148]. The
tively few studies on this latter issue [86, 146]. Canadian Dental Association recommends sup-
In the later systematic review, the authors also plements only for children who have high caries
evaluated the effectiveness of fluoride supple- experience and whose total intake of fluoride is
ments in preventing caries. They concluded that lower than 0.05–0.07 mg/kg [149]. This recom-
there is weak inconsistent evidence showing that mendation, however, is not practical because es-
fluoride supplements are effective at preventing timating the total fluoride intake from all sources
caries in primary dentition. However, they are is very difficult. A more practical view, recom-
able to help prevent caries in the permanent teeth mended by a group of European experts in 1991,
of school-aged children (>6 years) when used on states that ‘a dose of 0.5 mg/day fluoride should
a regular basis – primarily due to a topical effect be prescribed for at-risk individuals from the age
[86]. of 3 years’ [150].
From the available evidence regarding the as- Considering the available evidence indicating
sociations of supplements with dental caries and that fluoride supplements only help prevent caries
dental fluorosis, it is clear that consideration of when regularly used by children older than 6 years
the risk-benefit ratio is necessary when prescrib- of age, and that their use before this age (but es-
ing supplements, as discussed earlier for fluoride pecially during the first 3 years) is associated with
toothpastes. There is consensus that fluoride sup- dental fluorosis [86], the view of the Europeans
plements should not be prescribed in optimally seems to be the most rational one. However, for
fluoridated areas, for infants less than 6 months of remote/special populations not receiving other
age, nor for children who are at low risk of devel- fluoride and caries prevention measures, fluoride
oping dental caries. Different policies, however, supplementation may also be appropriate.
have been adopted by distinct countries and den-
tal associations regarding the recommendations Infant Formulas
for the appropriate use of supplements to prevent Despite breastfeeding being recommended in
caries. The conclusion of a workshop conducted health campaigns worldwide, in many cases it is
in Australia in 2006 on the use of fluorides for car- impractical. Additionally, as infants are weaned
ies prevention was that ‘fluoride supplements in from breast milk, most of them receive the
the form of drops or tablets to be chewed and/or majority of their nutrition from infant formula,
swallowed should not be used’ [147]. In the USA, especially in the first 4–6 months of life before

Fluoride Intake of Children 11

they start receiving solid foods. Commercially 1.0 ppm fluoridated water. Thus, it has been sug-
prepared infant formulas are available as gested that the intake of fluoride by infants from
powder and liquid concentrates that have to be formulas is influenced more by the water used
diluted with water before use, or as ready-to- to reconstitute the formula than by the formulas
feed formulations. While human milk [151] and themselves [53, 73, 101, 154, 156].
cow’s milk [152] have low fluoride concentra- Soy-based infant formulas have been reported
tions (typically <0.01 and <0.10 ppm, respec- to have somewhat higher fluoride concentrations
tively), this is not true for infant formulas that than milk-based ones [77, 101, 153, 154, 156]. A
can have a high intrinsic fluoride content due study conducted with Malaysian soy-based for-
to manufacturing procedures or an increased mulas found values ranging from 0.24 to 0.44 ppm
fluoride content due to the use of fluoridated when prepared with deionized water and from
water for reconstitution of powders or liquid 0.45 to 0.47 ppm when prepared with fluoridated
concentrates [77]. water (0.38 ppm) [153]. It has been reported that
Fluoride concentrations in infant formulas substantial consumption of some fluoride-rich
show wide variations when assessed in different soy-based infant formulas, even when reconsti-
countries. For infant powdered formulas market- tuted with deionized water, would provide a flu-
ed in Brazil, fluoride concentrations ranged from oride intake above the upper tolerable limit for
0.01 to 0.75 ppm when reconstituted with deion- 1-month-old children [77, 101, 156].
ized water, from 0.91 to 1.65 ppm when reconsti- Considering the fluoride concentrations pres-
tuted with fluoridated drinking water (containing ent in infant formulas themselves, as well as the
0.9 ppm), and from 0.02 to 1.37 ppm when recon- concentrations of fluoride present in the water
stituted with different brands of bottled mineral used to reconstitute them, the above-mentioned
water [101]. In Australia, fluoride concentrations studies have considered infant formula consump-
ranging from 0.03 to 0.53 ppm for powdered for- tion a potential risk factor for dental fluorosis. A
mulas prepared with non-fluoridated water were recent systematic review attempted to clarify the
reported [77]. In Thailand and Japan, values rang- association between use of infant formula from
ing from 0.14 to 0.64 and 0.37 to 1.00 ppm, re- birth to age 24 months and dental fluorosis [84].
spectively, were found [60]. In Malaysia, fluoride The authors compiled the results of 19 studies in-
concentrations ranging from 0.10 to 0.16 ppm cluding 17,429 subjects with ages ranging from 2
when prepared with deionized water, and from to 17 years. Among these studies, one was a pro-
0.35 to 0.40 ppm when prepared with water con- spective cohort study [157], five were retrospec-
taining 0.38 ppm, were observed for infant for- tive cohort studies [14, 21, 93, 143, 158], six were
mulas [153]. In the USA, fluoride concentrations case-control studies [36, 41, 115–117, 159], four
of ready-to-feed, concentrated liquid and pow- were cross-sectional studies [121, 160–162] and
dered formulas prepared with deionized water three were historical-control studies [48, 123,
were reported to be around 0.15, 0.12–0.27 and 163]. The summary OR from 17 studies relating
0.13 ppm, respectively [53, 154], and these fluo- infant formula use to dental fluorosis in the per-
ride levels would result in an intake well below manent dentition was 1.8 (95% CI 1.4–2.3), but
the upper limit of 0.10 mg/day established by the there was significant heterogeneity in the magni-
Institute of Medicine (Washington, D.C., USA) tude of the OR among the studies, indicating that
under normal consumption of formulas [154]. the summary OR must be interpreted with cau-
However, most infants are likely to exceed the tion. A meta-regression provided weak evidence
upper tolerable limit if they are exclusively fed that the fluoride in the infant formula resulted in
powdered infant formula reconstituted with 0.7– an increased risk of developing dental fluorosis.

12 Buzalaf · Levy
However, the dental fluorosis risk associated with fluoride concentrations of infant foods and bev-
the use of infant formula depended on the level erages span a wide range and depend mainly on
of fluoride in the water supply. An increase in the the fluoride concentration in the water used to
dental fluorosis OR of 5% was seen as the fluo- manufacture them [165].
ride level of the water supply increased by 0.1 ppm Beikost is a collective term for foods other than
(OR 1.05, 95% CI 1.02–1.09), such that a 1.0-ppm milk or formula fed to infants. A wide variation
increase in the fluoride concentration in the wa- (between 0.01 and 8.30 ppm) has been reported re-
ter supply is associated with a 67% increased OR garding the fluoride content of beikost. Chicken-
for dental fluorosis associated with infant formula based products usually present the highest values
(OR 1.67, 95% CI 1.18–2.36). The authors were due to the inclusion of bones in the manufactur-
not able to determine, however, whether liquid or ing process. In some studies, fish-based products
powder infant formulas with or without reconsti- also have been reported to have high fluoride
tution affected fluorosis risk differently, since only content. In general, the fluoride concentrations of
a few studies provided this information [84]. A re- most beikost is usually low [165]. However, some
cent cohort study (Iowa Fluoride Study) reported cereals commonly added to milk that are usually
that greater fluoride intakes from reconstituted consumed by infants in Brazil have been shown to
infant formulas at ages 3–9 months increased risk have high fluoride concentrations. This is the case
of mild dental fluorosis of the permanent maxil- for Mucilon and Neston (Nestlé), which have flu-
lary incisors [164]. oride concentrations of 2.4 and 6.2 ppm, respec-
In summary, considering that the increased tively [72, 74]. A relatively high fluoride concen-
risk of dental fluorosis posed by the use of infant tration was also found in ready-to-drink chocolate
formulas depends mainly on the fluoride level of milk (1.2 ppm, Toddynho, Quaker) [72, 74].
water supply [84] and that the reconstitution of It must be highlighted that fluoride present
formulas with 0.7–1.0 ppm fluoride may provide in the manufactured foods and beverages comes
infants with a daily fluoride intake above that like- as a ‘contaminant’. The manufacturers are usu-
ly to cause some degree of dental fluorosis [154], ally unaware of these fluoride concentrations,
it seems reasonable to advise that, for those re- which can vary among the different batches of
ceiving large quantities of reconstituted infant products. Thus, market basket studies are im-
formula, water containing <0.5 ppm fluoride portant for determining the fluoride concen-
should be used for reconstitution. Bottled water trations in manufactured foods and beverages,
with low fluoride concentrations could be used since there are no laws that require this informa-
for this purpose [77, 101, 102]. However, fluoride tion to be stated on the products’ labels. Health
concentrations both in infant formula and bottled professionals must be updated with respect to
water must be correctly displayed on product la- the available information, in order to adequately
bels. Periodical analyses of fluoride concentra- advise the parents of children at the age of risk
tions present in infant formula and bottled water for dental fluorosis. The general recommenda-
by government or private laboratories could con- tion is that children under the age of 7 years, but
tribute to ensure that the fluoride levels are ad- mainly in the first 3 years of life, avoid substan-
equately displayed on the labels. tial consumption of products with high-fluoride
content, since they can significantly contribute
Manufactured Infant Foods and Beverages to the total daily fluoride intake and increase the
During infancy, important sources of fluoride in- risk of dental fluorosis, especially when associ-
clude selected commercially available foods and ated with other fluoride sources.
beverages. Many studies have shown that the

Fluoride Intake of Children 13

Conclusion fluoride water, optimally fluoridated water, fluo-
ride toothpaste, dietary fluoride supplements and
Improved understanding of the mechanisms of ac- infant formulas, especially if these are reconstitut-
tion of fluoride for caries prevention and control ed with fluoridated water. Maintaining appropri-
have made it possible to obtain substantial benefit ate levels of fluoride in the water from the public
from the use of fluoride with a minimum risk of supply, avoiding the ingestion of substantial quan-
side effects, since it was established that fluoride tities of optimally fluoridated water with reconsti-
does not need to be ingested to be beneficial. This tuted infant formulas, placing a small amount of
observation, concomitant with the reported in- fluoride toothpaste onto the toothbrush and su-
crease in the prevalence of dental fluorosis, turned pervising toothbrushing of pre-school children,
the attention of researchers worldwide toward the as well as not routinely prescribing fluoride sup-
necessity of controlling fluoride intake. The most plements for children at low risk of developing
important risk factor for dental fluorosis is the to- caries, those living in fluoridated areas and those
tal amount of fluoride consumed from all sources below age 3 years (regardless of the status of com-
during the critical period of tooth development munity fluoridation) are the main measures rec-
(the first 3 years of life for the permanent maxil- ommended to reduce the total fluoride intake of
lary central incisors). The main sources of fluo- young children during the period of greatest risk
ride intake which are recognized as potential risk for developing dental fluorosis.
factors for developing dental fluorosis are high

1 Fejerskov O, Thylstrup A, Larsen MJ: 7 Khan A, Moola MH, Cleaton-Jones P: 13 Ophaug RH, Singer L, Harland BF: Esti-
Rational use of fluorides in caries pre- Global trends in dental fluorosis from mated fluoride intake of average two-
vention: a concept based on possible 1980 to 2000: a systematic review. SADJ year-old children in four dietary regions
cariostatic mechanisms. Acta Odontol 2005;60:418–421. of the United States. J Dent Res
Scand 1981;39:241–249. 8 Whelton HP, Ketley CE, McSweeney F, 1980;59:777–781.
2 ten Cate JM, Duijsters PP: Influence of O’Mullane DM: A review of fluorosis in 14 Forsman B: Early supply of fluoride and
fluoride in solution on tooth demineral- the European Union: prevalence, risk enamel fluorosis. Scand J Dent Res
ization. I. Chemical data. Caries Res factors and aesthetic issues. Community 1977;85:22–30.
1983;17:193–199. Dent Oral Epidemiol 15 Baelum V, Fejerskov O, Manji F,
3 ten Cate JM, Featherstone JD: Mechanis- 2004;32(suppl 1):9–18. Larsen MJ: Daily dose of fluoride and
tic aspects of the interactions between 9 Clark DC: Trends in prevalence of dental dental fluorosis. Tandlaegebladet
fluoride and dental enamel. Crit Rev fluorosis in North America. Community 1987;91:452–456.
Oral Biol Med 1991;2:283–296. Dent Oral Epidemiol 1994;22:148–152. 16 Angmar-Månsson B, Whitford GM:
4 Øgaard B, RØlla G, Ruben J, Dijkman T, 10 Burt BA: The changing patterns of sys- Environmental and physiological factors
Arends J: Microradiographic study of temic fluoride intake. J Dent Res affecting dental fluorosis. J Dent Res
demineralization of shark enamel in a 1992;71:1228–1237. 1990;69:706–713, discussion 21.
human caries model. Scand J Dent Res 11 McClure FJ: Ingestion of fluoride and 17 Irigoyen ME, Molina N, Luengas I: Prev-
1988;96:209–211. dental caries: quantitative relations alence and severity of dental fluorosis in
5 Øgaard B, RØlla G, Dijkman T, Ruben based on food and water requirements a Mexican community with above-
J, Arends J: Effect of fluoride for children 1 to 12 years old. Am J Dis optimal fluoride concentration in drink-
mouthrinsing on caries lesion Child 1943;66:8. ing water. Community Dent Oral Epide-
development in shark enamel: an in 12 Farkas CS, Farkas EJ: Potential effect of miol 1995;23:243–245.
situ caries model study. Scand J Dent food processing on the fluoride content 18 Mabelya L, König KG, van Palenstein
Res 1991;99:372–377. of infant foods. Sci Total Environ Helderman WH: Dental fluorosis, alti-
6 Featherstone JD: The science and prac- 1974;2:399–405. tude, and associated dietary factors
tice of caries prevention. J Am Dent (short communication). Caries Res
Assoc 2000;131:887–899. 1992;26:65–67.

14 Buzalaf · Levy
19 Mabelya L, van Palenstein Helderman 29 Rugg-Gunn AJ, al-Mohammadi SM, But- 41 Osuji OO, Leake JL, Chipman ML,
WH, van’t Hof MA, Konig KG: Dental ler TJ: Effects of fluoride level in drink- Nikiforuk G, Locker D, Levine N: Risk
fluorosis and the use of a high fluoride- ing water, nutritional status, and socio- factors for dental fluorosis in a fluori-
containing trona tenderizer (magadi). economic status on the prevalence of dated community. J Dent Res
Community Dent Oral Epidemiol developmental defects of dental enamel 1988;67:1488–1492.
1997;25:170–176. in permanent teeth in Saudi 14-year-old 42 Lalumandier JA, Rozier RG: The preva-
20 Manji F, Baelum V, Fejerskov O, boys. Caries Res 1997;31:259–267. lence and risk factors of fluorosis among
Gemert W: Enamel changes in two 30 Everett ET, McHenry MA, Reynolds N, patients in a pediatric dental practice.
low-fluoride areas of Kenya. Caries Res Eggertsson H, Sullivan J, Kantmann Pediatr Dent 1995;17:19–25.
1986;20:371–380. C, Martinez-Mier EA, Warrick JM, 43 Ismail AI, Messer JG: The risk of
21 Rwenyonyi C, Bjorvatn K, Birkeland J, Stookey GK: Dental fluorosis: variability fluorosis in students exposed to a higher
Haugejorden O: Altitude as a risk among different inbred mouse strains. than optimal concentration of fluoride
indicator of dental fluorosis in children J Dent Res 2002;81:794–798. in well water. J Public Health Dent
residing in areas with 0.5 and 2.5 mg 31 Huang H, Ba Y, Cui L, et al.: COL1A2 1996;56:22–27.
fluoride per litre in drinking water. gene polymorphisms (Pvu II and Rsa I), 44 Wang NJ, Gropen AM, Øgaard B: Risk
Caries Res 1999;33:267–274. serum calciotropic hormone levels, and factors associated with fluorosis in a
22 Yoder KM, Mabelya L, Robison VA, dental fluorosis. Community Dent Oral non-fluoridated population in Norway.
Dunipace AJ, Brizendine EJ, Stookey GK: Epidemiol 2008;36:517–522. Community Dent Oral Epidemiol
Severe dental fluorosis in a Tanzanian 32 Warren JJ, Levy SM, Broffitt B, 1997;25:396–401.
population consuming water with Cavanaugh JE, Kanellis MJ, Weber- 45 Bårdsen A, Bjorvatn K: Risk periods in
negligible fluoride concentration. Gasparoni K: Considerations on optimal the development of dental fluorosis. Clin
Community Dent Oral Epidemiol fluoride intake using dental fluorosis Oral Investig 1998;2:155–160.
1998;26:382–393. and dental caries outcomes – a longitu- 46 Hong L, Levy SM, Broffitt B, Warren JJ,
23 Akosu TJ, Zoakah AI: Risk factors dinal study. J Public Health Dent Kanellis MJ, Wefel JS, Dawson D: Timing
associated with dental fluorosis in 2009;69:111–115. of fluoride intake in relation to develop-
Central Plateau State, Nigeria. 33 Pendrys DG: The fluorosis risk index: a ment of fluorosis on maxillary central
Community Dent Oral Epidemiol method for investigating risk factors. incisors. Community Dent Oral Epide-
2008;36:144–148. J Public Health Dent 1990;50:291–298. miol 2006;34:299–309.
24 Pontigo-Loyola AP, Islas-Marquez A, 34 Pendrys DG: Analytical studies of 47 Hong L, Levy SM, Warren JJ, Broffitt B,
Loyola-Rodriguez JP, Maupome G, enamel fluorosis: methodological Cavanaugh J: Fluoride intake levels in
Marquez-Corona ML, Medina-Solis CE: considerations. Epidemiol Rev relation to fluorosis development in
Dental fluorosis in 12- and 15-year-olds 1999;21:233–246. permanent maxillary central incisors
at high altitudes in above-optimal fluori- 35 Larsen MJ, Richards A, Fejerskov O: and first molars. Caries Res
dated communities in Mexico. J Public Development of dental fluorosis accord- 2006;40:494–500.
Health Dent 2008;68:163–166. ing to age at start of fluoride administra- 48 Burt BA, Keels MA, Heller KE: The
25 Juncos LI, Donadio JV Jr: Renal failure tion. Caries Res 1985;19:519–527. effects of a break in water fluoridation
and fluorosis. JAMA 1972;222:783–785. 36 Pendrys DG, Katz RV: Risk of enamel on the development of dental caries and
26 Porcar C, Bronsoms J, Lopez-Bonet E, fluorosis associated with fluoride sup- fluorosis. J Dent Res 2000;79:761–769.
Valles M: Fluorosis, osteomalacia and plementation, infant formula, and fluo- 49 Burt BA, Keels MA, Heller KE: Fluorosis
pseudohyperparathyroidism in a patient ride dentifrice use. Am J Epidemiol development in seven age cohorts after
with renal failure. Nephron 1989;130:1199–1208. an 11-month break in water fluorida-
1998;79:234–235. 37 Pendrys DG, Morse DE: Use of fluoride tion. J Dent Res 2003;82:64–68.
27 Turner CH, Owan I, Brizendine EJ, supplementation by children living in 50 Evans RW, Darvell BW: Refining the esti-
Zhang W, Wilson ME, Dunipace AJ: fluoridated communities. ASDC J Dent mate of the critical period for suscepti-
High fluoride intakes cause osteomalacia Child 1990;57:343–347. bility to enamel fluorosis in human max-
and diminished bone strength in rats 38 Evans RW, Stamm JW: An epidemiologic illary central incisors. J Public Health
with renal deficiency. Bone estimate of the critical period during Dent 1995;55:238–249.
1996;19:595–601. which human maxillary central incisors 51 Ishii T, Suckling G: The appearance of
28 Ibarra-Santana C, Ruiz-Rodriguez Mdel are most susceptible to fluorosis. J Public tooth enamel in children ingesting water
S, Fonseca-Leal Mdel P, Gutierrez-Cantu Health Dent 1991;51:251–259. with a high fluoride content for a limited
FJ, Pozos-Guillen Ade J: Enamel hyp- 39 Bronckers AL, Lyaruu DM, DenBesten period during early tooth development. J
oplasia in children with renal disease in PK: The impact of fluoride on amelo- Dent Res 1986;65:974–977.
a fluoridated area. J Clin Pediatr Dent blasts and the mechanisms of enamel 52 Bårdsen A: ‘Risk periods’ associated with
2007;31:274–278. fluorosis. J Dent Res 2009;88:877–893. the development of dental fluorosis in
40 Holm AK, Andersson R: Enamel miner- maxillary permanent central incisors: a
alization disturbances in 12-year-old meta-analysis. Acta Odontol Scand
children with known early exposure to 1999;57:247–256.
fluorides. Community Dent Oral Epide-
miol 1982;10:335–339.

Fluoride Intake of Children 15

53 Van Winkle S, Levy SM, Kiritsy MC, 66 Phantumvanit P, Shinawatra V, Poshy- 76 Buzalaf MA, Pinto CS, Rodrigues MH,
Heilman JR, Wefel JS, Marshall T: Water achinda U, Phijaisanit P, Traisup C: Levy FM, Borges AS, Furlani TA, da
and formula fluoride concentrations: Dietary fluoride intake of 4–6 month old Silva Cardoso VE: Availability of fluoride
significance for infants fed formula. infants in Bangkok. J Dent Assoc Thai from meals given to kindergarten chil-
Pediatr Dent 1995;17:305–310. 1987;37:226–231. dren in Brazil. Community Dent Oral
54 Ophaug RH, Singer L, Harland BF: 67 Pessan JP, Pin ML, Martinhon CC, de Epidemiol 2006;34:87–92.
Dietary fluoride intake of 6-month and Silva SM, Granjeiro JM, Buzalaf MA: 77 Silva M, Reynolds EC: Fluoride content
2-year-old children in four dietary Analysis of fingernails and urine as bio- of infant formulae in Australia. Aust
regions of the United States. Am J Clin markers of fluoride exposure from den- Dent J 1996;41:37–42.
Nutr 1985;42:701–707. tifrice and varnish in 4- to 7-year-old 78 Levy SM: An update on fluorides and
55 Levy SM, Maurice TJ, Jakobsen JR: Feed- children. Caries Res 2005;39:363–370. fluorosis. J Can Dent Assoc
ing patterns, water sources and fluoride 68 de Almeida BS, da Silva Cardoso VE, 2003;69:286–291.
exposures of infants and 1-year-olds. J Buzalaf MA: Fluoride ingestion from 79 Levy SM, Kiritsy MC, Warren JJ: Sources
Am Dent Assoc 1993;124:65–69. toothpaste and diet in 1- to 3-year-old of fluoride intake in children. J Public
56 Levy SM, Kohout FJ, Guha-Chowdhury Brazilian children. Community Dent Health Dent 1995;55:39–52.
N, Kiritsy MC, Heilman JR, Wefel JS: Oral Epidemiol 2007;35:53–63. 80 Heilman JR, Kiritsy MC, Levy SM,
Infants’ fluoride intake from drinking 69 Rodrigues MH, Leite AL, Arana A, Vil- Wefel JS: Fluoride concentrations of
water alone, and from water added to lena RS, Forte FD, Sampaio FC, Buzalaf infant foods. J Am Dent Assoc
formula, beverages, and food. J Dent Res MA: Dietary fluoride intake by children 1997;128:857–863.
1995;74:1399–1407. receiving different sources of systemic 81 Heilman JR, Kiritsy MC, Levy SM, Wefel
57 Levy SM, Kohout FJ, Kiritsy MC, Heil- fluoride. J Dent Res 2009;88:142–145. JS: Assessing fluoride levels of carbon-
man JR, Wefel JS: Infants’ fluoride inges- 70 Buzalaf MAR, Rodrigues MHC, Pessan ated soft drinks. J Am Dent Assoc
tion from water, supplements and denti- JP, Leite AL, Arana A, Villena RS, Forte 1999;130:1593–1599.
frice. J Am Dent Assoc FD, Sampaio FC: Biomarkers of fluoride 82 Kiritsy MC, Levy SM, Warren JJ, Guha-
1995;126:1625–1632. in children exposed to different sources Chowdhury N, Heilman JR, Marshall T:
58 Levy SM, Warren JJ, Davis CS, Kirchner of systemic fluoride. J Dent Res Assessing fluoride concentrations of
HL, Kanellis MJ, Wefel JS: Patterns of 2011;90:215–219. juices and juice-flavored drinks. J Am
fluoride intake from birth to 36 months. 71 Miziara AP, Philippi ST, Levy FM, Dent Assoc 1996;127:895–902.
J Public Health Dent 2001;61:70–77. Buzalaf MA: Fluoride ingestion from 83 McDonagh MS, Whiting PF, Wilson PM,
59 Clovis J, Hargreaves JA: Fluoride intake food items and dentifrice in Sutton AJ, Chestnutt I, Cooper J, Misso
from beverage consumption. 2–6-year-old Brazilian children living in K, Bradley M, Treasure E, Kleijnen J:
Community Dent Oral Epidemiol a fluoridated area using a semiquantita- Systematic review of water fluoridation.
1988;16:11–15. tive food frequency questionnaire. Com- BMJ 2000;321:855–859.
60 Chittaisong C, Koga H, Maki Y, Takaesu munity Dent Oral Epidemiol 84 Hujoel PP, Zina LG, Moimaz SA, Cunha-
Y: Estimation of fluoride intake in rela- 2009;37:305–315. Cruz J: Infant formula and enamel fluo-
tion to F, Ca, Mg and P contents in infant 72 Buzalaf MA, Granjeiro JM, Duarte JL, rosis: a systematic review. J Am Dent
foods. Bull Tokyo Dent Coll 1995;36: Taga ML: Fluoride content of infant Assoc 2009;140:841–854.
19–26. foods in Brazil and risk of dental fluoro- 85 Parnell C, Whelton H, O’Mullane D:
61 Chowdhury NG, Brown RH, Shepherd sis. ASDC J Dent Child 2002;69:196–200, Water fluoridation. Eur Arch Paediatr
MG: Fluoride intake of infants in New 125–126. Dent 2009;10:141–148.
Zealand. J Dent Res 1990;69:1828–1833. 73 Buzalaf MA, Damante CA, Trevizani 86 Ismail AI, Hasson H: Fluoride supple-
62 Guha-Chowdhury N, Drummond BK, LM, Granjeiro JM: Risk of fluorosis asso- ments, dental caries and fluorosis: a sys-
Smillie AC: Total fluoride intake in chil- ciated with infant formulas prepared tematic review. J Am Dent Assoc
dren aged 3 to 4 years – a longitudinal with bottled water. J Dent Child (Chic) 2008;139:1457–1468.
study. J Dent Res 1996;75:1451–1457. 2004;71:110–113. 87 Wong MC, Glenny AM, Tsang BW, Lo
63 Kimura T, Morita M, Kinoshita T, 74 Buzalaf MA, de Almeida BS, Cardoso EC, Worthington HV, Marinho VC: Topi-
Tsuneishi M, Akagi T, Yamashita VE, Olympio KP, Furlani Tde A: Total cal fluoride as a cause of dental fluorosis
F, Watanabe T: Fluoride intake from and acid-soluble fluoride content of in children. Cochrane Database Syst Rev
food and drink in Japanese children infant cereals, beverages and biscuits 2010:CD007693.
aged 1–6 years. Caries Res 2001;35: from Brazil. Food Addit Contam 88 Cardoso VE, Olympio KP, Granjeiro
47–49. 2004;21:210–215. JM, Buzalaf MA: Fluoride content of
64 Lima YB, Cury JA: [Fluoride intake by 75 Buzalaf MA, Granjeiro JM, Cardoso VE, several breakfast cereals and snacks
children from water and dentifrice]. Rev da Silva TL, Olympio KP: Fluorine con- found in Brazil. J Appl Oral Sci
Saude Publica 2001;35:576–581. tent of several brands of chocolate bars 2003;11:306–310.
65 Paiva SM, Lima YB, Cury JA: Fluoride and chocolate cookies found in Brazil. 89 CDC: Achievements in public health.
intake by Brazilian children from two Pesqui Odontol Bras 2003;17:223–227. 1900–1999: fluoridation of drinking
communities with fluoridated water. water to prevent dental caries. Morb
Community Dent Oral Epidemiol Mort Wkly Rep 1999;48:933–940.

16 Buzalaf · Levy
90 Jones S, Burt BA, Petersen PE, Lennon 102 Ramires I, Grec RH, Cattan L, Moura 112 Tavener JA, Davies GM, Davies RM, Ell-
MA: The effective use of fluoride in pub- PG, Lauris JR, Buzalaf MA: [Evaluation wood RP: The prevalence and severity of
lic health. Bull WHO 2005;83:670–676. of the fluoride concentration and fluorosis in children who received tooth-
91 Dean HT, Arnold FA, Elvolve E: Addi- consumption of mineral water]. Rev paste containing either 440 or 1,450 ppm
tional studies of the relation of fluoride Saude Publica 2004;38:459–465. F from the age of 12 months in deprived
domestic waters to dental caries experi- 103 Grec RH, de Moura PG, Pessan JP, and less deprived communities. Caries
ence in 4,425 white children aged 12–14 Ramires I, Costa B, Buzalaf MA: [Fluo- Res 2006;40:66–72.
years in 13 cities in 4 states. Public ride concentration in bottled water on 113 Holt RD, Morris CE, Winter GB, Downer
Health Rep 1942;57:1155–1179. the market in the municipality of Sao MC: Enamel opacities and dental caries
92 Szpunar SM, Burt BA: Dental caries, Paulo]. Rev Saude Publica 2008;42:154– in children who used a low fluoride
fluorosis, and fluoride exposure in Mich- 157. toothpaste between 2 and 5 years of age.
igan schoolchildren. J Dent Res 104 Quock RL, Chan JT: Fluoride content Int Dent J 1994;44:331–341.
1988;67:802–806. of bottled water and its implications for 114 Franzman MR, Levy SM, Warren JJ,
93 Riordan PJ: Dental fluorosis, dental car- the general dentist. Gen Dent 2009;57: Broffitt B: Fluoride dentifrice ingestion
ies and fluoride exposure among 29–33. and fluorosis of the permanent incisors.
7-year-olds. Caries Res 1993;27:71–77. 105 Cochrane NJ, Saranathan S, Morgan MV, J Am Dent Assoc 2006;137:645–652.
94 Riordan PJ, Banks JA: Dental fluorosis Dashper SG: Fluoride content of still 115 Pendrys DG, Katz RV: Risk factors for
and fluoride exposure in Western Aus- bottled water in Australia. Aust Dent J enamel fluorosis in optimally fluoridated
tralia. J Dent Res 1991;70:1022–1028. 2006;51:242–244. children born after the US manufactur-
95 Buzalaf MA, de Almeida BS, Olympio 106 Ahiropoulos V: Fluoride content of bot- ers’ decision to reduce the fluoride con-
KP, da SCVE, de CSPSH: Enamel fluoro- tled waters available in Northern Greece. centration of infant formula. Am J Epi-
sis prevalence after a 7-year interruption Int J Paediatr Dent 2006;16:111–116. demiol 1998;148:967–974.
in water fluoridation in Jaú, Sao Paulo, 107 Martinez-Mier EA, Soto-Rojas AE, 116 Pendrys DG, Katz RV, Morse DE: Risk
Brazil. J Public Health Dent Buckley CM, Zero DT, Margineda J: Flu- factors for enamel fluorosis in a fluori-
2004;64:205–208. oride concentration of bottled water, tap dated population. Am J Epidemiol
96 Griffin SO, Beltran ED, Lockwood SA, water, and fluoridated salt from two 1994;140:461–471.
Barker LK: Esthetically objectionable communities in Mexico. Int Dent J 117 Pendrys DG, Katz RV, Morse DE: Risk
fluorosis attributable to water fluorida- 2005;55:93–99. factors for enamel fluorosis in a nonfluo-
tion. Community Dent Oral Epidemiol 108 Zohouri FV, Maguire A, Moynihan PJ: ridated population. Am J Epidemiol
2002;30:199–209. Fluoride content of still bottled waters 1996;143:808–815.
97 U.S. Department of Health and Human available in the North-East of England, 118 Skotowski MC, Hunt RJ, Levy SM: Risk
Services: Proposed HHS Recommenda- UK. Br Dent J 2003;195:515–518, factors for dental fluorosis in pediatric
tion for Fluoride Concentration in discussion 507. dental patients. J Public Health Dent
Drinking Water for Prevention of Dental 109 Leverett D: Prevalence of dental fluoro- 1995;55:154–159.
Caries. U.S. Federal Register sis in fluoridated and nonfluoridated 119 Beltrán-Valladares PR, Cocom-Tun H,
2011;76:2383–2388. communities – a preliminary Casanova-Rosado JF, Vallejos-Sanchéz
98 Sohn W, Heller KE, Burt BA: Fluid con- investigation. J Public Health Dent AA, Medina-Solis CE, Maupomé G:
sumption related to climate among chil- 1986;46:184–187. Prevalence of dental fluorosis and addi-
dren in the United States. J Public Health 110 Richards A, Banting DW: Fluoride tooth- tional sources of exposure to fluoride as
Dent 2001;61:99–106. pastes; in Fejerskov O, Ekstrand J, Burt risk factors to dental fluorosis in school-
99 Ramires I, Maia LP, Rigolizzo Ddos S, BA, (eds): Fluoride in Dentistry, ed 2. children of Campeche, Mexico (in Span-
Lauris JR, Buzalaf MA: [External control Copenhagen, Munksgaard, 1996, pp ish). Rev Invest Clin 2005;57:532–539.
over the fluoridation of the public water 328–344. 120 Bottenberg P, Declerck D, Ghidey W,
supply in Bauru, SP, Brazil]. Rev Saude 111 Clarkson J, Watt RG, Rugg-Gunn AJ, Bogaerts K, Vanobbergen J, Martens L:
Publica 2006;40: 883–889. Pitiphat W, Ettinger RL, Horowitz Prevalence and determinants of enamel
100 Buzalaf MA, Granjeiro JM, Damante CA, AM, Petersen PE, ten Cate JM, Vianna fluorosis in Flemish schoolchildren. Car-
Ornelas F: Fluctuations in public water R, Ferrillo P, Gugushe TS, Siriphant ies Res 2004;38:20–28.
fluoride level in Bauru, Brazil. J Public P, Pine C, Buzalaf MA, Pessan JP, Levy 121 Brothwell DJ, Limeback H: Fluorosis risk
Health Dent 2002;62:173–176. S, Chankanka O, Maki Y, Postma in grade 2 students residing in a rural
101 Buzalaf MA, Granjeiro JM, Damante CA, TC, Villena RS, Wang WJ, MacEntee area with widely varying natural fluo-
de Ornelas F: Fluoride content of infant MI, Shinsho F, Cal E, Rudd RE, Schou ride. Community Dent Oral Epidemiol
formulas prepared with deionized, L, Shin SC, Fox CH: Proceedings: 9th 1999;27:130–136.
bottled mineral and fluoridated drink- World Congress on Preventive Dentistry 122 Conway DI, MacPherson LM, Stephen
ing water. ASDC J Dent Child 2001;68: (WCPD): ‘Community Participation and KW, Gilmour WH, Petersson LG: Preva-
37–41, 10. Global Alliances for Lifelong Oral Health lence of dental fluorosis in children from
for All,’ Phuket, Thailand, September non-water-fluoridated Halmstad, Swe-
7–10, 2009. Adv Dent Res 2010;22:2–30. den: fluoride toothpaste use in infancy.
Acta Odontol Scand 2005;63:56–63.

Fluoride Intake of Children 17

123 Do LG, Spencer AJ: Decline in the preva- 134 Vilhena FV, Olympio KP, Lauris JR, Del- 145 Pendrys DG: Risk of enamel fluorosis in
lence of dental fluorosis among South bem AC, Buzalaf MA: Low-fluoride nonfluoridated and optimally fluori-
Australian children. Community Dent acidic dentifrice: a randomized clinical dated populations: considerations for the
Oral Epidemiol 2007;35:282–91. trial in a fluoridated area. Caries Res dental professional. J Am Dent Assoc
124 Mascarenhas AK, Burt BA: Fluorosis risk 2010;44:478–484. 2000;131:746–755.
from early exposure to fluoride tooth- 135 Buzalaf MA, Vilhena FV, Iano FG, Grizzo 146 Chankanka O, Levy SM, Warren JJ,
paste. Community Dent Oral Epidemiol L, Pessan JP, Sampaio FC, Oliveira RC: Chalmers JM: A literature review of
1998;26:241–248. The effect of different fluoride concen- aesthetic perceptions of dental fluorosis
125 Maupomé G, Shulman JD, Clark DC, trations and pH of dentifrices on plaque and relationships with psychosocial
Levy SM: Socio-demographic features and nail fluoride levels in young chil- aspects/oral health-related quality of life.
and fluoride technologies contributing dren. Caries Res 2009;43:142–146. Community Dent Oral Epidemiol
to higher fluorosis scores in permanent 136 Villena RS: An investigation of the trans- 2010;38:97–109.
teeth of Canadian children. Caries Res verse technique of dentifrice application 147 Australian Research Centre for Popula-
2003;37:327–334. to reduce the amount of fluoride denti- tion Oral Health: The use of fluorides in
126 Pereira AC, Da Cunha FL, Meneghim frice for young children. Pediatr Dent Australia: guidelines. Aust Dent J
MC, Werner CW: Dental caries and fluo- 2000;22:312–317. 2006;51:195–199.
rosis prevalence study in a nonfluori- 137 Vilhena FV, Silva HM, Peres SH, Caldana 148 Rozier RG, Adair S, Graham F, Iafolla
dated Brazilian community: trend analy- Mde L, Buzalaf MA: The drop technique: T, Kingman A, Kohn W, Krol D, Levy
sis and toothpaste association. ASDC J a method to control the amount of fluo- S, Pollick H, Whitford G, Strock S,
Dent Child 2000;67:132–135, 83. ride dentifrice used by young children. Frantsve-Hawley J, Aravamudhan
127 Riordan PJ: Dental fluorosis decline after Oral Health Prev Dent 2008;6:61–65. K, Meyer DM: Evidence-based clinical
changes to supplement and toothpaste 138 Ismail AI, Bandekar RR: Fluoride sup- recommendations on the prescription of
regimens. Community Dent Oral Epide- plements and fluorosis: a meta-analysis. dietary fluoride supplements for caries
miol 2002;30:233–240. Community Dent Oral Epidemiol prevention. J Am Dent Assoc
128 Rock WP, Sabieha AM: The relationship 1999;27:48–56. 2010;141:1480–1489
between reported toothpaste usage in 139 de Liefde B, Herbison GP: Prevalence of 149 Swan E: Dietary fluoride supplement
infancy and fluorosis of permanent inci- developmental defects of enamel and protocol for the new millennium. J Can
sors. Br Dent J 1997;183:165–170. dental caries in New Zealand children Dent Assoc 2000;66:362, discussion 3.
129 Sagheri D, McLoughlin J, Clarkson JJ: receiving differing fluoride supplemen- 150 Clarkson J: A European view of fluoride
The prevalence of dental fluorosis in tation. Community Dent Oral Epidemiol supplementation. Br Dent J
relation to water or salt fluoridation and 1985;13:164–167. 1992;172:357.
reported use of fluoride toothpaste in 140 Granath L, Widenheim J, Birkhed D: 151 Koparal E, Ertugrul F, Oztekin K:
school-age children. Eur Arch Paediatr Diagnosis of mild enamel fluorosis in Fluoride levels in breast milk and infant
Dent 2007;8:62–68. permanent maxillary incisors using two foods. J Clin Pediatr Dent 2000;24:
130 Tabari ED, Ellwood R, Rugg-Gunn AJ, scoring systems. Community Dent Oral 299–302.
Evans DJ, Davies RM: Dental fluorosis in Epidemiol 1985;13:273–276. 152 Buzalaf MA, Pessan JP, Fukushima R,
permanent incisor teeth in relation to 141 Bagramian RA, Narendran S, Ward M: Dias A, Rosa HM: Fluoride content of
water fluoridation, social deprivation Relationship of dental caries and fluoro- UHT milks commercially available in
and toothpaste use in infancy. Br Dent J sis to fluoride supplement history in a Bauru, Brazil. J Appl Oral Sci
2000;189:216–220. non-fluoridated sample of schoolchil- 2006;14:38–42.
131 Morgan L, Allred E, Tavares M, Bellinger dren. Adv Dent Res 1989;3:161–167. 153 Latifah R, Razak IA: Fluoride levels
D, Needleman H: Investigation of the 142 Woolfolk MW, Faja BW, Bagramian RA: in infant formulas. J Pedod 1989;13:
possible associations between fluorosis, Relation of sources of systemic fluoride 323–327.
fluoride exposure, and childhood behav- to prevalence of dental fluorosis. J Public 154 Siew C, Strock S, Ristic H, Kang P, Chou
ior problems. Pediatr Dent 1998;20: Health Dent 1989;49:78–82. HN, Chen JW, Frantsve-Hawley J, Meyer
244–252. 143 Clark DC, Hann HJ, Williamson MF, DM: Assessing a potential risk factor for
132 Walsh T, Worthington HV, Glenny AM, Berkowitz J: Influence of exposure to enamel fluorosis: a preliminary
Appelbe P, Marinho VC, Shi X: Fluoride various fluoride technologies on the evaluation of fluoride content in infant
toothpastes of different concentrations prevalence of dental fluorosis. formulas. J Am Dent Assoc
for preventing dental caries in children Community Dent Oral Epidemiol 2009;140:1228–1236.
and adolescents. Cochrane Database Syst 1994;22:461–464. 155 Trautner K, Einwag J: Human plasma
Rev 2010:CD007868. 144 Hiller KA, Wilfart G, Schmalz G: Devel- fluoride levels following intake of denti-
133 Petersson LG, Lodding A, Hakeberg M, opmental enamel defects in children frices containing aminefluoride or
Koch G: Fluorine profiles in human with different fluoride supplementation monofluorophosphate. Arch Oral Biol
enamel after in vitro treatment with – a follow-up study. Caries Res 1988;33:543–546.
dentifrices of different compositions and 1998;32:405–411. 156 McKnight-Hanes MC, Leverett DH,
acidities. Swed Dent J 1989;13:177–183. Adair SM, Shields CP: Fluoride content
of infant formulas: soy-based formulas
as a potential factor in dental fluorosis.
Pediatr Dent 1988;10:189–194.

18 Buzalaf · Levy
157 Hong L, Levy SM, Warren JJ, Dawson 162 van der Hoek W, Ekanayake L, Raja- 167 Tavener JA, Davies GM, Davies RM, Ell-
DV, Bergus GR, Wefel JS: Association of sooriyar L, Karunaratne R: Source of wood RP: The prevalence and severity of
amoxicillin use during early childhood drinking water and other risk factors for fluorosis and other developmental
with developmental tooth enamel dental fluorosis in Sri Lanka. Int J Envi- defects of enamel in children who
defects. Arch Pediatr Adolesc Med ron Health Res 2003;13:285–293. received free fluoride toothpaste con-
2005;159:943–948. 163 Larsen MJ, Senderovitz F, Kirkegaard E, taining either 440 or 1450 ppm F from
158 Forsman B: Dental fluorosis and caries Poulsen S, Fejerskov O: Dental fluorosis the age of 12 months. Community Dent
in high-fluoride districts in Sweden. in the primary and the permanent denti- Health 2004;21:217–223.
Community Dent Oral Epidemiol tion in fluoridated areas with consump- 168 Vallejos-Sanchéz AA, Medina-Solis CE,
1974;2:132–148. tion of either powdered milk or natural Casanova-Rosado JF, Maupomé G,
159 Villa AE, Guerrero S, Icaza G, Villalobos cow’s milk. J Dent Res 1988;67:822–825. Minaya-Sanchéz M, Perez-Olivares S:
J, Anabalon M: Dental fluorosis in Chil- 164 Levy SM, Broffitt B, Marshall TA, Dental fluorosis in cohorts born before,
ean children: evaluation of risk factors. Eichenberger-Gilmore JM, Warren JJ: during, and after the national salt fluori-
Community Dent Oral Epidemiol Associations between fluorosis of per- dation program in a community in Mex-
1998;26:310–315. manent incisors and fluoride intake ico. Acta Odontol Scand 2006;64:209–
160 Walton JL, Messer LB: Dental caries and from infant formula, other dietary 213.
fluorosis in breast-fed and bottle-fed sources and dentifrice during early 169 Pendrys DG, Haugejorden O, Bårdsen A,
children. Caries Res 1981;15:124–137. childhood. J Am Dent Assoc Wang NJ, Gustavsen F: The risk of
161 Ericsson Y, Ribelius U: Increased fluo- 2010;141:1190–1201. enamel fluorosis and caries among Nor-
ride ingestion by bottle-fed infants and 165 Fomon SJ, Ekstrand J: Fluoride intake by wegian children: implications for Nor-
its effect. Acta Paediatr Scand infants. J Public Health Dent way and the United States. J Am Dent
1970;59:424–426. 1999;59:229–234. Assoc 2010;141:401–414.
166 Kumar JV, Swango PA: Fluoride expo-
sure and dental fluorosis in Newburgh
and Kingston, New York: policy implica-
tions. Community Dent Oral Epidemiol

Marília Afonso Rabelo Buzalaf

Department of Biological Sciences
Bauru Dental School, University of São Paulo
Al. Octávio Pinheiro Brisolla, 9–75
Bauru-SP, 17012–901 (Brazil)
Tel. +55 14 3235 8346, E-Mail

Fluoride Intake of Children 19

Fluoride Intake, Metabolism and Toxicity
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 20–36

Fluoride Metabolism
Marília Afonso Rabelo Buzalafa ⭈ Gary Milton Whitfordb
aDepartment of Biological Sciences, Bauru Dental School, University of São Paulo, Bauru, Brazil;
bDepartment of Oral Biology, Medical College of Georgia, Augusta, Ga., USA

Abstract and acute acid-base disturbances, hematocrit, altitude,

Knowledge of all aspects of fluoride metabolism is essen- physical activity, circadian rhythm and hormones, nutri-
tial for comprehending the biological effects of this ion tional status, diet, and genetic predisposition. These will
in humans as well as to drive the prevention (and treat- be discussed in detail in this review.
ment) of fluoride toxicity. Several aspects of fluoride Copyright © 2011 S. Karger AG, Basel
metabolism – including gastric absorption, distribution
and renal excretion – are pH-dependent because the Fluorine is a natural component of the biosphere.
coefficient of permeability of lipid bilayer membranes to It is the thirteenth most abundant element in the
hydrogen fluoride (HF) is 1 million times higher than that earth’s crust, constituting in the combined state
of F–. This means that fluoride readily crosses cell mem- around 0.065% by weight of the crust. Due to the
branes as HF, in response to a pH gradient between adja- small radius of the fluorine atom, its effective sur-
cent body fluid compartments. After ingestion, plasma face charge is the highest among all elements. As a
fluoride levels increase rapidly due to the rapid absorp- consequence, fluorine is the most electronegative
tion from the stomach, an event that is pH-dependent and reactive of all elements and hardly occurs in
and distinguishes fluoride from other halogens and nature in its elemental form. Instead, it is found
most other substances. The majority of fluoride not most frequently as inorganic fluoride that is widely
absorbed from the stomach will be absorbed from the distributed [1]. Besides its ubiquitous natural oc-
small intestine. In this case, absorption is not pH-depen- currence, widespread acceptance of the cariostatic
dent. Fluoride not absorbed will be excreted in feces. properties of fluoride has led to its addition to sys-
Peak plasma fluoride concentrations are reached within temic (such as water, salt, sugar, milk and supple-
20–60 min following ingestion. The levels start declin- ments) and topical vehicles (such as toothpastes,
ing thereafter due to two main reasons: uptake in calci- gels, foams, mouth rinses and varnishes) which
fied tissues and excretion in urine. Plasma fluoride levels are widely employed for caries control [Buzalaf
are not homeostatically regulated and vary according to et al., this vol., pp. 97–114; Pessan et al., this vol.,
the levels of intake, deposition in hard tissues and excre- pp. 115–132; Sampaio and Levy, this vol., pp. 133–
tion of fluoride. Many factors can modify the metabolism 145]. It can be inferred therefore that the human
and effects of fluoride in the organism, such as chronic organism is broadly exposed to fluoride. The main
1 2 3 4 5 6 7 8 9 10 11 12 13 14 pH

1 2 3 4 5 6
>50% <50%
F– + H+
Fig. 1. pH-dependency of fluoride <50% >50%
metabolism. HF is a weak acid with
a pKa of 3.4. Thus, at pH 3.4, 50% of
50% 50%
fluoride is in the undissociated form
HF F– + H+
(HF) while the remaining 50% is in
the dissociated or ionic form (F–). As
pH decreases from 3.4, the concen-
tration of HF increases, and as pH
increases, the concentration of F–
increases. The coefficient of perme-
ability of lipid bilayer membranes
Acidic Alkaline
to HF is 1 million times higher than
that of F–. Therefore, fluoride crosses
cell membranes as HF, in response
to a pH gradient (goes from the
more acidic compartment to the
more alkaline compartment).

sources of fluoride intake were described in the General Features of Fluoride Metabolism
chapter by Buzalaf and Levy [this vol., pp. 1–19].
Despite its proven benefits for caries control Several aspects of fluoride metabolism – includ-
[2], there is a benefit/risk ratio that needs to be ing gastric absorption, distribution and renal ex-
taken into account. The acute ingestion of a large cretion – are pH-dependent. Hydrogen fluoride
dose can provoke gastric and kidney disturbanc- (HF) is a weak acid with a pKa of 3.4. Thus, at pH
es or even death in extreme cases [Whitford, this 3.4, 50% of fluoride is in the undissociated form
vol., pp. 66–80]. Lower levels of excessive intake (HF) while the remaining 50% is in the dissoci-
on a chronic basis can affect the quality of the de- ated or ionic form (F–). As pH decreases from 3.4,
veloping mineralized tissues, resulting in dental the concentration of HF increases, and as pH in-
or skeletal fluorosis, depending on the amount creases, the concentration of F– increases [3]. The
and duration of intake [DenBesten and Li, this coefficient of permeability of lipid bilayer mem-
vol., pp. 81–96]. Thus, knowledge of all aspects branes to HF is 1 million times higher than that of
of fluoride metabolism is essential for not only F– [4]. This means that fluoride crosses cell mem-
understanding the biological effects of this ion branes as HF, in response to a pH gradient be-
in humans, but also to optimize opportuni- tween adjacent body fluid compartments, i.e. HF
ties to prevent or treat cases of excess fluoride goes from the more acidic compartment to the
ingestion. more alkaline compartment (fig. 1).

Fluoride Metabolism 21
20–60 min

F Concentration
0 1 2 3 4 Time (h)

GI tract

Hard tissues Plasma Soft tissues

Sweat Feces


Fig. 2. Typical plasma fluoride concentration curve after ingestion of a small

amount of fluoride and general features of fluoride metabolism. After inges-
tion, plasma fluoride levels increase rapidly, reaching a peak within 20–60
min due to absorption of fluoride in the GI tract and lung (to a lesser extent).
Fluoride not absorbed will be excreted in feces. Plasma is the central com-
partment from which and into which fluoride must transit for its later distri-
bution to hard and soft tissues and excretion. In adults, approximately 50% of
an absorbed amount of fluoride will become associated with calcified tissues
(mainly bone), where 99% of fluoride in the body is found. However, fluoride
is not irreversibly bound to bone and can be released back into plasma when
plasma fluoride levels fall (bidirectional arrows). A small amount of fluoride is
found in soft tissues, where a steady-state distribution between extracellular
and intracellular fluids is established. Most of the fluoride absorbed and not
taken up by mineralized tissues is excreted in urine, while only a small amount
of absorbed fluoride is excreted in sweat and feces.

General features of fluoride metabolism are fluoride not absorbed from the stomach will be
described in figures 2 and 3. Figure 2 also illus- absorbed from the small intestine, but in this
trates a typical plasma fluoride concentration case absorption is not pH-dependent (fig. 3) [5,
curve after ingestion of a small amount of fluo- 6]. Fluoride not absorbed will be excreted in the
ride. After ingestion, plasma fluoride levels in- feces [3].
crease rapidly (fig. 2) due to the ready absorption Peak plasma fluoride concentrations are
from the stomach, an event that is pH-dependent reached within 20–60 min following ingestion
and distinguishes fluoride from other halogens (fig. 2) and the levels start declining thereaf-
and most other substances [3]. The majority of ter due to two main reasons: uptake in calcified

22 Buzalaf · Whitford


F Urine
Feces F
Fig. 3. General features of fluoride

tissues and excretion in urine (fig. 2, 3). Plasma to variation due to dietary, environmental, genet-
is the central compartment from which and into ic, physiological and pathological variables that
which fluoride must transit for its later distri- will be discussed later in this chapter.
bution to hard and soft tissues and excretion.
In adults, approximately 50% of an absorbed
amount of fluoride will become associated with Fluoride Absorption
calcified tissues (mainly bone), where 99% of
fluoride in the body is found [7]. However, fluo- In the absence of high amounts of bi- and tri-
ride is not irreversibly bound to bone and can valent cations such as calcium, aluminum and
be released back into plasma when plasma fluo- magnesium that may complex fluoride and form
ride levels fall (bidirectional arrows in fig. 2, 3). insoluble compounds, approximately 80–90% of
A small amount of fluoride absorbed is found in an amount of ingested fluoride is absorbed from
soft tissues, where a steady-state distribution be- the gastrointestinal tract [3]. Fluoride absorp-
tween extracellular and intracellular fluids is es- tion occurs by passive diffusion (not against a
tablished. Most of the absorbed fluoride not tak- concentration gradient), and is not affected by
en up by mineralized tissues is excreted in urine temperature changes or metabolic inhibitors.
while only a small amount of absorbed fluoride Fluoride absorption occurs rapidly, with a half
is excreted in sweat and feces. If the amount of time of approximately 30 min. Unlike most sub-
fluoride ingested is small, the plasma fluoride stances, roughly 20–25% of the total fluoride in-
concentrations return to baseline levels within gested is absorbed from the stomach, while the
3–6 h (fig. 2) [3]. remainder is absorbed from the proximal small
It is important to highlight that these general intestine [3, 6, 8, 9]. Although fluoride absorp-
characteristics of fluoride metabolism are subject tion from the stomach occurs rapidly, the rate

Fluoride Metabolism 23
of absorption is determined by gastric acidity small intestine (around 70–75% of absorbed
[10, 11] and velocity of gastric emptying [6, 12]. fluoride) [5, 6]. The small intestine has a huge
Other factors that influence fluoride absorption capacity for fluoride absorption and fluoride is
are fluoride intake with other foods [13–15] rapidly absorbed following emptying from the
and the specific salt of fluoride ingested [13, 15, stomach. Fluoride absorption from the small
16]. intestine, differently from what happens in the
Gastric fluoride absorption is inversely relat- stomach, is unaffected by pH and occurs pre-
ed to the pH of the stomach content because, in dominantly as the ionic fluoride (fig. 3) crosses
the stomach, fluoride is absorbed predominantly the leaky epithelia through the tight junctions
as HF [10]. When ionic fluoride enters the acidic between the cells or paracellular channels [5].
gastric lumen environment, it is converted into The massive fluoride absorption from the small
HF which is an uncharged molecule that read- intestine compensates for the low gastric absorp-
ily crosses cell membranes, including the gastric tion at high pH, so that overall fluoride absorp-
mucosa [4]. Thus, the higher the acidity of the tion is relatively unaffected by gastric acidity
gastric content, the faster the fluoride absorption [11].
from the stomach. As a consequence, peak plas- Fluoride absorption is affected by the compo-
ma concentrations will be reached more quick- sition of the diet and intake with foods. For a sol-
ly and sooner from an acidic environment than uble fluoride compound, such as sodium fluoride
from a more neutral environment. The pH of the (NaF) added to water, almost 100% of the fluo-
solution in which fluoride is administered, un- ride is absorbed. If fluoride is ingested with milk
der conditions of normal gastric acid secretion, (or baby formula) or with foods, especially those
has little or no effect on fluoride absorption. containing high amounts of divalent or trivalent
However, animal studies have suggested that the cations that can complex fluoride and form in-
pH of the solution exerts a profound short-term soluble compounds, the degree of absorption is
effect on fluoride absorption when drugs that in- reduced [13–15, 17]. This is the basis for using
hibit gastric acid secretion are used. Solutions calcium-containing solutions to lavage the stom-
with lower pH would lead to a greater rate of ach in cases of acute fluoride toxicity [Whitford,
fluoride absorption in the short term [11]. The this vol., pp. 66–80].
extent of fluoride absorption from the stomach Regarding the type of fluoride ingested,
as a function of pH has important implications most of the published studies are in agreement
both for the treatment of acute fluoride toxicity that the total amount of fluoride absorbed
[Whitford, this vol., pp. 66–80] and the therapeu- from disodium monofluorophosphate (SMFP)
tic use of fluoride. is similar to that absorbed from NaF [14, 16].
Another factor that interferes with gastric flu- However, since absorption of fluoride from
oride absorption is the rate of gastric emptying. SMFP requires enzymatic hydrolysis of the
Animal studies have shown that even at early time moiety by phosphatases, fluoride absorption
periods, while most of the fluoride dose still re- from SMFP occurs more slowly than from NaF.
mained in the stomach, the majority of fluoride This leads to lower and delayed peak plasma
absorption occurred from the proximal small in- fluoride levels compared to those seen after
testine. Thus, delayed gastric emptying might re- ingestion of NaF [13, 15, 16]. Similarly, the
sult in slower and smaller increases in plasma flu- bioavailability of fluoride when ingested from
oride levels [6, 12]. naturally or artificially fluoridated water, which
Most of fluoride that is not absorbed from usually have different fluoride compounds, does
the stomach will be absorbed from the proximal not differ [18, 19].

24 Buzalaf · Whitford
Fluoride Distribution of different types of lipid-like molecules that bind
to plasma proteins; (2) can only be detected in
After absorption, fluoride is rapidly distributed plasma by the electrode after ashing; (3) is not ex-
throughout the organism. Plasma fluoride levels pected to increase with increasing levels of chron-
start to increase within 10 min following fluoride ic fluoride intake, suggesting little or no exchange
intake and peak concentrations are reached with- between the two pools. Together, the non-ionic
in 20–60 min. Baseline plasma fluoride levels are and ionic fractions constitute the so-called ‘total’
usually reached within 3–11 h after ingestion, de- plasma fluoride [3, 20].
pending on the amount ingested [3]. It is important to highlight that plasma ionic
From a pharmacokinetic point of view, plasma fluoride concentrations, unlike most other bio-
is regarded as the central compartment for fluo- logically relevant ions, are not homeostatically
ride distribution, since it is the fluid from which regulated. Instead they increase or decrease ac-
and into which fluoride must pass to be distrib- cording to the amount of fluoride intake, depo-
uted to hard and soft tissues and excreted. A small sition and removal in soft and hard tissues and
part (<1%) of absorbed fluoride is found in soft urinary excretion [3]. As a consequence, plasma
tissues, where a steady-state distribution between fluoride levels have been used as contemporary
extracellular and intracellular fluids is established biomarkers of exposure to fluoride (indicate pres-
[3]. This means that when there is an increase or ent exposure), although many physiological fac-
decrease in plasma fluoride levels, a proportional tors can influence plasma concentrations, regard-
change occurs in the fluoride concentrations of less of fluoride intake [Rugg-Gunn et al., this vol.,
the extracellular and intracellular fluids. Most flu- pp. 37–51].
oride absorbed (around 35% for healthy adults) is
taken up by calcified tissues where fluoride is re- Distribution to Soft Tissues
versibly bound and can be released back into plas- Fluoride in plasma is rapidly distributed to all tis-
ma when plasma fluoride levels fall (fig. 2) [7]. sues and organs. The velocity of distribution is
The quantitative and qualitative aspects of flu- determined by the rate of blood flow to the dif-
oride distribution to each of these compartments ferent tissues [20]. When considering fluoride
will be detailed below. distribution to soft tissues, it is useful to keep in
mind that fluoride accumulates in the more al-
Fluoride in Blood Plasma kaline compartment in response to a pH gradi-
There are two general forms of fluoride in human ent (diffusion equilibrium of HF across cell mem-
plasma. One fraction is ionic fluoride (also called branes). In other words, fluoride goes from the
inorganic or free fluoride) that can be detected by more acidic to the more alkaline environment
the fluoride ion-specific electrode. Ionic fluoride (fig. 1). Considering that the cytosol of mamma-
is not bound to other plasma constituents and is lian cells is usually more acidic than extracellu-
the form of significance in dentistry, medicine lar fluid, intracellular fluoride levels are typically
and public health. In the blood, ionic fluoride is 10–50% lower than those found in plasma and
not equally distributed between plasma and blood extracellular fluid (fig. 4, 5), as shown by short-
cells (its concentration in plasma is twice as high term experiments with radioactive fluoride in
as that found in the cells). The other fraction is the laboratory animals. However, intracellular fluo-
non-ionic fluoride whose biological function has ride concentrations change simultaneously and
not been established yet, although its concentra- in proportion to changes in plasma fluoride lev-
tion is usually higher than that of ionic fluoride. els [21]. Considering that the pH gradient across
This fraction: (1) seems to be composed chiefly the membranes of most cells can be changed by

Fluoride Metabolism 25
5 Plasma
Concentration ratios 25% 75%
AB muscle
SM gland

Total tissue

Fig. 4. Tissue/plasma fluoride (18F) concentration ratios

of soft tissues from the rat. AB = Abdominal; SM = sub- (0.25 × 100) + (0.75 × 100)
T/P = = 0.625
mandibular [21]. 100

Fig. 5. Distribution of fluoride in the water spaces of soft

tissues. The concentrations of fluoride in plasma and in-
terstitial fluid are assumed to be the same. The intracel-
altering extracellular pH, it is possible to promote
lular (IC) fluoride concentration is lower than that of the
the net flux of fluoride into or out of cells. For this extracellular (EC) fluid. T/P = Tisue/plasma. Modified from
reason, the recommended treatment in cases of Whitford [3].
acute and potentially toxic levels of fluoride in-
gestion includes alkalinization of the body fluids
as a means to promote a net flux of fluoride out
of cells, favoring fluoride elimination in the urine submandibular ductal saliva are slightly lower.
[22; Whitford, this vol., pp. 66–80]. Ductal salivary-to-plasma fluoride concentration
Figure 4 shows tissue/plasma fluoride (18F) ratios have been reported to be around 0.9 and
concentration ratios of different soft tissues from 0.8 for submandibular and parotid secretions, re-
published animal studies. The ratios are typically spectively [7]. Ductal saliva has been employed as
between 0.4 and 0.9 [21]. Exceptions are the brain a contemporary biomarker of fluoride exposure
(<0.1), because the blood-brain barrier is relative- rather than plasma to estimate the bioavailabil-
ly impermeable to fluoride, and the kidney (>4.0), ity of fluoride from fluoridated products or fluo-
due to the high fluoride concentrations within the ridated water [23–25]. Whole saliva usually has
tubular and interstitial fluids. fluoride concentrations more variable and higher
than those seen in ductal saliva due to exogenous
Distribution to Specialized Body Fluids contamination and is not recommended to esti-
Fluoride concentrations in some specialized body mate plasma fluoride levels [26]. For more details,
fluids are different from those found in plasma, see the chapter by Rugg-Gunn et al. [this vol., pp.
but the concentrations change simultaneously 37–51].
and in proportion to those found in plasma. This
is the case for cerebrospinal fluid and milk, which Distribution to Mineralized Tissues
have fluoride concentrations 50% or less than Fluoride is an avid mineralized tissue seeker.
that of plasma [3]. Gingival crevicular fluid fluo- Approximately 99% of all fluoride retained in the
ride levels are slightly higher than those in plas- human body is found in mineralized tissues, mainly
ma, whereas the concentrations in parotid and in bone but also in enamel and dentin [7]. Fluoride

26 Buzalaf · Whitford
concentration in bone is not uniform. In long bone, bone remodeling in the young, bone resorption
the concentrations are higher in the periosteal and and bone remodeling in the adult [35].
endosteal regions. Cancellous bone has higher flu- Dentin fluoride concentrations are similar to
oride concentrations than compact bone due to its bone fluoride concentrations and both tend to in-
greater surface area in contact with the surround- crease with age, i.e. they are proportional to the
ing extracellular fluid [27]. Bone fluoride concen- long-term level of fluoride intake. Dentin fluo-
trations tend to increase with age due to continu- ride levels are higher close to the pulp and reduce
ous fluoride uptake throughout life [27–29]. progressively towards the dentin-enamel junction
It is estimated that approximately 36% of the [36]. Enamel fluoride concentrations are usually
fluoride absorbed each day by healthy adults (18– lower than the levels found in dentin; no corre-
75 years) becomes associated with the skeleton, lation has been found between the fluoride con-
while the remainder is excreted in urine. In chil- centrations in these two dental tissues [37, 38].
dren (<7 years), the degree of retention is much Enamel fluoride concentrations tend to decrease
higher (around 55%) [30] due to the richer blood with age in areas subjected to tooth wear, but in-
supply and larger surface area of bone crystallites, crease in areas that accumulate dental biofilm [39].
which are smaller, more loosely organized, and The fluoride concentrations of tooth enamel gen-
more numerous than those of mature bone [7]. erally reflect the level of fluoride exposure during
Fluoride uptake by bone occurs in different its formation [36]. However, a significant correla-
stages [31]. The initial uptake occurs by iso- and tion between the severity of dental fluorosis and
heteroionic exchange on the hydration shells of tooth fluoride concentrations has been found for
bone crystallites. These ion-rich shells are con- dentin, but not for enamel [37, 38, 40].
tinuous with the extracellular fluids. In fact, it
is believed that a steady-state relationship exists
between the fluoride concentrations in the extra- Renal handling of Fluoride
cellular fluids and the hydration shells of bone
crystallites. According to this concept, there is a Kidneys represent the major route of fluoride re-
net transfer of fluoride from plasma to the hydra- moval from the body. Under normal conditions,
tion shells when the plasma concentration is ris- roughly 60% of fluoride absorbed each day by
ing and in the opposite direction when the plasma healthy adults (18–75 years) is excreted in urine.
concentration is falling [7]. For this reason, bone The corresponding percentage for children is 45%
surface has been suggested as a terminal biomark- [30]. As a consequence, plasma and urinary ex-
er of acute fluoride exposure [32–34; Rugg-Gunn cretion reflect a physiologic balance determined
et al., this vol., pp. 37–51]. Later stages involve flu- by previous fluoride intake, rate of fluoride uptake
oride association with or incorporation into pre- and removal from bone and the efficiency with
cursors of hydroxyfluorapatite and finally into the which the kidneys excrete fluoride.
apatitic lattice itself [31]. Since ionic fluoride is not bound to plasma pro-
A physiologically based pharmacokinet- teins, its concentration in the glomerular filtrate is
ic model considers that bone has two compart- the same found in plasma. After entering the renal
ments: a small, flow-limited, rapidly exchangeable tubules, a variable amount of the ion is reabsorbed
surface bone compartment and a bulk, virtually (from 10 to 90%) and returns to the systemic cir-
non-exchangeable, inner bone compartment. culation, while the remainder in excreted in urine
Fluoride associated with the inner bone compart- [20]. This process, together with glomerular filtra-
ment is not irreversibly bound. Over time, it may tion rate, is the main determinant of the amount
be mobilized through the continuous process of of fluoride excreted in urine. The reduction in

Fluoride Metabolism 27
glomerular filtration rate that occurs in chronic of ingested fluoride. Thus, more than 90% of in-
renal dysfunction as well as in the last decades of gested fluoride is usually absorbed [47, 48].
life, when the number of functional nephrons is Fluoride present in feces, however, does not
declining, will result in lower excretion and in- correspond solely to fluoride that was not ab-
creased plasma fluoride levels [20, 41]. sorbed. In two other situations, increased fecal
The renal clearance of fluoride (around 35 ml/ fluoride excretion has been reported in rats: when
min in healthy adults) is unusually high when plasma fluoride levels are high and when the diet
compared with the clearance of the other halogens contains high amounts of calcium (1% or higher).
(usually less than 1 or 2 ml/min). There is, howev- High plasma fluoride levels would cause net mi-
er, a high variation among individuals [7] that is at- gration of fluoride from the systemic circulation
tributed to alterations in glomerular filtration rate into the intestinal tract. On the other hand, when
[42], urinary pH [43–45] and flow rate [45, 46]. diets containing high amounts of calcium are con-
The mechanism of renal tubular reabsorption sumed, it is believed that unabsorbed calcium in
of fluoride, as happens for gastric absorption and the chyme binds fluoride entering the intestinal
transmembrane migration of fluoride, is also pH- tract, thus reducing the concentration of diffus-
dependent and occurs by diffusion of HF [44]. ible fluoride and allowing the migration of more
Thus, when the pH of the tubular fluid is relative- fluoride into the tract [49].
ly high, the proportion of fluoride as HF is lower
while there is a higher proportion of fluoride as
F–. As a consequence, only a small amount of HF Factors That Modify the Metabolism or
crosses the epithelium of the renal tubule to be re- Effects of Fluoride
absorbed and a high amount of fluoride is excreted
in urine as F–. On the other hand, when the pH By analyzing the general features of fluoride me-
of the tubular fluid is lower, high amounts of HF tabolism, it becomes clear that any condition –
cross the tubular epithelium into the interstitial flu- systemic, metabolic or genetic – which interferes
id where the pH is relatively high (around 7) which with the absorption or excretion of fluoride, will
promotes the dissociation of HF. F– is then released influence its fate in the body, and ultimately may
and diffuses into the peritubular capillaries return- alter the relationship between fluoride intake and
ing to the systemic circulation. The renal clearance the risk of dental or skeletal fluorosis. Variables
rate, in this case, is low (fig. 6). Thus, all conditions that have been reported to modify the general fea-
that alter urinary pH can affect the metabolic bal- tures of fluoride metabolism in the organism in-
ance and tissue concentrations of fluoride. These clude chronic and acute acid-base disturbances,
include diet composition, certain drugs (such as hematocrit, high altitude, physical activity, circa-
ascorbic acid, ammonium chloride, chlorothiazide dian rhythm and hormones [3]. Other predispos-
diuretics and methenamine mandelate), metabol- ing factors suggested are impaired kidney func-
ic and respiratory disorders, and altitude of resi- tion, genetic predisposition and nutritional status.
dence. These will be discussed later. These will be discussed in more detail below.

Acid-Base Disturbances
Fecal Fluoride Due to the effects of urinary pH on the efficien-
cy of kidneys to remove fluoride from the body,
Most of the fluoride found in feces corresponds to chronic acid-base disturbances play an important
the fraction that was not absorbed. Fecal fluoride role on the balance and tissue concentrations of
usually accounts for less than 10% of the amount fluoride. Factors that chronically alter the acid-

28 Buzalaf · Whitford
1 2 3 4 5 6 7 8 9 10 11 12 13 14 pH


Alkaline urine
F– HF H+ + F– F–



Tubule Interstitium Blood

1 2 3 4 5 6 7 8 9 10 11 12 13 14 pH

Fig. 6. Mechanism of fluoride re-

absorption from the renal tubule.
When urine is alkaline, there is a HF H+ + F– F–
Acidic urine

low concentration of HF and most F– HF H+ + F– F–

of fluoride remains in the tubule to HF H+ + F– F–
be excreted. When urine is acidic,
there is a high concentration of HF
that crosses the tubule membrane
towards the interstitium where it
dissociates originating F- that dif-
fuses into the peritubular capillaries Tubule Interstitium Blood
and returns to the systemic circula-
tion. Modified from Whitford [3].

base equilibrium include diet composition (veg- hypoplasia. The effects of uremia (increased con-
etarian diet tends to increase urinary pH, while centrations of urea in blood) on tooth formation
a diet with a high composition of meat tends to were evaluated in nephrectomized rats exposed to
decrease urinary pH), certain drugs, a variety of 0 or 50 ppm NaF in drinking water [50]. Intake of
metabolic and respiratory disorders, the level of fluoride by nephrectomized rats increased plas-
physical activity and the altitude of residence [3]. ma F levels twofold. It was also shown that ure-
Acute respiratory acid-base disorders affect renal mia affected the formation of dentin and enamel,
excretion of fluoride in the same manner as the and was more extensive than the effect of fluoride
metabolic disorders [3]. alone, demonstrating that intake of fluoride by
rats with reduced renal function impairs fluoride
Renal Impairment clearance from the plasma and aggravates the al-
Renal impairment in children has been associated ready negative effects of uremia on incisor tooth
with tooth defects that include enamel pitting and development. In humans, several studies have

Fluoride Metabolism 29
shown a direct relationship between renal impair- especially skeletal muscle cells, which promotes
ment and enamel defects, which include hypopla- the diffusion of fluoride (as HF) from the extra-
sia [51–53]. In a study comparing the frequency cellular to the intracellular fluid. In addition, re-
of dental fluorosis in children with renal disease nal vasoconstriction can occur due to increased
and healthy children, although no significant dif- secretion of catecholamines and muscular blood
ference was observed in the frequency of dental flow during exercise. Depending on the balance of
fluorosis between the 2 groups, patients with re- several factors, exercise could be associated with
nal disease presented more severe dental fluorosis either decreased or increased circulating fluoride
than children without renal disease [54]. levels [3]. It must be considered, however, that al-
though physical activity may alter the pattern of
Altitude of Residence fluoride excretion, the impact of such findings on
Researchers have noted that enamel disturbances the development of dental fluorosis seem to be
are exacerbated in rats raised in hypobaric cham- negligible, as prolonged physical activity in chil-
bers which simulated high altitudes, regardless of dren at the age risk for fluorosis is uncommon.
the levels of ingested fluoride [3]. Alterations in
acid-base balance, caused by hypobaric hypoxia Circadian Rhythm and Hormones
during residence at high altitude, were cited as the The possibility of existence of a biological rhythm
cause of decreased urinary excretion of fluoride in plasma fluoride levels was raised based on re-
and therefore greater retention of fluoride [55]. In ports of circadian rhythms for calcium and phos-
humans, a significantly higher prevalence of fluo- phate [61, 62]. The daily variations of these ions
rosis has been observed in Tanzanian communi- are partially attributed to the balance between
ties at a high altitude (1,463 m), in contrast with bone accretion and resorption, which are influ-
a low altitude area (100 m), but with similar food enced by bone-active hormones. As the bulk of
habits and low levels of fluoride in the drinking fluoride is contained in the skeleton, it was hy-
water [56]. The authors concluded that the sever- pothesized that plasma fluoride levels would ex-
ity of enamel disturbances at the high altitude area hibit a circadian rhythm similar to, and in phase
was not consistent with the low fluoride concen- with, that of calcium and phosphate. Such rhyth-
tration in drinking water, suggesting that altitude, micity was verified in dogs, with a mean peak flu-
along with other factors, is a variable which may oride concentration around 9 a.m., followed by a
be contributing to the severity of dental enamel decrease around 9 p.m. [3].
disturbances occurring in that area. Studies con- The administration of parathormone or salm-
ducted in other countries confirmed this finding on thyrocalcitonin to humans demonstrated for
[57–60], suggesting that physiological changes as- the first time that alterations in hormone-medi-
sociated with residence at high altitude are able to ated bone accretion and resorption are reflected
exacerbate the effects of fluoride in mineralized in plasma and urinary fluoride levels. However,
tissues. Such disturbances may be due to hypoxia as reported in published animal studies [61, 62],
in high altitude areas. This ultimately leads to a the rhythmic pattern for calcium and phosphate
decrease in urinary pH, reducing fluoride renal occurred in the opposite way of that verified for
excretion and, therefore, increasing fluoride con- fluoride, suggesting that a physiological system,
centrations in the body. other than bone, would be the responsible for the
characteristics of the biological rhythmicity of
Physical Activity fluoride in plasma. A recent study suggested that
In prolonged physical activity, there is a reduc- the renal system is involved with such rhythmic-
tion in the pH gradient across cell membranes, ity in humans. Cardoso et al. [63] demonstrated

30 Buzalaf · Whitford
a rhythmicity for fluoride concentrations in plas- considered with caution to avoid misinterpreta-
ma, with mean peak (0.55 μmol/l) at 11 a.m. and tion. The authors correlated their findings (high
the lowest concentrations (mean of 0.50 μmol/l) prevalence of dental fluorosis) with previous in-
occurring between 5 and 8 p.m. Plasma fluoride formation on nutrition in 2 of the 3 areas evalu-
concentrations were positively correlated with ated, but no direct comparison between children
urinary fluoride excretion rates and with serum with or without malnutrition regarding the preva-
parathormone levels, suggesting that both the lence of dental fluorosis was carried out.
renal system and hormones might be involved Correia Sampaio et al. [67] demonstrated that
in the rhythmicity for plasma fluoride concen- dental fluorosis is independent of nutritional sta-
trations in humans. It was also recently demon- tus. Nutritional status was assessed by the height-
strated that the diurnal average fractional urinary for-age (chronic malnutrition) and weight-for-age
fluoride excretion is significantly lower than the (general malnutrition) indexes, recommended
average nocturnal one [64], which is in line with by the WHO. A significant relationship between
the findings of Cardoso et al. [63] and the sug- dental fluorosis and water fluoride concentration
gested rhythmicity for plasma fluoride concentra- was found, but not with regard to nutritional sta-
tions. The existence of this rhythmicity may alter tus or sex. Dental fluorosis may be related to other
the relationship between fluoride intake and the factors, like infant dietary habits or increased con-
risk of dental or skeletal fluorosis. sumption of fluoridated water. Future studies on
this subject should consider a longitudinal study
Nutritional Status design where nutritional status, infant dietary
Although an association between malnutrition habits and fluoride intake are assessed during the
and dental fluorosis prevalence and severity has tooth formation period. This is particularly im-
been suggested for decades, the evidence for such portant for developing countries, where malnutri-
a relationship is controversial and difficult to in- tion and dental fluorosis are prevalent and fluo-
terpret. If a fasting child may absorb fluoride from ride-containing products are introduced in order
water or other sources more quickly than a well- to control dental caries.
fed child (due to the inexistence of complexes of
fluoride in an empty stomach), a malnourished Diet Composition
child, on the other hand, may have low fluoride The acidification and subsequent alkalinization
deposition over a long-term period of time (due of urine by ingestion of NH4Cl and NaHCO3, re-
to slower bone growth). spectively, led to significant differences in urinary
A statistically significant relationship between fluoride clearance and plasma half-lives of 5 adult
water fluoride concentration, socioeconomic sta- volunteers [68]. Similar findings were obtained
tus, nutritional status and the prevalence of diffuse by acidifying and alkalinizing urine by following
enamel lesions (DDE index) in boys from Saudi a protein-rich (meat/dairy products) and a veg-
Arabia has been demonstrated by Rugg-Gunn et etarian diet, respectively. These results strongly
al. [65]. Although the diffuse enamel defects of suggested that long-term diet-induced changes
the DDE index are considered as an indicator of in urinary pH could decrease (alkaline urine) or
dental fluorosis, direct comparisons between the increase (acidic urine) the risk of dental fluorosis
DDE index and specific dental fluorosis indi- [55]. The prevalence and severity of dental fluo-
ces have been discouraged [66]. In a study with rosis were compared among vegetarian and non-
Tanzanian children, Yoder et al. [56] suggested a vegetarian children and adolescents living in an
direct relationship between malnutrition and den- area with endemic dental fluorosis in India [69].
tal fluorosis. Such assumptions, however, must be Vegetarianism was inversely associated with the

Fluoride Metabolism 31
prevalence of dental fluorosis. The prevalence to the prevalence of dental fluorosis [73, 74]. It
and severity (Thylstrup and Fejersko index ≥4) was suggested that calcium supplementation
of dental fluorosis were 67 and 21%, respective- should be implemented in areas with endemic
ly, in the vegetarian group, and 95 and 35%, re- fluorosis in order to minimize the effects of fluo-
spectively, in the nonvegetarian group (p < 0.05). ride on mineralized tissues [74]. However, there
In addition, multiple logistic regression analysis is not enough evidence to support this, since none
showed that the risk of developing dental fluo- of the above-mentioned studies were able to de-
rosis was 7 times higher among nonvegetarians termine the effect of calcium alone in communi-
than among vegetarians. Tamarind has also been ties with similar background exposure to fluoride
shown to increase urinary fluoride excretion by from water.
increasing urinary pH in schoolchildren [70]. The usual diet also appears to be important.
In a study conducted in a fluoride endemic area Fluoride retention and resulting toxicity were
in South India, a significant decrease in urinary found to be higher with sorghum (also called
fluoride excretion was seen after volunteers were jowar) or sorghum-based diets than with rice- or
supplied with defluoridated water for 2 weeks. wheat-based diets when the fluoride intakes were
Then half of the subjects were supplemented with similar. Fluoride excretion in urine was signifi-
tamarind for 3 weeks, while the control group re- cantly high on rice-based diets as compared with
ceived defluoridated water for the same period. A the sorghum-based diet [75].
significant increase in fluoride excretion and uri-
nary pH was observed in the experimental group Genetic Factors
[71]. Tartaric acid is a major component of the Epidemiological observations of marked varia-
tamarind paste (8.4–12.4%), which does not get tion in dental fluorosis prevalence in subjects
metabolized and is excreted as such through the from areas with comparable levels of fluoride in-
urine. take [56], or even in studies showing different de-
Other dietary constituents also seem to play grees of susceptibility to fluorosis between certain
an important role in the balance between fluo- ethnic groups [76–78] have led to the assumption
ride and fluorosis. High dietary concentrations of that the predisposition to dental fluorosis is ge-
certain cations, especially calcium, can reduce the netically determined [79, 80].
extent of fluoride absorption [49]. In a study con- In a study conducted with Tanzanian children
ducted in the province of Jiangxi, China, where from three distinct areas, which differed regard-
the prevalence of dental fluorosis is reported to ing water fluoride concentrations and altitude, it
be above 50%, the incidence rates of dental flu- was observed that even in the two sites with more
orosis were found to differ markedly, depending severe fluorosis, several children had very little ev-
on whether or not the children consumed milk. idence of enamel disturbances [56]. These ‘resis-
The rate of dental fluorosis of the milk-drinking tant’ children were lifelong residents of the same
group was 7.2%, whereas that of the non-milk- area of the ‘susceptible’ children. Urinary fluoride
drinking group was 37.5% [72]. In India, where values and meal fluoride values from children
approximately 62 million people (including 6 were also similar between the two groups. The
million children) have dental fluorosis (mainly question of possible genetic influence became
endemic), some studies have been conducted in more evident due to the tribal homogeneity in the
order to identify components other than fluoride area were fluorosis prevalence was unexpectedly
associated with an increased risk of dental fluoro- high (no fluoridated drinking water).
sis. Low calcium concentrations in the drinking The possibility of genetic predisposition
water were demonstrated to be inversely related to dental fluorosis was demonstrated using a

32 Buzalaf · Whitford
mouse model system where genotype, age, gen- gene were genotyped. Calcitonin and osteocalcin
der, food, housing and drinking water fluoride levels in the serum were measured. Children car-
levels were under control [81]. Examination rying the homozygous genotype PP of COL1A2
of 12 inbred strains of mice showed differenc- PvuII had a significantly increased risk of dental
es in susceptibilities to dental fluorosis. The fluorosis (OR = 4.85, 95% CI: 1.22–19.32) com-
A/J mouse strain was highly susceptible, with pared to children carrying the homozygous geno-
a rapid onset and severe development of dental type pp in an endemic fluorosis village. However,
fluorosis compared to the other strains tested, the risk was not elevated when the control pop-
whereas the 129P3/J mouse strain was less af- ulation was recruited from a non-endemic fluo-
fected, with minimal dental fluorosis. It was lat- rosis village. Additionally, fluoride levels in urine
ter demonstrated that these 2 strains also have and osteocalcin levels in serum were found to be
different bone responses to fluoride exposure significantly lower in controls from non-endemic
[82]. It was hypothesized that the different sus- villages compared to cases. However, the differ-
ceptibility to dental fluorosis between these two ences in fluoride and osteocalcin levels were not
2 strains was due to differences in fluoride me- observed when cases were compared to a control
tabolism, i.e. it was expected that the resistant population from endemic fluorosis villages. This
strain would excrete more fluoride which in turn study provided the first evidence of an association
would lead to decreased susceptibility to den- between polymorphisms in the COL1A2 gene
tal fluorosis. Thus, a metabolic study was con- with dental fluorosis in high-fluoride-exposed
ducted to test this hypothesis. Surprisingly, the populations [85].
resistant strain (129P3/J) excreted a significantly
lower amount of fluoride in urine than the sus-
ceptible strain (A/J) and, as a result, had signifi- Conclusion
cantly higher plasma and bone fluoride concen-
trations. Despite this, the amelogenesis in the In view of the diverse effects that fluoride can
129P3/J strain was remarkably unaffected by produce in biological systems, it is not surpris-
fluoride [83]. Dental fluorosis-associated quan- ing that it has been the subject of thousands of
titative trait loci were detected on mouse chro- scientific reports. It is clear that the beneficial as
mosomes 2 and 11. Histological examination of well as the adverse effects of fluoride can be attrib-
maturing enamel showed that fluoride treatment uted to the magnitude and duration of the con-
resulted in accumulation of amelogenins in the centration of the ion at specific tissue or cellular
maturing enamel of A/J mice, but not of 129P3/J sites. In addition to the level of prior fluoride ex-
mice [84]. The physiological, biochemical and/ posure, these concentrations are determined by
or molecular mechanisms underlying this resis- the characteristics of the general metabolism of
tance remain to be determined. fluoride within the individual. As has been made
In humans, the possibility of gene-environ- clear in this chapter, these characteristics are not
ment interaction was assessed by determining constant within or among individuals or popula-
differential susceptibility to fluorosis at a given tions. Instead they are subject to the effects of di-
level of fluoride exposure based upon genetic verse environmental, biochemical, physiological
background. A case-control study was conduct- and pathological factors. While much has been
ed among children between 8 and 12 years of age learned during the last few decades, much re-
with (n = 75) and without (n = 165) dental fluoro- mains to be done – particularly in clearly defining
sis in two counties in Henan Province, China. The the mechanisms underlying the metabolism and
PvuII and RsaI polymorphisms in the COL1A2 biological effects of fluoride. With the continuing

Fluoride Metabolism 33
development of advanced analytical, diagnostic, Acknowledgments
molecular and genetic techniques, we can expect
The authors thank Prof. Heitor Marques Honório for de-
our knowledge to grow and, with that growth, the signing figures 1, 2, 5 and 6 and Prof. Juliano Pelim Pessan
beneficial effects of fluoride will be enhanced and for help with the section ‘Factors that modify the metabo-
the unwanted effects minimized. lism or effects of fluoride’.

1 Smith FA, Ekstrand J: The occurrence 14 Trautner K, Siebert G: An experimental 24 Olympio KP, Bardal PA, Cardoso VE,
and chemistry of fluoride; in Fejerskov study of bio-availability of fluoride from Oliveira RC, Bastos JR, Buzalaf MA:
O, Ekstrand J, Burt BA (eds): Fluoride in dietary sources in man. Arch Oral Biol Low-fluoride dentifrices with reduced
Dentistry, ed 2. Copenhagen, 1986;31:223–228. pH: fluoride concentration in whole
Munksgaard, 1996, pp 17–26. 15 Trautner K, Einwag J: Influence of milk saliva and bioavailability. Caries Res
2 Bratthall D, Hansel-Petersson G, Sund- and food on fluoride bioavailability from 2007;41:365–370.
berg H: Reasons for the caries decline: NaF and Na2FPO3 in man. J Dent Res 25 Wilson AC, Bawden JW: Salivary fluo-
what do the experts believe? Eur J Oral 1989;68:72–77. ride concentrations in children with var-
Sci 1996;104:416–422, discussion 16 Buzalaf MAR, Leite AL, Carvalho NTA, ious systemic fluoride exposures. Pediatr
423–425, 430–432. et al.: Bioavailability of fluoride adminis- Dent 1991;13:103–105.
3 Whitford GM: The metabolism and tox- tered as sodium fluoride or monofluoro- 26 Whitford GM, Thomas JE, Adair SM:
icity of fluoride. Monogr Oral Sci phosphate to humans. J Fluorine Chem Fluoride in whole saliva, parotid ductal
1996;16:1–153. 2008;129:691–694. saliva and plasma in children. Arch Oral
4 Gutknecht J, Walter A: Hydrofluoric and 17 Spak CJ, Ekstrand J, Zylberstein D: Bio- Biol 1999;44:785–788.
nitric acid transport through lipid availability of fluoride added by baby 27 Weidmann SM, Weatherell JA: The
bilayer membranes. Biochim Biophys formula and milk. Caries Res 1982;16: uptake and distribution of fluorine in
Acta 1981;644:153–156. 249–256. bones. J Pathol Bacteriol 1959;78:
5 Nopakun J, Messer HH: Mechanism of 18 Maguire A, Zohouri FV, Mathers JC, 243–255.
fluoride absorption from the rat small Steen IN, Hindmarch PN, Moynihan PJ: 28 Parkins FM, Tinanoff N, Moutinho M,
intestine. Nutr Res 1990;10:771–9. Bioavailability of fluoride in drinking Anstey MB, Waziri MH: Relationships of
6 Nopakun J, Messer HH, Voller V: Fluo- water: a human experimental study. human plasma fluoride and bone fluo-
ride absorption from the gastrointestinal J Dent Res 2005;84:989–993. ride to age. Calcif Tissue Res 1974;16:
tract of rats. J Nutr 1989;119:1411–1417. 19 Whitford GM, Sampaio FC, Pinto CS, 335–338.
7 Whitford GM: Intake and metabolism of Maria AG, Cardoso VE, Buzalaf MA: 29 Richards A, Mosekilde L, Sogaard CH:
fluoride. Adv Dent Res 1994;8:5–14. Pharmacokinetics of ingested fluoride: Normal age-related changes in fluoride
8 Wagner MJ: Absorption of fluoride by lack of effect of chemical compound. content of vertebral trabecular bone –
the gastric mucosa in the rat. J Dent Res Arch Oral Biol 2008;53:1037–1041. relation to bone quality. Bone 1994;15:
1962;41:667–671. 20 Ekstrand J: Fluoride metabolism; in 21–26.
9 Stookey GK, Dellinger EL, Muhler JC: In Fejerskov O, Ekstrand J, Burt BA (eds): 30 Villa A, Anabalon M, Zohouri V, Magu-
vitro studies concerning fluoride absorp- Fluoride in Dentistry, ed 2. Copenhagen, ire A, Franco AM, Rugg-Gunn A: Rela-
tion. Proc Soc Exp Biol Med 1964;115: Munksgaard, 1996, pp 55–68. tionships between fluoride intake, uri-
298–301. 21 Whitford GM, Pashley DH, Reynolds nary fluoride excretion and fluoride
10 Whitford GM, Pashley DH: Fluoride KE: Fluoride tissue distribution: short- retention in children and adults: an
absorption: the influence of gastric acid- term kinetics. Am J Physiol 1979;236: analysis of available data. Caries Res
ity. Calcif Tissue Int 1984;36:302–307. F141–F148. 2010;44:60–68.
11 Messer HH, Ophaug RH: Influence of 22 Whitford GM, Reynolds KE, Pashley 31 Neuman WF, Neuman MW: The Chemi-
gastric acidity on fluoride absorption in DH: Acute fluoride toxicity: influence of cal Dynamics of Bone Mineral, ed 1.
rats. J Dent Res 1993;72:619–622. metabolic alkalosis. Toxicol Appl Phar- Chicago, University of Chicago Press,
12 Messer HH, Ophaug R: Effect of delayed macol 1979;50:31–39. 1958.
gastric emptying on fluoride absorption 23 Zero DT, Raubertas RF, Fu J, Pedersen 32 Bezerra de Menezes LM, Volpato MC,
in the rat. Biol Trace Elem Res 1991;31: AM, Hayes AL, Featherstone JD: Fluo- Rosalen PL, Cury JA: Bone as a bio-
305–315. ride concentrations in plaque, whole marker of acute fluoride toxicity. Foren-
13 Ekstrand J, Ehrnebo M: Influence of saliva, and ductal saliva after application sic Sci Int 2003;137:209–214.
milk products on fluoride bioavailability of home-use topical fluorides [published 33 Buzalaf MA, Caroselli EE, Cardoso de
in man. Eur J Clin Pharmacol 1979;16: erratum appears in J Dent Res 1993 Oliveira R, Granjeiro JM, Whitford GM:
211–215. Jan;72(1):87]. J Dent Res 1992;71: Nail and bone surface as biomarkers for
1768–1775. acute fluoride exposure in rats. J Anal
Toxicol 2004;28:249–252.

34 Buzalaf · Whitford
34 Buzalaf MA, Caroselli EE, de Carvalho 46 Chen PS Jr, Gardner DE, Hodge HC, 58 Martinez-Mier EA, Soto-Rojas AE,
JG, de Oliveira RC, da Silva Cardoso VE, O’Brien JA, Smith FA: Renal clearance of Urena-Cirett JL, Katz BP, Stookey GK,
Whitford GM: Bone surface and whole fluoride. Proc Soc Exp Biol Med Dunipace AJ: Dental fluorosis and alti-
bone as biomarkers for acute fluoride 1956;92:879–883. tude: a preliminary study. Oral Health
exposure. J Anal Toxicol 2005;29: 47 Ekstrand J, Hardell LI, Spak CJ: Fluoride Prev Dent 2004;2:39–48.
810–813. balance studies on infants in a 1-ppm- 59 Pontigo-Loyola AP, Islas-Marquez A,
35 Rao HV, Beliles RP, Whitford GM, water-fluoride area. Caries Res Loyola-Rodriguez JP, Maupomé G, Mar-
Turner CH: A physiologically based 1984;18:87–92. quez-Corona ML, Medina-Solis CE:
pharmacokinetic model for fluoride 48 Ekstrand J, Ziegler EE, Nelson SE, Dental fluorosis in 12- and 15-year-olds
uptake by bone. Regul Toxicol Pharma- Fomon SJ: Absorption and retention of at high altitudes in above-optimal fluori-
col 1995;22:30–42. dietary and supplemental fluoride by dated communities in Mexico. J Public
36 Weatherell JA: Uptake and distribution infants. Adv Dent Res 1994;8:175–180. Health Dent 2008;68:163–166.
of fluoride in bones and teeth and the 49 Whitford GM: Effects of plasma fluoride 60 Akosu TJ, Zoakah AI: Risk factors asso-
development of fluorosis; in Barltrop W, and dietary calcium concentrations on ciated with dental fluorosis in Central
Burland WL (eds): Mineral Metabolism GI absorption and secretion of fluoride Plateau State, Nigeria. Community Dent
in Paediatrics, ed 1. Oxford, Blackwell, in the rat. Calcif Tissue Int 1994;54: Oral Epidemiol 2008;36:144–148.
1969, pp 53–70. 421–425. 61 Talmage RV, Roycroft JH, Anderson JJ:
37 Vieira AP, Hancock R, Dumitriu M, 50 Lyaruu DM, Bronckers AL, Santos F, Daily fluctuations in plasma calcium,
Limeback H, Grynpas MD: Fluoride’s Mathias R, DenBesten P: The effect of phosphate, and their radionuclide con-
effect on human dentin ultrasound fluoride on enamel and dentin forma- centrations in the rat. Calcif Tissue Res
velocity (elastic modulus) and tubule tion in the uremic rat incisor. Pediatr 1975;17:91–102.
size. Eur J Oral Sci 2006;114:83–88. Nephrol 2008;23:1973–1979. 62 Perault-Staub AM, Staub JF, Milhaud G:
38 Vieira AP, Hancock R, Limeback H, Maia 51 Koch MJ, Buhrer R, Pioch T, Scharer K: A new concept of plasma calcium
R, Grynpas MD: Is fluoride concentra- Enamel hypoplasia of primary teeth in homeostasis in the rat. Endocrinology
tion in dentin and enamel a good indica- chronic renal failure. Pediatr Nephrol 1974;95:480–484.
tor of dental fluorosis? J Dent Res 1999;13:68–72. 63 Cardoso VES, Whitford GM, Aoyama H,
2004;83:76–80. 52 Al-Nowaiser A, Roberts GJ, Trompeter Buzalaf MAR: Daily variations in human
39 Weatherell JA, Robinson C, Hallsworth RS, Wilson M, Lucas VS: Oral health in plasma fluoride concentrations. J Fluo-
AS: Changes in the fluoride concentra- children with chronic renal failure. Pedi- rine Chem 2008;129:1193–1198.
tion of the labial enamel surface with atr Nephrol 2003;18:39–45. 64 Villa A, Anabalon M, Cabezas L, Rugg-
age. Caries Res 1972;6:312–324. 53 Farge P, Ranchin B, Cochat P: Four-year Gunn A: Fractional urinary fluoride
40 Vieira A, Hancock R, Dumitriu M, follow-up of oral health surveillance in excretion of young female adults during
Schwartz M, Limeback H, Grynpas M: renal transplant children. Pediatr Neph- the diurnal and nocturnal periods. Car-
How does fluoride affect dentin micro- rol 2006;21:851–855. ies Res 2008;42:275–281.
hardness and mineralization? J Dent Res 54 Ibarra-Santana C, Ruiz-Rodriguez Mdel 65 Rugg-Gunn AJ, al-Mohammadi SM, But-
2005;84:951–957. S, Fonseca-Leal Mdel P, Gutierrez-Cantu ler TJ: Effects of fluoride level in drink-
41 Schiffl HH, Binswanger U: Human uri- FJ, Pozos-Guillen Ade J: Enamel hyp- ing water, nutritional status, and socio-
nary fluoride excretion as influenced by oplasia in children with renal disease in economic status on the prevalence of
renal functional impairment. Nephron a fluoridated area. J Clin Pediatr Dent developmental defects of dental enamel
1980;26:69–72. 2007;31:274–278. in permanent teeth in Saudi 14-year-old
42 Spak CJ, Berg U, Ekstrand J: Renal clear- 55 Whitford GM: Determinants and mech- boys. Caries Res 1997;31:259–267.
ance of fluoride in children and adoles- anisms of enamel fluorosis. Ciba Found 66 Clarkson J, O’Mullane D: A modified
cents. Pediatrics 1985;75:575–579. Symp 1997;205:226–241, discussion DDE Index for use in epidemiological
43 Jarnberg PO, Ekstrand J, Ehrnebo M: 241–245. studies of enamel defects. J Dent Res
Renal excretion of fluoride during water 56 Yoder KM, Mabelya L, Robison VA, 1989;68:445–450.
diuresis and induced urinary pH- Dunipace AJ, Brizendine EJ, Stookey GK: 67 Correia Sampaio F, Ramm von der Fehr
changes in man. Toxicol Lett 1983;18: Severe dental fluorosis in a Tanzanian F, Arneberg P, Petrucci Gigante D, Hatloy
141–146. population consuming water with negli- A: Dental fluorosis and nutritional status
44 Whitford GM, Pashley DH, Stringer GI: gible fluoride concentration. Commu- of 6- to 11-year-old children living in
Fluoride renal clearance: a pH-depen- nity Dent Oral Epidemiol 1998;26: rural areas of Paraiba, Brazil. Caries Res
dent event. Am J Physiol 1976;230: 382–393. 1999;33:66–73.
527–532. 57 Rwenyonyi C, Bjorvatn K, Birkeland J, 68 Ekstrand J, Ehrnebo M, Whitford GM,
45 Ekstrand J, Spak CJ, Ehrnebo M: Renal Haugejorden O: Altitude as a risk indica- Jarnberg PO: Fluoride pharmacokinetics
clearance of fluoride in a steady state tor of dental fluorosis in children resid- during acid-base balance changes in
condition in man: influence of urinary ing in areas with 0.5 and 2.5 mg fluoride man. Eur J Clin Pharmacol 1980;18:
flow and pH changes by diet. Acta Phar- per litre in drinking water. Caries Res 189–194.
macol Toxicol (Copenh) 1982;50: 1999;33:267–274.

Fluoride Metabolism 35
69 Awadia AK, Haugejorden O, Bjorvatn K, 74 Khandare AL, Harikumar R, Sivakumar 81 Everett ET, McHenry MA, Reynolds N,
Birkeland JM: Vegetarianism and dental B: Severe bone deformities in young Eggertsson H, Sullivan J, Kantmann
fluorosis among children in a high fluo- children from vitamin D deficiency and C, Martinez-Mier EA, Warrick JM,
ride area of northern Tanzania. Int J Pae- fluorosis in Bihar-India. Calcif Tissue Int Stookey GK: Dental fluorosis: variability
diatr Dent 1999;9:3–11. 2005;76:412–418. among different inbred mouse strains.
70 Khandare AL, Rao GS, Lakshmaiah N: 75 Lakshmaiah N, Srikantia SG: Fluoride J Dent Res 2002;81:794–798.
Effect of tamarind ingestion on fluoride retention in humans on sorghum and 82 Mousny M, Banse X, Wise L, Everett
excretion in humans. Eur J Clin Nutr rice based diets. Indian J Med Res 1977; ET, Hancock R, Vieth R, Devogelaer
2002;56:82–85. 65:543–548. JP, Grynpas MD: The genetic influence
71 Khandare AL, Kumar PU, Shanker RG, 76 Russell AL: Dental fluorosis in Grand on bone susceptibility to fluoride. Bone
Venkaiah K, Lakshmaiah N: Additional Rapids during the seventeenth year of 2006;39:1283–1289.
beneficial effect of tamarind ingestion fluoridation. J Am Dent Assoc 1962; 83 Carvalho JG, Leite AL, Yan D, Everett ET,
over defluoridated water supply to ado- 65:608–612. Whitford GM, Buzalaf MA: Influence of
lescent boys in a fluorotic area. Nutrition 77 Butler WJ, Segreto V, Collins E: Preva- genetic background on fluoride metabo-
2004;20:433–436. lence of dental mottling in school-aged lism in mice. J Dent Res 2009;88:
72 Chen YX, Lin MQ, Xiao YD, Gan WM, lifetime residents of 16 Texas communi- 1054–1058.
Min D, Chen C: Nutrition survey in den- ties. Am J Public Health 1985;75: 84 Everett ET, Yan D, Weaver M, Liu L,
tal fluorosis-afflicted areas. Fluoride 1408–1412. Foroud T, Martinez-Mier EA: Detection
1997;30:77–80. 78 Williams JE, Zwemer JD: Community of dental fluorosis-associated quantita-
73 Bhargavi V, Khandare AL, Venkaiah K, water fluoride levels, preschool dietary tive trait Loci on mouse chromosomes 2
Sarojini G: Mineral content of water and patterns, and the occurrence of fluoride and 11. Cells Tissues Organs 2009;189:
food in fluorotic villages and prevalence enamel opacities. J Public Health Dent 212–218.
of dental fluorosis. Biol Trace Elem Res 1990;50:276–281. 85 Huang H, Ba Y, Cui L, et al.: COL1A2
2004;100:195–203. 79 Anand JK, Roberts JT: Chronic fluorine gene polymorphisms (Pvu II and Rsa I),
poisoning in man: a review of literature serum calciotropic hormone levels, and
in English (1946–1989) and indications dental fluorosis. Community Dent Oral
for research. Biomed Pharmacother Epidemiol 2008;36:517–522.
80 Polzik EV, Zinger VE, Valova GA,
Kazantsev VS, Yakusheva MY: A method
for estimating individual predisposition
to occupational fluorosis. Fluoride 1994;

Marília Afonso Rabelo Buzalaf

Department of Biological Sciences
Bauru Dental School, University of São Paulo
Al. Octávio Pinheiro Brisolla, 9–75
Bauru-SP, 17012–901 (Brazil)
Tel. +55 14 3235 8346, E-Mail

36 Buzalaf · Whitford
Fluoride Intake, Metabolism and Toxicity
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 37–51

Contemporary Biological Markers of Exposure

to Fluoride
Andrew John Rugg-Gunna ⭈ Alberto Enrique Villab ⭈ Marília Afonso Rabelo Buzalafc
aNewcastle University, Newcastle upon Tyne, UK; bInstitute of Nutrition and Food Technology, University of Chile,
Santiago, Chile; cDepartment of Biological Sciences, Bauru Dental School, University of São Paulo, Bauru, Brazil

Abstract have some ability to predict fluoride exposure, present

Contemporary biological markers assess present, or very data are insufficient to recommend utilizing fluoride con-
recent, exposure to fluoride: fluoride concentrations in centrations in these body fluids as biomarkers of contem-
blood, bone surface, saliva, milk, sweat and urine have porary fluoride exposure for individuals. Daily fluoride
been considered. A number of studies relating fluoride excretion in urine can be considered a useful biomarker
concentration in plasma to fluoride dose have been of contemporary fluoride exposure for groups of people,
published, but at present there are insufficient data on and normal values have been published.
plasma fluoride concentrations across various age groups Copyright © 2011 S. Karger AG, Basel
to determine the ‘usual’ concentrations. Although bone
contains 99% of the body burden of fluoride, attention The concept of biological markers of fluoride ex-
has focused on the bone surface as a potential marker posure came to prominence in the 1994 Technical
of contemporary fluoride exposure. From rather limited Report on Fluorides and Oral Health [1]. The
data, the ratio surface-to-interior concentration of fluo- WHO stated that ‘a fluoride biomarker is of
ride may be preferred to whole bone fluoride concen- value primarily for identifying and monitoring
tration. Fluoride concentrations in the parotid and sub- deficient or excessive intakes of biologically avail-
mandibular/sublingual ductal saliva follow the plasma able fluoride’, and acknowledged a workshop on
fluoride concentration, although at a lower concentra- this topic, organized by the US National Institute
tion. At present, there are insufficient data to establish a of Dental Research, the year before [2]. In its re-
normal range of fluoride concentrations in ductal saliva port in 2002, the UK Medical Research Council
as a basis for recommending saliva as a marker of fluoride [3] identified biomarkers of fluoride exposure as
exposure. Sweat and human milk are unsuitable as mark- a research priority.
ers of fluoride exposure. A proportion of ingested fluoride During the last two decades, there has been a
is excreted in urine. Plots of daily urinary fluoride excre- rapid expansion in the availability and use of bio-
tion against total daily fluoride intake suggest that daily markers in health care, such that they now occupy
urinary fluoride excretion is suitable for predicting fluo- a central position in the armamentarium of the
ride intake for groups of people, but not for individuals. clinician for screening, diagnosis and manage-
While fluoride concentrations in plasma, saliva and urine ment of disease [4]. For example, there is now
extensive literature on biomarkers of bone forma- concentration, and fluoride concentration in nails
tion [5], bone turnover [6] and bone resorption (finger and toe) and hair might reflect past blood
[7]. This expansion in bioinformatics has been fluoride concentration and the body burden of flu-
due, in no small part, to rapid developments in oride. Fluoride toxicity can be ‘acute’ or ‘chronic’
chip technology in routine biological analysis [8]. – the former involving a very recent high or very
However, people using biomarkers should be able high dose of fluoride, whilst the later might occur
to interpret results appropriately; for example, after modest yet excessive ingestion over a longer
they should be able to answer the question ‘Is it period of time. Biomarkers are needed for both
(the value obtained) normal?’ The ideal method situations – present exposure being assessed by
should be accurate, precise, sensitive and spe- ‘contemporary’ biomarkers, while more chronic
cific [9]. This chapter and the chapter by Pessan fluoride exposure might be assessed by ‘recent’ or
and Buzalaf [this vol., pp. 52–65] will review the ‘historic’ biomarkers. Contemporary biomarkers
literature on biological markers of fluoride expo- might be fluoride concentrations in blood, bone
sure. As described below, this chapter will con- surface, saliva, milk, sweat and urine, while his-
sider ‘contemporary’ biomarkers and the other toric biomarkers might include bone, teeth, nail
will consider ‘historical and recent’ biomarkers of and hair. This chapter will consider contemporary
fluoride exposure. markers.
Almost all fluoride entering the body is ab-
sorbed via the intestinal tract [Buzalaf and
Whitford, this vol., pp. 20–36]. While most foods Early Investigations
contain fluoride, fluoride is also ingested from
fluoride-containing vehicles designed to control The presence of fluoride in ‘notable proportions’
the development of dental caries. A proportion in the blood of humans and other animals was re-
of fluoride entering the body is excreted in urine, ported by Nickles in 1856 [10]. In 1888, Tammann
and nearly all of the retained fluoride accumu- recorded fluoride in the milk and blood of a cow
lates in calcified tissues, mainly bone. To maxi- and, in 1913, Gautier and Clausmann recorded
mize health benefits, there has to be a balance the fluoride content of a large number of animal
between too little fluoride (with increased risk of tissues including blood, milk and urine [10]. The
dental caries and its sequelae) and too much fluo- first experimental study of fluoride deposition in
ride (with increased risk of fluorosis). The ‘thera- soft tissues was reported in 1891 by Brandl and
peutic ratio’ is relatively low – the space between Tappeiner, who gave dogs daily food additives of
the two disbenefits (insufficiency and excess) is 0.1–1.0 g sodium fluoride [10]. It is now known
small. It is, therefore, very desirable to know what that the fluoride concentrations published were
the body burden of fluoride is, in order to assess far too high due to inaccurate analytical methods
the risk/benefit ratio and maximize benefit and [11], but these studies indicate that early inves-
minimize disbenefit. tigators were aware of the possibility of fluoride
Buzalaf and Whitford [this vol., pp. 20–36] de- toxicity. This lack of accuracy in fluoride analy-
scribed how ingested fluoride was absorbed into sis was particularly important for tissues such as
the bloodstream, readily entering calcified tissue blood and saliva where the fluoride concentra-
and excreted in urine. Thus, the body burden of tion is low compared with mineralized tissues
fluoride might be estimated by examining fluo- or urine [11]. Awareness grew during the 1930s
ride concentrations in blood, bone, teeth and and 1940s that water, along with sprayed pesti-
urine. In addition, fluoride concentrations in sa- cides and chemical contamination, were the prin-
liva, milk and sweat might reflect blood fluoride cipal potential sources of fluoride intake in man.

38 Rugg-Gunn · Villa · Buzalaf

Controlled fluoride balance studies were con- 1.26 μmol/l). Most of these studies reported the
ducted and, in a good summary, McClure et al. effect on plasma fluoride concentration of ingest-
[12] stated that ‘chronic cumulative toxic effects ing a dose of fluoride: the maximum plasma fluo-
of fluorine may be predicted to a large extent by ride concentration was usually reported and these
the quantity of fluorine which the body regularly are also given in table 1. The effect of fluoride dose
retains’. They noted ‘that fluorine concentration of on plasma fluoride concentration is given in fig-
spot urine specimens has a strikingly close corre- ure 1. The data points are taken from the mean
lation with the fluorine concentration in domestic values given in table 1: readers should be aware
water, through the range of 0.5 to 4.5 ppm fluo- that this limits the interpretation of the plot, since
rine in drinking water’. A crucial development oc- the studies varied considerably in location, num-
curred in the mid-1960s with the availability of ber and age of the subjects, and background fluo-
the specific fluoride ion electrode which allowed ride exposure.
accurate measurement of low concentrations of When interpreting plasma fluoride concentra-
fluoride [11], although considerable care is need- tion information, it is important to be aware that
ed for concentrations in the region of 0.01 mg/l several factors, independent of fluoride dose, in-
(approximately 0.5 μmol/l) [13]. fluence the concentration value. These include:
site of blood collection [13], age [13], acid-base
balance [13], altitude [13], hematocrit [13] and
Blood genetic background [27]. The effects of circadian
rhythm [13, 25] and hormones [13, 25] are rather
It is usual to report fluoride concentration in plas- discrete. Of these, site of blood collection, age and
ma, rather than in whole blood or serum. Blood hematocrit have the most influence. It is advisable
cells contain about half the fluoride concentra- to collect venous or capillary blood [13]. Plasma
tion recorded in plasma. Arguments for select- fluoride concentration increases with age, and
ing plasma include: (1) plasma concentration this is likely to be a reflection of increasing bone
establishes interstitial and intracellular fluoride fluoride concentration with age [13]. Hematocrit
concentrations in soft tissues, and (2) plasma is values are likely to be lower in females than males
the fluid from which fluoride is filtered into the [13].
nephron [13]. While both ionic and non-ionic
fluoride forms exist in plasma, the ionic form is
of far greater significance [13], and it is detectable Bone Surface
by the ion-specific electrode.
Plasma fluoride concentration returns to the The human skeleton, which weighs approximate-
resting value about 3–6 h after ingestion of a small ly 3–4 kg, consists of 60% inorganic material and
fluoride dose; the half-life is about 30 min [13]. 30% organic matrix (95% type I collagen). The
Fasting (baseline or resting) values are, there- mineral phase consists chiefly of hydroxyapatite.
fore, usually determined after overnight fasting. Bone is metabolically active and about 5–10% of
There have been a number of reports of fasting existing bone in adults is replaced through mod-
plasma fluoride concentrations in humans [13– eling each year. The skeleton consists of two types
26], and these are listed in table 1, excluding de- of bone: cortical bone (sometimes called compact
terminations made before the availability of the bone) which comprises about 80% of the skeletal
ion-specific electrode. For less than optimal water mass, and trabecular bone (also called spongy or
fluoride concentrations, the resting fluoride con- cancellous bone) which forms the internal scaf-
centrations ranged from 9.3 to 24.0 ng/ml (0.49– folding of bone. Peak bone mass is achieved at

Contemporary Biomarkers 39
Table 1. Studies of plasma fluoride concentrations in human subjects

Study Subjects Resting After fluoride dose (maximum)

(first author)
n age, years water ng/ml μmol/l fluoride ng/ml μmol/l
fluoride, ppm dose, mg

Ekstrand [14] 1 27 0.25 10.3 0.54 10 300 15.8

Ekstrand [15] 5 24–28 0.25 10.0 0.50

5 27–56 1.20 20.01 1.00

5 10–38 9.60 35.01 1.84

Ekstrand [16] 5 27–36 n.a. 13.3 0.701 1.5 75 4

Oliveby [17] 5 26–38 0.2 9.3 0.492 1.0 51.1 2.692

Oliveby [18] 5 26–38 0.2 9.5 0.503 1.0 52.3 2.753

Oliveby [19] 5 26–38 0.2 12.4 0.653 1.0 52.6 2.773

Whitford [13] 5 adults n.a. 12.7 0.67 10.0 289 15.2

Whitford [20] 17 5–10 n.a. 16.9 0.89

Levy [21] 15 2–6 0.6–0.8 19.0 1.00

15 2–6 0.1–0.2 24.0 1.26

Maguire [22] 20 20–35 0.02 19.8 1.04 0.5 32.34 1.74

Cardoso [23] 5 25–35 0.035 9.7 0.51

5 25–35 0.70 6.8 0.366

5 25–35 0.30 10.5 0.557

Whitford [24] 5 24–32 0.85 17.3 0.91 0.33 28.38 1.498

5 24–32 0.85 20.0 1.05 2.73 137.28 7.228

Cardoso [25] 5 27–33 low F9 10.0 0.53

Buzalaf [26] 4 19–29 0.6–0.810 21.0 1.11 2.0011 106.0 5.58

4 19–29 0.6–0.810 22.0 1.16 2.0012 100.0 5.26

1 Estimated from graph.

2 For study 1 and low flow rate.
3 For unstimulated saliva.
4 For subjects receiving naturally fluoridated soft water.
5 Subjects drank bottled water.
6 Low fluoride intake from dentifrice.
7 High fluoride intake from dentifrice.
8 For subjects receiving naturally fluoridated water.
9 Subjects drank bottled water with low fluoride content.
10 Personal communication from Buzalaf.
11 Fluoride administered as sodium fluoride.
12 Fluoride administered as disodium monofluorophosphate.

40 Rugg-Gunn · Villa · Buzalaf

bone may reflect contemporary fluoride expo-
sure, while mature bone reflects historic fluoride
Concentration (ng/ml)

There is a steady-state relationship between the
fluoride concentrations in the extracellular fluids
and the hydration shells of bone crystallites [13].
According to this concept, there is a net transfer
of fluoride from plasma to the hydration shells
0 2 4 6 8 10 12 when the plasma concentration is rising and in
Dose (mg)
the opposite direction when the plasma concen-
tration is falling. In other words, the surface bone
Fig. 1. Plot of the maximum concentration of fluoride compartment is considered to be small but rap-
in plasma after ingestion against the dose. Data are the
mean values obtained from the 11 studies listed in table
idly exchangeable when compared with the bulk,
1 (right-most columns). Since these data are mean values, virtually non-exchangeable, inner compartment
not values for individuals, a regression line is not given. [30]. Thus, the initial uptake of ingested fluoride
by bone occurs on the endosteal and periosteal
surfaces which also have the highest concentra-
tions in rats [31–33] and humans [34]. In this
about age 38 years, with 90% of peak bone mass sense, bone surface fluoride concentrations have
being achieved by 18 years. Genetic and lifestyle been shown to increase in rats in the first hours
factors influence bone mineral accrual during or even days following an acute high dose of flu-
growth. Bone mineral density is usually measured oride [33, 35, 36], and could be useful to clarify
by dual energy X-ray absorption, and the rate of the causa mortis when fluoride is suspected as the
bone remodeling is assessed by biomarkers (vide cause of death. However, the suitability of this bio-
supra). marker to humans has not been evaluated so far.
A proportion of ingested fluoride is retained It is important to consider that bone surface fluo-
in the body. This has been shown in fluoride bal- ride concentrations are not expected to be homo-
ance studies where, in addition to urinary fluoride geneous among humans in defined populations
excretion, fecal fluoride was measured in infants or regions. Bone fluoride concentrations increase
[28] and adults [29]. At least 99% of the body with increase of past fluoride intake [37] and age
burden of fluoride is associated with the skeleton [38–40]. They also seem to be influenced by acid-
[13]. Incorporation of fluoride into bone occurs base balance (decrease in cases of metabolic al-
in several stages: first, exchange of ions in the kalosis and increase in high altitude) and genetic
loosely integrated sheath surrounding the bone background [27, 41]. Because of these variables,
crystallite; second, incorporation into the hydra- surface bone fluoride concentrations per se may
tion shell; and, finally, migration of the fluoride not be the best indicators of acute exposure to le-
ion into the crystal structure during recrystalliza- thal amounts of fluoride and the ratio surface-to-
tion. The first stage is rapid, occurring within 60 interior concentrations should be preferred.
min of intravenous injection of fluoride. Uptake The choice of site for bone sampling needs
by bone is continuous, unlike in other (soft) tis- careful consideration, since there is much varia-
sues where a plateau is quickly reached [13]. In tion in fluoride concentration between sites [42].
addition, the crystallites of developing bone are Since cancellous bone is much more biological-
small in size (compared with mature bone), large ly active than compact bone, the highest values
in number and heavily hydrated [13]. Thus, new might be obtained in the metaphysis cancellous

Contemporary Biomarkers 41
bone (>9,000 mg/kg in ash), while within a bone were not significantly related with plasma fluoride
such as the rib, cancellous bone might have a fluo- concentrations of 5- to 10-year-old children (n =
ride concentration of 3,500 mg/kg (in ash) com- 17) while parotid fluoride concentrations were
pared with 1,700 mg/kg (in ash) in rib compact (by a proportionality constant of 0.8). A further
bone [38]. The suitability of bone as a marker of comparison of fluoride concentrations in parotid
historic fluoride exposure is considered further in and submandibular ductal saliva was published
the chapter by Pessan and Buzalaf [this vol., pp. by Twetman et al. [50]; the subjects were 12 young
52–65]. adolescent girls. Fluoride concentrations (stimu-
lated flow) were higher in submandibular saliva
(0.55 μmol/l, 10.5 ng/ml) than in parotid saliva
Saliva (0.25 μmol/l, 4.8 ng/ml), and concentrations re-
mained higher in submandibular duct saliva after
Knowledge of the fluoride concentration in saliva ingestion of fluoride. Thus, fluoride concentra-
has been considered important, principally be- tions appear to be slightly higher in submandibu-
cause it influences plaque fluoride concentration lar saliva than in parotid saliva.
strongly and hence the control of dental caries. The series of studies by Oliveby et al. [17–19]
There are many examples of the fluoride concen- and the study by Whitford et al. [20] clearly dem-
tration in whole saliva increasing in response to onstrated the relationship between fluoride con-
the provision of caries-prevention agents [43–48]. centrations in plasma and parotid or subman-
Because whole saliva (usually collected by drool- dibular ductal saliva. In agreement with previous
ing into a container) is contaminated with fluo- research [51], Oliveby et al. [17, 18] reported that
ride from food and therapeutic agents, saliva has fluoride concentrations were lower in ductal sa-
been collected from parotid and submandibular/ liva than in plasma. The ratios of saliva to plasma
sublingual ducts, by specially constructed devic- fluoride concentrations, under resting conditions,
es. The fluoride concentration in ductal saliva has were 0.32 to 0.55 for parotid saliva [17] and 0.61 to
been studied in relation to: (1) plasma fluoride 0.88 for submandibular saliva [18]. In contrast, in
concentration, (2) salivary stimulation and flow the same series of experiments, the ratio for fluo-
rate, and (3) fluoride ingestion. ride concentrations in whole saliva (as collected)
Fluoride concentrations in whole saliva and and plasma was 1.10 [19], suggesting that whole
ductal saliva were compared by Yao and Gron saliva had acquired fluoride from the oral envi-
[49]. The concentration in whole saliva (mean of ronment. Whitford et al. [20] also found the fluo-
6 subjects) was 15.8 ng/ml (0.83 μmol/l), while af- ride concentration in whole saliva much higher
ter centrifuging, the concentration was 9.3 ng/ml than in parotid ductal saliva. The fluoride concen-
(0.49 μmol/l). The corresponding values for pa- tration in ductal parotid saliva was strongly cor-
rotid and submandibular ductal saliva were 6.5 related with plasma fluoride concentration, with
ng/ml (0.34 μmol/l) and 6.3 ng/ml (0.33 μmol/l), a proportionality constant of 0.80 for the saliva/
respectively. In a series of three experiments, al- plasma relation. It follows that, for comparisons
though seemingly using the same subjects, Oliveby with plasma fluoride concentration and therefore
et al. [17–19] recorded fluoride concentrations in the body burden of fluoride, it is preferable to ex-
parotid and submandibular ductal saliva and in amine ductal saliva rather than whole saliva.
whole saliva (as collected) of 0.17 μmol/l (3.23 ng/ After ingestion of fluoride, there is a close re-
ml), 0.46 μmol/l (8.7 ng/ml) and 0.71 μmol/l (13.5 lationship between the rise in fluoride concentra-
ng/ml), respectively. Also in the study by Whitford tion in plasma and the rise in fluoride concentra-
et al. [20], fluoride concentrations in whole saliva tion in parotid and submandibular ductal salivas

42 Rugg-Gunn · Villa · Buzalaf

[17, 18, 51]. In the study of Ekstrand [51], where et al. [17, 18] also found fluoride concentrations
3 young adults received 3 mg fluoride, both plas- in parotid [17] and submandibular ductal saliva
ma and salivary (parotid) fluoride concentrations [18] were little affected by stimulation of salivary
peaked at about 30 min and remained elevated for flow, for up to 2 h after fluoride ingestion, while
8 h. Salivary fluoride concentration followed that Twetman et al. [50] reported slightly higher fluo-
of plasma closely throughout the 8 h, with an av- ride concentrations in unstimulated ductal saliva,
erage saliva/plasma ratio of 0.63. Commenting on before and after fluoride ingestion.
the stability of this ratio during the experiment, Thus, in conclusion, as a marker of plasma
the author commented ‘that saliva [fluoride] may fluoride concentration, it would appear that sub-
be used as a substitute for blood sampling in stud- mandibular/sublingual duct saliva is preferable
ies concerning the pharmacokinetics of fluoride’. to parotid duct saliva, and both are preferable to
Rather similar results were reported by Oliveby whole saliva, because (1) fluoride concentrations
et al. [17, 18] and Whitford et al. [20] for parot- are higher, and (2) saliva to plasma ratios are more
id and submandibular saliva. While the stability stable after fluoride ingestion. However, it should
of the saliva/plasma ratio was good for subman- be noted that collection of submandibular/sublin-
dibular saliva (approximately 0.4–0.6), it was less gual saliva is technically more difficult [18].
good for parotid saliva (approximately 0.30–0.65)
[17, 18]. Whitford [13] reported that the saliva to
plasma ratio for fluoride concentration, after in- Sweat
gestion of 10 mg fluoride by adults, was higher
for submandibular (about 0.88) than for parotid Early work suggested that the concentration of
(about 0.78) ductal saliva. This was very close to fluoride in sweat was substantial: McClure et al.
the value of 0.80 reported by Whitford et al. [20] [12] reported 0.3–1.8 mg/l, Crosby and Shepherd
for children. [53] 0.3–0.9 mg/l, and Largent [54] 0.3–0.9 mg/l.
The times of the peak concentrations of fluo- However, estimates made after the introduction
ride in plasma and ductal saliva, after fluoride in- of the ion-specific electrode were very much low-
gestion, have been studied. Ekstrand [51] found er. Even after ingesting 10 mg fluoride, which
that both peaked at around 30 min after a 3 mg raised plasma concentration to 0.24 mg/l (12.6
dose. In slight contrast, Oliveby et al. [17, 18] μmol/l), the concentration in sweat was only
found: (1) that peaks were not reached until about about 0.05 mg/l (2.6 μmol/l) [11]. In a brief sum-
40 min after ingestion of 1 mg fluoride, and (2) mary, Whitford [13] stated that fluoride concen-
a delay of about 10–15 min in the fluoride con- trations in sweat were similar to concentrations in
centration peak in parotid saliva compared with plasma (1–3 μmol/l; 0.019–0.057 mg/l; 19–57 ng/
plasma, although the results for submandibular ml). Presently, issues of collection, including con-
saliva were variable. tamination, and lack of supporting data, preclude
The concentrations of many constituents of sa- the use of sweat as a viable marker of contempo-
liva change markedly after stimulation of salivary rary fluoride exposure.
flow [13]. However, a number of researchers have
shown that fluoride concentration is remarkably
stable. Shannon et al. [52] found that the fluoride Human Milk
concentration in parotid saliva fell from 22 ng/ml
(1.16 μmol/l) before stimulation to 17 ng/ml (0.89 Backer Dirks et al. [55] estimated the fluoride
μmol/l) after stimulation, reducing only slightly concentration in human milk of mothers living
as the intensity of stimulation increased. Oliveby in fluoridated or non-fluoridated communities,

Contemporary Biomarkers 43
using both gas-liquid chromatography and the from an assessment of that property, fluoride
ion-specific electrode. The total fluoride concen- exposure might be reliably inferred. Another
trations in the milk of mothers in the two areas, important point that has to be considered is
respectively, were 52 and 46 ng/ml (2.7 and 2.4 whether the fluoride biomarker can be used on
μmol/l), while the ionic fluoride concentrations an individual or group (e.g. community) basis
were 8 and 4 ng/ml (0.42 and 0.21 μmol/l). A (vide infra).
more recent estimate of the fluoride concentra- Based on pharmacokinetic findings, urinary
tion in milk from 57 lactating Turkish women was fluoride is considered a contemporary biomark-
19 ng/ml (1.0 μmol/l) [56]. The only direct com- er of fluoride exposure, since varying propor-
parison of the fluoride concentrations in human tions of a given fluoride dose are completely
plasma and milk was published by Ekstrand et al. excreted with the urine in less than 24 h in chil-
[16]. Five mothers were given 1.5 mg fluoride af- dren and adults [13, 15]. It is important that the
ter fasting, and fluoride concentrations in plasma selected biomarker is clearly related to the fluo-
and milk were followed for 2 h. While the plasma ride exposure. Regarding urine as a possible bio-
fluoride concentration rose from about 0.6 μmol/l marker, early studies [44, 60, 61] attempted to es-
(11 ng/ml) to a peak of about 5 μmol/l (95 ng/ml) tablish such a relationship by comparing urinary
after 30 min, there was virtually no change in the fluoride concentration and fluoride exposure.
fluoride concentration in milk which remained at In some studies, exposure was estimated semi-
about 0.26 μmol/l (4.9 ng/ml). In conclusion, on quantitatively, e.g. reporting the fluoride con-
the basis of these limited data, human milk would centration of the drinking water that the study
seem unsuitable for estimating the body burden subjects ingested [44, 60, 61]. In these studies,
of fluoride. no simple numerical relationship could be es-
tablished between fluoride exposure and uri-
nary fluoride concentration, and thus they are
Urine of limited use. Other studies have measured ex-
perimentally the amount of fluoride ingested
Fluoride concentrations in body fluids (e.g. urine, from drinking water, foods, other beverages and,
plasma, serum, saliva) are generally recognized as eventually, other fluoride sources such as dental
being the most suitable for evaluating short-term hygiene products [29, 62–65]. As discussed by
fluoride exposures or fluoride balance (intake mi- Villa et al. [66] a 24-hour urinary collection is the
nus excretion), while some earlier sources suggest minimal recommended period of time in order
that samples obtained from fasting persons may to obtain good estimations of the daily amount
be useful for estimating chronic fluoride intake or of fluoride excretion. The daily urinary fluoride
potential bone fluoride concentrations [57, 58]. excretion is the variable generally recommended
Examples of the association between estimated for the estimation of the daily fluoride exposure.
fluoride intakes (or mass-normalized intakes) The amount of excreted fluoride is easily ob-
and measured fluoride concentrations in urine, tained multiplying the 24-hour urinary volume
plasma, and serum for individuals and groups by its fluoride concentration [62, 67].
were shown in a recent review [table 2.16; 59].
In order to be considered a viable fluoride Factors Affecting Urinary Fluoride Excretion
biomarker, a relationship must be established Changes in chronic exposure to fluoride will
between some property (mass, concentration, tend to alter plasma and bone fluoride concen-
others) of the candidate biomarker and fluoride trations (as discussed in previous sections of this
exposure or intake over a period of time: thus, chapter), and a number of factors can modify

44 Rugg-Gunn · Villa · Buzalaf

Table 2. Observed range of values for 24-hour urinary fluoride excretion and estimated 24-hour fluoride exposure in
young children and adults according to different fluoridation conditions

Range of urinary fluoride excretion found, Range of predicted fluoride intake, Fluoridation condition
mg/24 h mg/24 h

Young children (≤6 years)

0.17–0.31 0.40–0.80 low-fluoridated areas1
0.31–0.50 0.80–1.34 optimally fluoridated areas2
>0.60 >1.63 higher than ‘optimal’3

Adults (18 to 50+ years)

1.00–1.40 1.31–2.05 low-fluoridated areas1
1.50–2.50 2.24–4.10 optimally fluoridated areas2
>3.00 >5.00 higher than ‘optimal’4

Data taken from Villa et al. [66], with permission (n = 212 young children and 269 adults).
1 Fluoride concentrations in drinking water ≤0.4 mg/l.
2 Fluoride concentrations in drinking water in the range 0.6–1.0 mg/l.
3 On a chronic basis, fluoride exposure might cause an objectionable prevalence of moderate and severe enamel

4 On a chronic basis, fluoride exposure might cause a preclinical stage of skeletal fluorosis [59].

fluoride pharmacokinetics, providing another glomerular filtration rates are reduced to around
way to change fluoride tissue concentrations 20% of normal, as measured via creatinine
[13]. Fluoride clearance tends to increase with clearance or serum creatinine concentrations
urinary pH [68]. One proposed mechanism for [58, 69, 70].
this is decreased reabsorption in the renal tu-
bule, since hydrogen fluoride (which is formed Urine as a Contemporary Fluoride Biomarker
at lower pH values) easily crosses cell mem- While the first reports of simultaneous measure-
branes, which are nearly impermeable to the ment of total daily fluoride intake (TDFI) and dai-
fluoride ion [68]. Thus, increasing urinary pH ly urinary fluoride excretion (DUFE) were pub-
tends to reduce fluoride retention [68]. As a re- lished many years ago [29, 71, 72], several studies
sult, fluoride retention might be affected by en- have been undertaken more recently in children
vironments or conditions that chronically affect [28, 62, 63, 65, 73, 74] and adults [64, 67]. A recent
urinary pH, including diet, drugs, altitude and paper [66] reviewed all of the available published
certain diseases, e.g. chronic obstructive pulmo- data on the simultaneous experimental assess-
nary disease [13]. ment of TDFI and DUFE using individuals’ data
Since the kidney is the major route of fluoride from each study, and the relationship between
excretion, it is not surprising that increased these variables was examined in order to assess
plasma and bone fluoride concentrations have the suitability of DUFE as a predictor (biomarker)
been observed in patients with kidney disease. of TDFI.
Plasma fluoride concentrations have also been Results obtained from nine independent stud-
demonstrated to be elevated in patients with ies (n = 212) carried out in young children (0.15–
severely compromised kidney function, where 7 years old) from six different Western countries,

Contemporary Biomarkers 45
2.0 10

Excretion (mg/day)

Excretion (mg/day)

1.0 5

0.5 2

0 0 2 4 6 8 10 12 14 16
0 1 2 3 4 5 –1
Intake (mg/day)
Intake (mg/day)

Fig. 3. Relationship between daily urinary fluoride ex-

Fig. 2. Relationship between daily urinary fluoride excre-
cretion and total daily fluoride intake for 269 data pairs
tion and total daily fluoride intake for 212 young children
from adults aged 18–75 years recorded in 8 studies in 2
aged 0.15–7 years recorded in 9 studies in 6 countries. The
countries. The full line is the best fit; the inner interrupted
full line is the best fit; the inner interrupted lines indicate
lines indicate the 95% CI of the regression, and the out-
the 95% CI of the regression, and the outer interrupted
er interrupted lines indicate the 95% prediction interval.
lines indicate the 95% prediction interval. Reproduced
Reproduced from Villa et al. [66], with permission.
from Villa et al. [66], with permission.

and six independent studies (n = 269) in adults Individual or Group Biomarker?

(18- to 75-years-old) from two American coun- Statistical analysis of the above-mentioned lin-
tries, were available for the analysis (fig. 2 and 3). ear relationships clearly show that the 95% pre-
Highly significant linear relationships were found diction intervals associated with the regression
between DUFE and TDFI for both children and lines do not suggest that DUFE is viable as a pre-
adults. The values of the intercepts and slopes of cise estimator of TDFI on an individual basis.
the best fit regressions for both age groups were However, the 95% CI bands do suggest DUFE
significantly different [66], indicating that the is appropriate when considering fluoride ex-
proportion of fluoride retained in children’s hard posure on a group (or community) basis [66].
tissues is higher than the corresponding values Thus, it can be concluded that, at this time, uri-
for adults, in agreement with previous knowl- nary fluoride excretion has a very limited value
edge. No differences due to gender were observed as a biomarker of individual fluoride exposure.
for both age groups. These results strongly sug- This situation is similar to that of other fluoride
gest that the daily urinary fluoride excretion is a biomarkers discussed in this chapter and that of
reasonably good biomarker of contemporary flu- Pessan and Buzalaf [this vol., pp. 52–65]. From
oride exposure. an epidemiological point of view, estimating

46 Rugg-Gunn · Villa · Buzalaf

the average (and 95% CI) daily fluoride expo- US Environmental Protection Agency [chapter
sure on a community basis might have some 3; 59]. In addition, it is important to take into
merit. When a 24-hour urinary fluoride excre- account that, among healthy subjects, the effect
tion study is performed on a relatively high (n of diet on their urinary pH might have a signifi-
≥20 according to general practice in most pub- cant effect on observed urinary fluoride excre-
lished reports on this type of study) number of tion values for a certain range of average fluoride
subjects, and accepting that the distribution of exposures [13].
frequencies of urinary excretion values is nearly
normal, the average and 95% CI values might be Fluoride Urinary Excretion: Normal or Observed
considered reliable estimators of 24-hour fluo- Range of Values?
ride exposure for the group or community from Fluoride is not homeostatically controlled [13].
which the sample was taken. This is especially This means that fluoride levels in body fluids
valid when such a group or community of indi- or tissues will show a range of values that will
viduals is under ‘customary’ fluoride intake con- be highly variable and dependent on short- and
ditions, i.e. when the different fluoride sources long-term fluoride exposure. Thus, a ‘normal’
remain stable over time. It might be argued that (as used in clinical chemistry) urinary fluo-
24-hour urine collection provides a ‘snap-shot’ ride excretion range of values cannot be estab-
of the fluoride exposure for the particular study lished. In the introductory section of this chap-
day, since each individual might present a value ter, the WHO statement on the usefulness of
that is different from the one he/she would have a fluoride biomarker said that ‘a fluoride bio-
presented the following day or the day before. marker is of value primarily for identifying and
However, using a statistical approach, it can be monitoring deficient or excessive intakes of
considered that most of the within- and inter- biologically available fluoride’ [1]. In this con-
individual variation over time would be virtu- text, a semi-quantitative use of fluoride urinary
ally cancelled out when average values are con- excretion as a biomarker was presented in an
sidered. Under the above-mentioned conditions, earlier monograph [75]. The latter publication
several urinary fluoride excretion studies have presents ranges of provisional standard values
been considered useful for evaluating the safety for daily fluoride urinary excretion in young
of fluoride-based systemic preventive caries pro- children in low and ‘optimally’ fluoridated areas
grams [75]. Alternatively, an indication of a high- [table 5; 75].
er than safe fluoride exposure (in terms of the Based on the available data, a quantitative es-
risk of an undesirable prevalence of enamel fluo- timation of fluoride exposure can now be pro-
rosis in children, or skeletal fluorosis in adults, as posed using fluoride urinary excretion values and
defined in table 2) can be also established from the numerical relationships between DUFE and
community studies of urinary fluoride excretion, TDFI [66]. Thus, for sets of observed ranges of
provided that updated guidelines on ‘expected’ values of 24-hour urinary fluoride excretion, the
average value ranges are available for different estimated daily fluoride exposure in young chil-
fluoride exposure conditions. Several health dren and adults under different fluoridation con-
conditions that cause fluoride renal clearance ditions can be obtained (table 2).
disturbances have been mentioned previously, The predicted values for 24 h fluoride expo-
but a detailed discussion of particularly vulner- sures in young children and adults (table 2) were
able subgroups of subjects is beyond the scope estimated using published numerical models [66]
of this section. For a thorough discussion of this and are presented here as a provisional guideline
subject, see a recently published review by the that shows the feasibility of using daily fluoride

Contemporary Biomarkers 47
urinary excretion as a biomarker of daily fluoride include bone, teeth, nail and hair. Contemporary
exposure. However, certain limitations apply: biomarkers have been discussed in this chapter.
first, the numerical values of predicted fluoride There have been a number of reports of fluo-
exposures appearing in table 2 have estimated ride concentration in plasma, both after fasting
95% CI of 10–15%; second, the available studies and after ingestion of known doses of fluoride.
from which the experimental data were taken in- Maximum concentration occurs at about 30 min
cluded subjects who consumed ‘westernized’ di- and resting values may not be reached until 3–6 h
ets [66]. Thus, for those communities where ‘non- after exposure. Apart from fluoride dose, plasma
westernized’ more vegetarian diets are consumed, fluoride concentration varies with site of blood
it is reasonable to assume that urinary pH values collection, age, acid-base balance, altitude, hema-
could be higher [13] and, consequently, the pro- tocrit and genetic background. At present, there
portion of fluoride excreted with the urine might are insufficient data on plasma fluoride concen-
be higher. Further studies on this latter issue are tration, for various age groups, to determine ‘nor-
needed. It is conceptually reasonable to point out mal’ plasma fluoride concentrations.
that different age subgroups of the two age groups Much of the unexcreted fluoride that is even-
(young children and adults) from which the pres- tually incorporated into bone is found within the
ent estimates were obtained, might present differ- bone crystal structure. Although there has been
ent numerical values for the relationship between some attention focused on bone surface as a po-
DUFE and TDFI, especially so among young chil- tential marker of contemporary fluoride exposure
dren when the different rates of bone formation and, in particular, the use of the ratio of surface-
of infants and 6-year-olds are considered, as dis- to-interior concentrations, there are insufficient
cussed in previous reports [13, 66, 76]. However, data for recommending the use of bone as a viable
the imprecision associated with the variability marker for estimating contemporary fluoride ex-
caused by inter-individual physiological differ- posure in humans.
ences might obscure these effects. Fluoride concentrations in parotid and sub-
mandibular/sublingual ductal saliva follow plas-
ma fluoride concentration, although at a lower
Conclusion concentration. Submandibular saliva may have
advantages over parotid saliva, although its col-
Interest in the identification of viable and accurate lection is more difficult. At present, there are in-
biomarkers for fluoride exposure has increased sufficient data to establish a normal range of fluo-
significantly over the past several decades. This ride concentrations in ductal saliva as a basis for
is partly due to the increasing recognition that recommending saliva as a marker of fluoride ex-
the therapeutic ratio for fluoride is relatively low posure. Sweat and human milk are unsuitable as
and the need to avoid disbenefits, partly because biomarkers of fluoride exposure.
of technical advances in measurement, and partly A proportion of ingested fluoride is excreted
because of the expansion in knowledge of fluo- in urine. This proportion is influenced by age,
ride metabolism. Fluoride toxicity can be acute or urinary pH, and the several conditions that af-
chronic, and biomarkers are needed for assessing fect urinary pH. Plots of daily urinary fluoride
both situations. Contemporary biomarkers as- excretion against total daily fluoride intake, for
sess present (acute) exposure and might include young children and adults separately, reveal dif-
fluoride concentration in blood, bone surface, sa- ferent slopes of the regression lines, indicating
liva, milk, sweat and urine, while past exposure the relationship is different in the two age groups.
is assessed by historic biomarkers, which might The plots also suggest that daily urinary fluoride

48 Rugg-Gunn · Villa · Buzalaf

excretion is suitable for predicting fluoride intake biomarker of contemporary fluoride exposure for
for groups of people, but not for individuals. groups of people, and normal values have been
While fluoride concentrations in plasma, sa- published.
liva and urine have some ability to predict fluo- Future research on biomarkers for contempo-
ride exposure, data are, at present, insufficient rary fluoride exposure should focus on urine, sa-
to recommend fluoride concentration in these liva, plasma and bone, listed in order of impor-
body fluids as viable biomarkers of contempo- tance. Areas of uncertainty have been indicated
rary fluoride exposure for individuals. Fluoride within the relevant sub-sections.
concentration in urine can be considered a useful

1 WHO: Fluorides and oral health. Techni- 8 Jones R: Using intelligent systems in 17 Oliveby A, Lagerlof F, Ekstrand J, Dawes
cal Report Series 846. Geneva, WHO, clinical decision support; in Trull AK, C: Influence of flow rate, pH and plasma
1994. Demers LM, Holt DW, Johnston A, fluoride concentrations on fluoride con-
2 Selwitz RH: Introduction to the work- Tredger JM, Price CP (eds): Biomarkers centration in human parotid saliva. Arch
shop on methods for assessing fluoride of Disease: An Evidence-Based Oral Biol 1989;34:191–194.
accumulation and effects in the body. Approach. Cambridge, Cambridge Uni- 18 Oliveby A, Lagerlof F, Ekstrand J, Dawes
Adv Dent Res 1994;8:3–4. versity Press, 2002, pp 32–44. C: Studies on fluoride concentrations in
3 Medical Research Council: Water fluori- 9 Marshall WJ, Bangert SK: Clinical human submandibular/sublingual saliva
dation and health (working group Chemistry, ed 6. Edinburgh, Mosby, and their relation to flow rate and
report). London, Medical Research 2008, pp 1–13. plasma fluoride levels. J Dent Res 1989;
Council, 2002. 10 Armstrong WD, Gedalia I, Singer L, 68:146–149.
4 Trull AK, Demers LM, Holt DW, John- Weatherell JA, Weidmann SM: Distribu- 19 Oliveby A, Lagerlof F, Ekstrand J, Dawes
ston A, Tredger JM, Price CP: Biomark- tion of fluorides; in WHO (ed): Fluo- C: Studies on fluoride excretion in
ers of Disease: An Evidence-Based rides and Human Health. Geneva, WHO, human whole saliva and its relation to
Approach. Cambridge, Cambridge Uni- 1970, pp 93–139. flow rate and plasma fluoride levels.
versity Press, 2002. 11 Murray JJ, Rugg-Gunn AJ, Jenkins GN: Caries Res 1989;23:243–246.
5 Risteli J, Kauppila S, Jukkola A, Marjoni- Fluorides in caries prevention, ed 3. 20 Whitford GM, Thomas JE, Adair SM:
emi E, Melkko J, Risteli L: Biomarkers of Oxford, Butterworth-Heinemann, 1991, Fluoride in whole saliva, parotid ductal
bone formation; in Trull AK, Demers pp 262–294. saliva and plasma in children. Arch Oral
LM, Holt DW, Johnston A, Tredger JM, 12 McClure FJ, Mitchell HH, Hamilton TS, Biol 1999;44:785–788.
Price CP (eds): Biomarkers of Disease: Kinser CA: Balances of fluorine ingested 21 Levy FM, Bastos JR, Buzalaf MA: Nails
An Evidence-Based Approach. Cam- from various sources in food and water as biomarkers of fluoride in children of
bridge, Cambridge University Press, by five young men. J Ind Hyg Toxicol fluoridated communities. J Dent Child
2002, pp 115–121. 1945;27:159–170. 2004;71:121–125.
6 Eastell R, Hart S: Bone turnover markers 13 Whitford GM: The Metabolism and Tox- 22 Maguire A, Zohouri FV, Mathers JC,
in clinical practice; in Trull AK, Demers icology of Fluoride, ed 2. Basel, Karger, Steen IN, Hindmarsh PN, Moynihan PJ:
LM, Holt DW, Johnston A, Tredger JM, 1996. Bioavailability of fluoride in drinking
Price CP (eds): Biomarkers of Disease: 14 Ekstrand J, Alvan G, Boreus LO, Norlin water: a human experiment. J Dent Res
An Evidence-Based Approach. Cam- A: Pharmacokinetics of fluoride in man 2005;84:989–993.
bridge, Cambridge University Press, after single and multiple oral doses. 23 Cardoso VE, Whitford GM, Buzalaf MA:
2002, pp 99–114. Europ J Clin Pharmacol 1977;12: Relationship between daily fluoride
7 Robins SP: Biochemical markers of bone 311–317. intake from diet and the use of dentrifice
resorption; in Trull AK, Demers LM, 15 Ekstrand J: Relationship between fluo- and human plasma fluoride concentra-
Holt DW, Johnston A, Tredger JM, Price ride in the drinking water and the tions. Arch Oral Biol 2006;51:552–557.
CP (eds): Biomarkers of Disease: An plasma fluoride concentration in man. 24 Whitford GM, Sampaio FC, Pinto CS,
Evidence-Based Approach. Cambridge, Caries Res 1978;12:123–127. Maria AG, Cardoso VES, Buzalaf MAR:
Cambridge University Press, 2002, pp 16 Ekstrand J, Boreus LO, de Chateau P: No Pharmacokinetics of ingested fluoride:
122–132. evidence of transfer of fluoride from lack of effect of chemical compound.
plasma to breast milk. BMJ 1981;283: Arch Oral Biol 2008;53:1037–1041.

Contemporary Biomarkers 49
25 Cardoso VES, Whitford GM, Aoyama H, 37 Charen J, Taves DR, Stamm JW, Parkins 49 Yao K, Gron P: Fluoride concentrations
Buzalaf MAR: Daily variations in human FM: Bone fluoride concentration associ- in duct saliva and in whole saliva. Caries
plasma fluoride concentrations. J Fluo- ated with fluoridated drinking water. Res 1970;4:321–331.
rine Chem 2008;129:1193–1198. Calcif Tissue Int 1979;27:95–99. 50 Twetman S, Nederfors, Petersson LG:
26 Buzalaf MAR, Leite AL, Carvalho NTA, 38 Weidmann SM, Weatherell JA: Distribu- Fluoride concentration in whole saliva
Rodrigues MHC, Takamori ER, Nicolielo tion in hard tissues; in WHO (ed): Fluo- and separate gland secretions in school-
DB, Levy FM, Cardoso VES: Bioavail- rides and Human Health. Geneva, WHO, children after intake of fluoridated milk.
ability of fluoride administered as 1970, pp 104–128. Caries Res 1998;32:412–416.
sodium fluoride or monofluorophos- 39 Parkins FM, Tinanoff N, Moutinho M, 51 Ekstrand J: Fluoride concentrations in
phate to humans. J Fluoride Chem Anstey MB: Relationships of human saliva after single oral doses and their
2008;129:691–694. plasma fluoride and bone fluoride to relation to plasma fluoride. Scand J Dent
27 Carvalho JG, Leite AL, Yan D, Everett ET, age. Calcif Tiss Res 1974;16:335–338. Res 1977;85:16–17.
Whitford GM, Buzalaf MAR: Influence 40 Richards A, Mosekilde L, Sogaard CH: 52 Shannon IL, Suddick RP, Edmonds EJ:
of genetic background on fluoride Normal age-related changes in fluoride Effect of rate of gland function on
metabolism in mice. J Dent Res content of vertebral trabecular bone – parotid saliva fluoride concentration in
2009;88:1054–1058. relation to bone quality. Bone 1994; the human. Caries Res 1973;7:1–10.
28 Ekstrand J, Zeigler EE, Nelson SE, 15:21–26. 53 Crosby ND, Shepherd PA: Studies on
Fomon SJ: Absorption and retention of 41 Everett ET, McHenry MA, Reynolds N, patterns of fluid intake, water balance
dietary and supplemental fluoride by Eggertsson H, Sullivan J, Kantmann C, and fluoride. Med J Aust 1957;ii:
infants. Adv Dent Res 1994;8:175–180. Martinez-Mier EA, Warrick JM, Stookey 341–346.
29 Spencer H, Lewin I, Wistrowski E, GK: Dental fluorosis: variability among 54 Hodge HC, Smith FA, Gedalia I: Excre-
Samachson J: Fluoride metabolism in different inbred mouse strains. J Dent tion of fluorides; in WHO (ed): Fluo-
man. Am J Med 1970;49:807–813. Res 2002;81:794–798. rides and human health. Geneva, WHO,
30 Rao HV, Beliles RP, Whitford GM, 42 Zipkin I, McClure FJ, Leone NC, Lee 1970, pp 141–161.
Turner CH: A physiologically based WA: Fluoride deposition in human 55 Backer Dirks O, Jongeling-Eijndhoven
pharmacokinetic model for fluoride bones after prolonged ingestion of fluo- JMPA, Flissebaalje TD, Gedalia I: Total
uptake by bone. Reg Toxicol Pharmacol ride in drinking water. Publ Health Rep and free ionic fluoride in human and
1995;22:30–42. 1958;73:732–740. cow’s milk as determined by gas-liquid
31 Narita N, Kato K, Nakagaki H, Ohno N, 43 Brunn C, Givskov H, Thylstrup A: Whole chromatography and the fluoride elec-
Kameyama Y, Weatherell JA: Distribu- saliva fluoride after toothbrushing with trode. Caries Res 1974;8:181–186.
tion of fluoride’s concentration in rat’s NaF and MFP dentrifices with different 56 Koparal E, Ertugrul F, Oztekin K: Fluo-
bone. Calcif Tiss Int 1990;46:200–204. F concentrations. Caries Res 1984;18: ride levels in breast milk and infant
32 Li J, Nakagaki H, Kato K, Ishiguro K, 282–288. foods. J Clin Pediatr Dent 2000;24:
Ohno N, Kameyama Y, Mukai M, Ikeda 44 Schamschula RG, Sugar E, Un PSH, Toth 299–302.
M, Weatherell J, Robinson C: Distribu- K, Barmes DE, Adkins BL: Physiological 57 Ericsson Y, Gydell K, Hammarskioeld T:
tion profiles of fluoride in three different indicators of fluoride exposure and utili- Blood plasma fluoride: an indicator of
kinds of rats bone. Bone 1993;14: zation: an epidemiological study. Com- skeletal fluoride content. J Int Res Com-
835–842. munity Dent Oral Epidemiol 1985;13: mun Syst 1973;1:33–35.
33 Buzalaf MAR, Caroselli EE, Carvalho JG, 104–107. 58 Waterhouse C, Taves D, Munzer A:
Oliveira RC, Cardoso VES, Whitford 45 Boros I, Keszler P, Banoczy J: Fluoride Serum inorganic fluoride: changes
GM: Bone surface and whole bone as concentrations of unstimulated whole related to previous fluoride intake, renal
biomarkers of acute fluoride exposure. and labial gland saliva in young adults function and bone resorption. Clin Sci
J Analyt Toxicol 2005;29:810–813. after fluoride intake with milk. Caries 1980;58:145–152.
34 Ishiguro K, Nakagaki H, Tsuboi S, Narita Res 2001;35:167–172. 59 EPA: Fluoride in drinking water; a scien-
N, Kato K, Li J, Kamei H, Yoshioka I, 46 Petersson LG, Arvidsson I, Lynch E, tific review of EPA’s standards (2006).
Miyauchi K, Hosoe H, et al: Distribution Engstrom K, Twetman S: Fluoride con- Committee on Fluoride in Drinking
of fluoride in cortical bone of human rib. centrations in saliva and dental plaque Water. Board on Environmental Studies
Calcif Tiss Int 1993;52:278–282. in young children after intake of fluori- and Toxicology. Division on Earth and
35 Bezerra de Menezes LM, Volpato MC, dated milk. Caries Res 2002;36:40–43. Life Studies. National Research Council
Rosalen PL, Cury JA: Bone as a bio- 47 Hedman J, Sjoman R, Sjostrom I, Twet- of the National Academies. Washington,
marker of acute fluoride toxicity. Foren- man S: Fluoride concentration in saliva National Academies Press Washington,
sic Sci Int 2003;137:209–214. after consumption of a dinner meal pre- 2006.
36 Buzalaf MAR, Caroselli EE, Oliveira RC, pared with fluoridated salt. Caries Res 60 Czarnowski W, Krechniak J: Fluoride in
Granjeiro JM, Whitford GM: Nail and 2006;40:158–162. the urine, hair, and nails of phosphate
bone surface as biomarkers of acute flu- 48 Pessan JP, Silva SM, Lauris JR, Sampaio fertiliser workers. Br J Ind Med
oride exposure in rats. J Analyt Toxicol FC, Whitford GM, Buzalaf MA: Fluo- 1990;47:349–351.
2004;28:249–252. ride uptake by plaque from water and
from dentifrice. J Dent Res 2008;87:

50 Rugg-Gunn · Villa · Buzalaf

61 Czarnowski W, Krechniak J, Urbanska B, 65 Maguire A, Zohouri FV, Hindmarch PN, 71 Ekstrand J, Hardell LI, Spak CJ: Fluoride
Stolarska K, Taraszewska-Czarnowska Hatts J, Moynihan PJ: Fluoride intake balance studies on infants in a 1 ppm
M, Muraszko-Klaudel A: The impact of and urinary excretion in 6- to 7-year-old water fluoride area. Caries Res
water-borne fluoride on bone density. children living in optimally, sub- 1984;18:87–92.
Fluoride 1999;32:91–95. optimally and non-fluoridated areas. 72 Maheshwari UR, McDonald JT, Sch-
62 Villa AE, Anabalón M, Cabezas L: The Community Dent Oral Epidemiol 2007; neider VS, Brunetti AJ, Leybin L, New-
fractional urinary fluoride excretion in 35:479–488. brun E, Hodge HC: Fluoride balance
young children under stable fluoride 66 Villa A, Anabalon M, Zohouri V, Magu- studies in ambulatory healthy men with
intake conditions. Community Dent ire A, Franco AM, Rugg-Gunn A: Rela- and without fluoride supplements.
Oral Epidemiol 2000;28:344–355. tionships between fluoride intake, uri- Am J Clin Nutr 1981;34:2679–2684.
63 Haftenberger M, Viergutz G, Neumeister nary fluoride excretion and fluoride 73 Zohouri FV, Rugg-Gunn AJ: Total fluo-
V, Hetzer G: Total fluoride intake and retention in children and adults: an ride intake and urinary excretion in
urinary excretion in German children analysis of available data. Caries Res 4-year-old Iranian children residing in
aged 3–6 years. Caries Res 2001;35: 2010;44:60–68. low-fluoride areas. Br J Nutr 2000;83:
451–457. 67 Villa AE, Anabalón M, Rugg-Gunn A: 15–25.
64 Villa AE, Cabezas L, Anabalón M, Garza Fractional urinary fluoride excretion of 74 Franco AM, Saldarriaga A, Martignon S,
E: The fractional urinary fluoride excre- young female adults during the diurnal González MC, Villa AE: Fluoride intake
tion of adolescents and adults under and nocturnal periods. Caries Res 2008; and fractional urinary fluoride excretion
customary fluoride intake conditions, in 42:275–281. of Colombian preschool children. Com-
a community with 0.6 mg F/L in its 68 Whitford GM, Pashley DH, Stringer GI: munity Dent Health 2005;22:272–278.
drinking water. Community Dent Health Fluoride renal clearance: a pH- 75 Marthaler TM: Monitoring of renal fluo-
2004;21:11–18. dependent event. Am J Physiol 1976;230: ride excretion in community preventive
527–532. programmes on oral health. Geneva,
69 Schiffl HH, Binswanger U: Human uri- World Health Organization, 1999.
nary fluoride excretion as influenced by 76 Whitford G: Fluoride metabolism and
renal functional impairment. Nephron excretion in children. J Public Health
1980;26:69–72. Dent 1999;59:224–228.
70 Hanhijärvi H, Penttilä I: The relation-
ship between human ionic plasma fluo-
ride and serum creatinine concentra-
tions in cases of renal and cardiac
insufficiency in a fluoridated commu-
nity. Proc Finn Dent Soc 1981;77:

Prof. Andrew John Rugg-Gunn

Morven, Boughmore Road
Devon EX10 8SH (UK)
Tel. +44 1395 578746, E-Mail

Contemporary Biomarkers 51
Fluoride Intake, Metabolism and Toxicity
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 52–65

Historical and Recent Biological Markers of

Exposure to Fluoride
Juliano Pelim Pessana ⭈ Marília Afonso Rabelo Buzalafb
aDepartment of Pediatric Dentistry and Public Health, Araçatuba Dental School, São Paulo State University, Araçatuba, and
bDepartment of Biological Sciences, Bauru Dental School, University of São Paulo, Bauru, Brazil

Abstract that are commonly extracted, is advantageous. However,

Recent and historical biomarkers assess chronic or sub- mean values also span a wide range and reference con-
chronic exposure to fluoride. The most studied recent centrations have not been published yet.
biomarkers are nails and hair. Both can be non-invasively Copyright © 2011 S. Karger AG, Basel
obtained, although collection of nails is more accepted by
the subjects. External contamination may be a problem Measuring fluoride intake is an important tool
for both biomarkers and still needs to be better evaluated. when controlling the risk factors for dental flu-
Nails have been more extensively studied. Although the orosis. The determination of ingestion levels of
available knowledge does not allow their use as predic- fluoride, however, is becoming increasingly dif-
tors of dental fluorosis by individual subjects, since ref- ficult, as fluoride comes from different sources.
erence values of fluoride have not yet been established, Considering that only the absorbed fluoride is
they have a strong potential for use in epidemiological implicated in the development of dental fluorosis,
surveys. Toenails should be preferred instead of finger- the monitoring of fluoride absorption, instead of
nails, and variables that are known to affect nail fluoride fluoride intake, seems to be more accurate [1].
concentrations – such as age, gender and geographical Monitoring fluoride exposures can be accom-
area –should be considered. The main historical biomark- plished through the analysis of several biological
ers that could indicate total fluoride body burden are bone tissues or fluids, with varying degrees of accuracy.
and dentin. Of these, bone is more studied, but its fluo- While studies evaluating the time course of body
ride concentrations vary according to the type of bone fluid fluoride concentrations following the in-
and subjects’ age and gender. They are also influenced by gestion of a fluoride compound typically involve
genetic background, renal function and remodeling rate, timed collections of blood plasma or parotid duc-
variables that complicate the establishment of a normal tal saliva, the analysis of urinary fluoride concen-
range of fluoride levels in bone that could indicate ‘desir- trations can also provide useful information, al-
able’ exposure to fluoride. The main issue when attempt- though the data are less precise and more difficult
ing to use bone as biomarker of fluoride exposure is the to interpret than those derived from the analysis
difficulty and invasiveness of sample collection. In this of plasma [2]. In order to have more precise indi-
aspect, collection of dentin, especially from 3rd molars cators of the fluoride levels in the organism and,
therefore, to be able to predict the risk of dental last few hours or differences in glomerular filtra-
fluorosis, the search for biomarkers of exposure tion rate, urinary pH or urinary flow rate. Such
to fluoride has been intensified over the last years. advantages make the analysis of fingernail clip-
Biological markers or ‘biomarkers’ are defined as pings an attractive alternative to other body fluids
indicators signaling events in biological systems or tissues for the purpose of monitoring fluoride
or samples [3]. According to Grandjean [4], bio- exposure [2].
markers are not used as diagnostic tests, but in- The idea of using nails to monitor fluoride ex-
stead as indicators of changes that could lead to a posure is not new. Studies conducted in the 1970s
clinical disease. and 1980s reported that fluoride concentrations
Biomarkers of fluoride can be arranged ac- in nails could reflect differences in chronic flu-
cording to two classifications. The Committee oride exposure from the atmosphere [16] and
of Biomarkers of the National Research Council from water [17]. The possibility of using nails as
[3] divided biomarkers into biomarkers of: (1) biomarkers of subchronic exposure to fluoride
effect, (2) susceptibility, and (3) exposure; while was first described by Whitford et al. [7], about
the World Health Organization [5] proposes a 10 years ago. In that study, one of the authors in-
time-perspective classification, in which bio- creased his fluoride intake for 1 month (3–6 mg/
markers are divided into: (1) historical, (2) con- day) and this was reflected in fingernail fluoride
temporary, and (3) recent markers. The present concentrations 3.5 months later (fig. 1). That study
review will adopt the WHO criteria and will fo- also helped to clarify how fluoride is incorporated
cus on recent (nails and hair) and historical (bone in nails. Although incorporation of fluoride from
and teeth) biological markers of exposure to fluo- the environment is also possible, it became clear
ride. Contemporary biomarkers are addressed by that fluoride enters the nail mainly via the growth
Rugg-Gunn et al. [this vol., pp. 37–51]. end, and that the concentration in the nail clip-
ping is determined by the average plasma fluoride
concentration that existed while the clipping was
Recent Biomarkers of Exposure to Fluoride forming [2]. Incorporation of fluoride through
the nail bed seems to contribute at a lesser extent
Nails to the total nail fluoride concentration. In a study
Among the short- and long-term biomarkers conducted in 2- to 3-year-old children, a fluoride-
studied, nails seem to be promising both for acute free toothpaste was used for 1 month, then a fluo-
[6], subchronic [7–9] and chronic [7, 10–15] ex- ridated toothpaste (1,570 ppm as sodium mono-
posure to fluoride. Nail sampling has some advan- fluorophosphate) for another month, and finally
tages, since samples can be accessed and collected a fluoride-free toothpaste for an extra month [8].
in a non-invasive manner, besides the possibility Although the highest peak fluoride concentration
of storage for long periods of time without degra- was seen 16 weeks after starting use of the fluori-
dation. Also, the concentration of fluoride reflects dated toothpaste, a smaller peak was also seen 12
the average level of intake and plasma concentra- weeks earlier, which could suggest incorporation
tion over a protracted period, in contrast to the of fluoride also through the nail bed, although to
analysis of urine, plasma or ductal saliva, whose a lesser extent (fig. 2). In a subsequent study, using
fluoride concentrations are more like ‘snapshots’, the same protocol as that described by Whitford
and therefore subject to change due to recent flu- et al. [7], subjects increased their fluoride intake at
oride intake and certain physiological variables. a much lower level (1.8 mg/day), and nail growth
Finally, fingernail concentrations are not affect- rates and lengths were also evaluated. The increas-
ed by variables such as fluoride intake within the es verified in nails occurred within the limits of the

Biomarkers of Fluoride Exposure 53


Fingernail [F], mg/kg ( )


Fig. 1. Fingernails fluoride concen- 3

trations over time after 1 subject
increased his fluoride intake by 3
mg/day for 30 days (horizontal bar). 2
Open circles indicate mean urinary

Jan 22
Feb 6
Feb 22
Mar 9
Mar 25
Apr 10
Apr 26
May 9
May 22
Jun 4
Jun 18
Jul 2
Jul 18
Aug 4
Aug 19
Sep 3
Sep 21
fluoride concentrations before (left)
and after (right) increased fluoride
intake. Date
Source: Whitford et al. [7].

Mean fluoride concentration (μg/g)

10 Toenails

Fig. 2. Time course of fluoride con- 6

centrations in fingernail and toe-
nail clippings. Placebo dentifrice
was used between 29-11-01 and
27-12-01. Fluoride dentifrice (1,570
ppm) was used between 28-12-01
and 24-01-02. Placebo dentifrice
was used again between 25-01-02 25_07
and 21-02-02, when the children
returned to their usual dentifrices. Dates
Source: Correa Rodrigues et al. [8].

95% CI of the mean lag times for fluoride detec- and its concentrations in nails. According to the
tion in nails (fig. 3, 4), indicating that nail growth table, fingernails were found to have higher F con-
rates and lengths are important determinants when centrations than toenails in 3 of the 7 studies that
evaluating subchronic fluoride exposure [9]. compared fluoride concentrations between the
Table 1 summarizes the main published stud- two types of nails [7, 12, 14]. The authors attri-
ies on the relationship between fluoride exposure bute the higher F concentrations in fingernails to

54 Pessan · Buzalaf
Fig. 3. Time course of fluoride con-
centrations in fingernails clippings
(n = 10). The clippings were ob-
tained every 14 days, totalizing 15 1.9
fingernails clippings. An additional 1.8

Fingernails [F] (μg/g)

1.8 mg/day of fluoride was ingested ab
in 3 divided doses for 30 days from 1.7 ab
the beginning of the study (hori- 1.6 ab
zontal bar). Fluoride intake from
the diet and dentifrice was estimat- 1.5 ac
ed in 1.5 mg/day throughout the 1.4 ac ac ac
6-month period. After 98, 112 and ac
1.3 ac
154 days the additional intake of 1.8 ac
mg/day was reflected in increased 1.2 c
fluoride concentrations in toenails, c
although this difference was not 0 14 28 42 56 70 84 98 112 126 140 154 168 182 196
statistically significant from base- Days
line values. After day 154, fluoride
concentrations decreased, and the
values obtained in days 182 and 210
were significantly lower. Data are
presented as mean ± SE. Different
letters indicate statistical signifi-
cance (p < 0.05). Vertical bars indi-
cate expected mean lag time for flu-
oride detection in fingernails (solid)
and 95% confidence interval (dots).
Source: Buzalaf et al. [9].

Fig. 4. Time course of fluoride con-

centrations in toenails clippings
(n = 10). The clippings were
obtained every 14 days, totalizing
15 toenails clippings. An additional 2.0
1.8 mg/day of fluoride was ingested
Toenails [F] (μg/g)

in 3 divided doses for 30 days from 1.8
the beginning of the study (hori- bc ac
ac ac
zontal bar). Fluoride intake from the 1.6 ac ac
diet and dentifrice was estimated in
1.5 mg/day throughout the 6-month
1.4 ac ac ac
period. After 126 and 154 days the
additional intake of 1.8 mg/day was a a a
1.2 a
reflected in significantly increased a
fluoride concentrations in toenails.
Data are presented as mean ± SE. 0 14 28 42 56 70 84 98 112 126 140 154 168 182 196
Different letters indicate statistical Days
significance (p < 0.05). Vertical bars
indicate expected mean lag time for
fluoride detection in toenails (solid)
and 95% confidence interval (dots).
Source: Buzalaf et al. [9].

Biomarkers of Fluoride Exposure 55

Table 1. Summary of the main studies published (in reverse chronological order) on fluoride concentrations in fin-
gernails and/or toenails

Reference Exposure Age range, Site Sample size Main findings


Buzalaf et al. chronic 4–6 fingernails/ 121 correlations for F intake and nail
[15] toenails concentrations were higher than for
intake and urinary excretion (in children
exposed to fluoridated water, salt and

Lima-Arsati subchronic 1–3 fingernails 23 no detectable difference in nail F

et al. [54] concentrations when using a 1,500-ppm
F dentifrice instead of a placebo

Fukushima chronic 3–7 fingernails/ 300 (1) geographical area and water F
et al. [13] 14–20 toenails concentration exerted the most
30–40 influence on finger/toenail
50–60 concentrations;
(2) higher F concentrations in nails from
older subjects and females

Buzalaf et al. chronic 5–6 fingernails/ 60 significant differences between F

[14] toenails concentrations in nails from children
using conventional toothpaste (1,100
ppm, pH 7.0) and low-F toothpaste
(550 ppm, pH 4.5)

de Almeida chronic 1–3 fingernails 33 no correlation between F intake and

et al. [21] concentration in nails

Buzalaf et al. subchronic 20–30 fingernails/ 10 increased F concentrations in toenails

[9] toenails 3.7 months after increased F intake
(1.8 mg/day, 30 days)

Pessan et al. subchronic 4–7 fingernails 20 no detectable difference in nail

[55] concentrations when using a 1,500-ppm
F dentifrice instead of a placebo

Levy et al. chronic 2–6 fingernails/ 30 significant differences in fingernail and

[12] toenails toenail F concentrations between
children living in areas with different
F concentrations in the drinking

Correa subchronic 2–3 fingernails/ 10 increased F concentrations in nails

Rodrigues toenails when using a 1,500 ppm F dentifrice
et al. [8] instead of a placebo toothpaste

56 Pessan · Buzalaf
Table 1. Continued

Reference Exposure Age range, Site Sample size Main findings

Whitford et al. (a) chronic 6–7 fingernails 46 (a) significant differences in nail F
[7] concentrations among residents in
communities with 0.1, 1.6 and 2.3
ppm F in the drinking water

(b) chronic adult fingernails/ 1 (b) F concentrations in fingernails

toenails higher than in toenails

(c) subchro- adult fingernails 1 (c) Increased F concentrations in

nic fingernails 3.5 months after increased
F intake (3 mg/day, 30 days)

Spate et al. chronic adult toenails 25 higher F concentrations in nails from

[56] (women) residents in an area with 1 ppm F in the
water when compared to those from a
community with 0.1 ppm F

Schmidt and chronic 42–86 fingernails 38 higher F concentrations in nails of

Leuschke [57] individuals exposed to polluted air

Czarnowski subchronic 21–61 fingernails 110 significant correlation (r = 0.99)

and between F concentrations in nails and
Krechniak urine, which were directly related to
[10] exposure to F from the environment

Schamschula chronic children fingernails 139 F concentrations in nails directly

et al. [17] proportional to F concentrations in the
drinking water of 3 communities

Balazova [16] chronic 6–14 info. not info. not higher F concentrations in nails of
available available children living in close proximity of an
aluminum smelter in comparison to

F = Fluoride. Source: Clarkson et al. [58] (updated).

a higher blood supply. However, as that pattern in future studies evaluating fluoride levels in nails
could not be demonstrated by other investigators, [9, 13].
this question remains to be answered. One aspect The analytical method used for sample prepa-
that must be taken into account, however, is that ration and analysis also seems to be an important
toenails may be less prone to external contami- variable to be considered [10]. All papers pub-
nation by fluoride than fingernails, which might lished after the study by Whitford et al. [7] used
partially explain the lower values found in some HMDS (hexamethyldisiloxane)-facilitated dif-
studies. Therefore, it has been suggested that toe- fusion (Taves [18], modified by Whitford [19]),
nails, especially from the big toes (which provide so if the results obtained by those investigators
enough mass for fluoride analysis) should be used are used as reference values, it can be seen that

Biomarkers of Fluoride Exposure 57

Table 2. Multivariate linear regression analysis of factors associated to fingernail fluoride concentration

Variable Coefficient SE coeff. β SE β p value Adjusted R2

Water fluoride level 1.95 0.33 0.30 0.05 <0.001 0.33

Age 0.03 0.01 0.15 0.05 0.002

0 = Female –2.15 0.41 –0.25 0.05 <0.001
1 = Male

Geographical area
0 = Southeast (A, C) 3.78 0.55 0.45 0.07 <0.001
1 = Northeast (B, D, E)

0 = Urban (A, C, D) –2.43 0.54 –0.29 0.06 <0.001
1 = Rural (B, E)

The variable growth rate was included in the model, but was removed since it was not statistically significant. A–E =
5 Brazilian communities used in the original study. Source: Fukushima et al. [13].

the results are somewhat comparable, given the influence the fluoride concentrations in finger-
differences in the background fluoride exposure nails and toenails [13]. The effects of water fluo-
(diet and use of dental products). Whitford [2] ride concentration, age, gender, nail growth rate
compared his previous results (Whitford et al. and geographical area on the fluoride concentra-
[7]) with those obtained by other investigators tion in the fingernail and toenail clippings were
using different preparative methods and analyt- evaluated in 300 volunteers, distributed into 4
ical techniques. Although a clear dose-response age groups. Among the tested factors, geograph-
relationship could be observed between nail flu- ical area and water F concentration exerted the
oride concentrations and the level of exposure most influence on finger- and toenail fluoride
to fluoride in each report, the author observed concentrations (tables 2, 3). Subjects of older age
markedly lower or higher values, showing that groups from communities located at a warmer re-
care must be taken when evaluating results from gion and with naturally fluoridated drinking wa-
investigators using different methodologies. The ter showed higher nail fluoride concentrations
HMDS method presents some advantages, as than the others. Females presented higher nail
samples are not required to be minced or ashed, fluoride concentrations than males. The authors
and cleaning the nails by brushing with deion- concluded that water fluoride concentration, age,
ized water can be done without loss of intrinsic gender and geographical area influenced the flu-
fluoride, reinforcing the concept that fluoride is oride concentrations of fingernails and toenails,
mainly incorporated into nails from the systemic and should be taken into account when using this
circulation [7]. biomarker of exposure to predict risk for dental
In order to provide stronger evidence for the fluorosis.
validation of nails as biomarkers of fluoride expo- Although the use of nails seems to be promis-
sure, a recent study evaluated factors that might ing for monitoring fluoride exposure in humans,

58 Pessan · Buzalaf
Table 3. Multivariate linear regression analysis of factors associated to toenail fluoride concentration

Variable Coefficient SE coeff. β SE β p value Adjusted R2

Water fluoride level 0.404 0.034 0.484 0.040 <0.001 0.58

Age 0.003 0.001 0.101 0.039 0.009

0 = Female –0.140 0.042 –0.129 0.039 0.001
1 = Male

Geographical/social- economic area

0 = Southeast (A, C) 0.454 0.044 0.419 0.040 <0.001
1= Northeast (B, D, E)

The variable growth rate was included in the model but was removed since it was not statistically significant. A–E = 5
Brazilian communities used in the original study. Source: Fukushima et al. [13].

some points need to be addressed. The main is- different populations, it is not uncommon to ob-
sues about the validation of nails as biomarkers serve overlaps among the 95% CI in the analyzed
of fluoride exposure are related to the sensitivity populations.
and specificity of the method. It is possible that It must be highlighted, however, that no study
nails could only be used to differentiate levels of correlated dental fluorosis severity with fluoride
fluoride exposure when there is a broad varia- concentrations to which children were exposed
tion among them. A clinical study demonstrated during the formation of the permanent dentition,
significant differences in fluoride concentrations so the sensitivity and specificity of nails to pre-
in fingernails among children with Thylstrup- dict dental fluorosis still need to be determined.
Fejerskov scores of 0 and 5, but not in children To date, only one ongoing study conducted
with Thylstrup-Fejerskov scores of 1, 2, 3 and 4 with Brazilian children has addressed this issue.
[20]. Another study, which evaluated fluoride in- Fingernail clippings were collected from chil-
take from diet and dentifrice in 1- to 3-year-old dren during the period of formation of perma-
children, also demonstrated that small variations nent teeth and analyzed for fluoride content, and
in the daily dose of fluoride intake cannot be de- this information was later correlated with dental
tected in fingernails, suggesting that fingernails fluorosis prevalence in the permanent dentition
give an indication of fluoride intake over the long (unpublished data). Fluoride concentrations in
term and are unlikely to be sufficiently sensitive fingernails of children presenting dental fluorosis
to distinguish small day-to-day variations in flu- were significantly higher than those not present-
oride intake [21]. Another problem concerning ing fluorosis. More importantly, although some
sensitivity and specificity is evidenced by studies values could be classified as outliers, fluoride con-
which evaluated fluoride concentrations in nails centrations in fingernails were directly related to
of subjects residing in areas with different levels the severity of dental fluorosis in the majority of
of exposure to fluoride. Although most of those the cases. These exciting results, although they
studies were able to determine significant dif- should be interpreted with caution, indicate that
ferences in nails fluoride concentrations among nails could be used in the near future as indicators

Biomarkers of Fluoride Exposure 59

of systemic exposure to fluoride that could lead to determination of the variables that affect individ-
the development of dental fluorosis. ual variations would be instructive.
Another issue involving the use of nails as bio- Finally, based on the results from the exist-
markers of exposure to fluoride is the possibil- ing literature, future studies should use toenails
ity of fluoride uptake from exogenous sources. when monitoring fluoride exposure, and growth
Although Whitford et al. [7] showed no effect of rate and length must be taken into account when
prolonged immersion of nails in deionized or flu- monitoring subchronic exposure. Also, water flu-
oridated water (1 ppm) on the resulting fluoride oride concentration, age, gender and geographi-
concentration, a more recent investigation dem- cal area should be considered when using this
onstrated that soaking nails in a 1,100-ppm fluo- biomarker of exposure to predict risk for dental
ride dentifrice slurry for 3 minutes, or in water fluorosis.
with 100 ppm fluoride for 2 h can dramatically
increase the fluoride concentration in nails [22]. Hair
Vigorous cleaning of nail fragments by brushing As described for nails, the use of hair as an indica-
with deionized water and sonication were shown tor of systemic exposure to fluoride has been the
to be unable to remove fluoride incorporated subject of investigations for over 4 decades, both
from the dentifrice slurry and from the solution, in studies with animals [23, 24] and humans [10,
leading to the conclusion that an inappropriate 16, 17, 25–28]. The rationale for the use of hair
external decontamination can lead to a misclassi- as a biomarker of fluoride (and other elements)
fication of exposure when using nails as a fluoride exposure is the same of that for nails: the endog-
biomarker. Other disadvantages include: (1) the enous trace element composition of hair and nails
need of fluoride extraction (in contrast to other is believed to reflect the metabolic environment
biomarkers), which increases both time and cost; during formation and to be relatively isolated
(2) the unsuitability of some nails for analysis, when the finished structure is expelled from the
such as nails covered with polish (as some prod- skin [29].
ucts contain fluoride); (3) the possible effects of Hair has been reported to be a suitable
some variables on the rate of fluoride incorpora- biomarker to monitor fluoride exposure from dif-
tion into fingernails remain to be determined (as ferent environmental sources. Balazova [16] eval-
nail diseases). uated fluoride concentrations in the hair of chil-
Considering all the advantages and disadvan- dren after the start of operation of an aluminum
tages of monitoring chronic and subchronic ex- smelter, and compared these to values obtained
posure to fluoride using nails, it is evident that in children from a control area. Mean fluoride
this biomarker is a promising tool to be used in concentration in the hair of exposed children
dentistry. Although nail clippings still cannot be was 16.0 μg/g, about twice the amount in non-
used as predictors of dental fluorosis in individ- exposed children (7.5 μg/g). Such increases were
ual subjects, they have a strong potential for epi- also observed for fluoride content of teeth, nails
demiological surveys, as demonstrated by sev- and urine of exposed children. According to the
eral reports using different research protocols. author, that study prompted the introduction of
More studies are still needed before nails can be measures for the protection of the life and health
fully validated as biomarkers of exposure to fluo- of the population affected. After 10 years, medi-
ride. The appropriate method for decontamina- cal examinations along with analysis of fluoride
tion of fluoride from external sources, without in teeth, nails and urine were repeated, and the
removing fluoride incorporated from the blood results showed that mean fluoride concentrations
supply, would be extremely important. Also, the in hair had decreased from 16.0 to 4.0 μg/g [30].

60 Pessan · Buzalaf
Other studies also reported the usefulness of hair comparing the studies of Schamschula et al. [17]
in monitoring the levels of exposure to fluoride in and Czarnowski and Krechniak [10]. In both stud-
hydrofluoric acid workers [26], as well as in phos- ies, fluoride was analyzed with a fluoride-selective
phate fertilizer workers [10]. In both cases, mean electrode using the same extraction method, and
fluoride concentrations in hair of exposed work- both studies found a positive correlation between
ers were significantly higher than those observed fluoride exposure and the fluoride content of the
for the control subjects. hair and nails. However, as described by Ophaug
Hair has also been shown to be effective in de- [29], an 8-fold increase in systemic fluoride ex-
tecting significant differences among children ex- posure in the study of Schamschula et al. [17] in-
posed to water containing different fluoride lev- creased the fluoride content of hair and nails by a
els (≤0.11 ppm; 0.5–1.1 ppm; 1.6–3.1 ppm) [17]. factor of approximately 2 and 3, respectively; in
Mean fluoride concentrations in hair increased contrast, a 3-fold increase in systemic fluoride ex-
consistently and significantly with increasing wa- posure in the study by Czarnowski and Krechniack
ter fluoride levels, and such a trend was also seen [10] led to a 200-fold increase in fluoride concen-
for urine, nails, saliva, enamel and plaque samples. trations in hair, and a 14-fold increase in nail fluo-
A direct relationship was observed between water ride concentration. In the review by Ophaug [29],
fluoride concentration and dental fluorosis, while issues concerning the length of hair used for anal-
an inverse relationship was seen between water ysis, as well as the decontamination procedures
fluoride concentration and dental caries. used are regarded as the possible causes for the
Although the use of hair as biomarker of fluo- conflicting results obtained.
ride exposure started to be investigated at about In addition to issues regarding external contam-
the same time as nails, there seems to have been a ination of nails, it is worth mentioning that despite
lack of interest in the subject in the last few years. the significant differences observed among groups
With the exception of one study, the majority of exposed to different sources of fluoride, some in-
works assessing the use of hair as a biomarker dividual values overlap between the groups, so
of exposure to fluoride were conducted over 15 no reference value is currently available to indi-
years ago. The most recent report addressing the cate safety levels of systemic exposure to fluoride.
use of hair as indicator of systemic fluoride expo- Also, as hair sampling needs to be done as close to
sure showed that fluoride content in the drinking the scalp as possible, it may not be accepted by all
water is highly correlated with fluoride content in subjects, especially those with long hair.
hair and with dental fluorosis levels in 12-year-old
Serbian children [28]. The authors stated that hair
might be regarded as a biomaterial of high infor- Historical Biomarkers of Exposure to Fluoride
mative potential in evaluating prolonged expo-
sure to fluorides and to individuate children at Bone
risk of fluorosis, regardless of the phase of teeth General characteristics of bone tissue are des-
eruption. cribed by Rugg-Gunn et al. [this vol., pp. 37–51],
As for nails, despite the promising results men- while the role played by bone in the metabolism
tioned above, the assessment of fluoride exposure of fluoride is described by Buzalaf and Whitford
using hair presents issues regarding the method [this vol., pp. 20–36]. Briefly, around 40% of an ab-
of fluoride extraction and external contamina- sorbed amount of fluoride will become incorpo-
tion. Ophaug [29] discussed possible factors that rated into bone in normal adults and even higher
could influence the results obtained depend- percentages (around 55%) in children [31]. Thus,
ing on the method of pretreatment sampling, by bone is the main site of fluoride accumulation in

Biomarkers of Fluoride Exposure 61

the body. Fluoride levels in bone throughout life in cancellous bone when compared with com-
tend to increase because 99% of the body bur- pact bone due to the higher blood supply of the
den of fluoride is associated with calcified tissues, former [31, 38, 39]. Gender also seems to influ-
mainly with the skeleton [32]. This makes bone a ence bone fluoride concentrations. Men usually
natural candidate as a fluoride biomarker. present higher concentrations than women, who
In order to understand the potential applica- appear to have delayed bone fluoride uptake [36,
tions of bone as a biomarker of exposure to fluo- 39]. All these variables make bone fluoride con-
ride, it is important to comprehend in which sites centrations extremely variable, turning difficult
fluoride accumulates within the tissue. A physio- the establishment of a normal range of fluoride
logically based pharmacokinetic model considers levels in bone that would indicate ‘desirable’ ex-
that bone has two compartments: a small, flow- posure to fluoride. However, it is possible to de-
limited, rapidly exchangeable surface bone com- tect differences in bone fluoride concentrations of
partment, and a bulk, virtually non-exchangeable, individuals living in areas with distinct fluoride
inner bone compartment. However, it is known concentrations provided that samples matched
that fluoride associated with the inner bone com- for age are compared [45–47].
partment is not irreversibly bound. In the longer The main issue when attempting to use bone
time frame it may be mobilized through the con- as biomarker of fluoride exposure is the difficulty
tinuous process of bone remodeling in the young, and invasiveness of sample collection. Most of the
bone resorption and bone remodeling in the studies that evaluated bone fluoride concentra-
adult. These features were elegantly described in tions collected post-mortem samples or samples
a pharmacokinetics study related to chronic expo- from subjects undergoing orthopedic surgery,
sure to fluoride [33]. It is important to highlight which considerably limits the usefulness of this
that fluoride is not taken up by fully mineralized biomarker.
bone. It accumulates solely in bone formed dur-
ing exposure to fluoride. Due to this, incorpora- Teeth
tion of fluoride in adult bone occurs only during As discussed by Buzalaf and Whitford [this vol.,
remodeling. pp. 20–36], mineralized tissues are the main site
In the 1990s, a mathematical model for fluo- of fluoride retention in the organism. Thus, they
ride uptake by the skeleton was proposed [34]. are considered to be biomarkers for total fluoride
This model suggested that: (1) binding of fluoride body burden or historical biomarkers of exposure
to bone is nonlinear; smaller percentages of fluo- to fluoride. Despite the fact that most of the fluo-
ride bind to bone upon higher levels of fluoride ride retained in the organism is associated with the
intake; (2) bone resorption rate is inversely related skeleton, part of it is deposited in the teeth. Since
to bone fluoride content since it is proportional to the collection of bone is difficult and invasive, teeth
the solubility of hydroxyfluorapatite; (3) fluoride have emerged as potential historical biomarkers of
clearance from the skeleton by bone remodeling exposure to fluoride, in particular third molars or
takes over four times longer than fluoride uptake, premolars that are commonly extracted.
and, as a consequence, bone fluoride concentra- Unlike bone, teeth comprise two distinct types
tions increase with age [35–39]. of mineralized tissues: enamel and dentin. The
Other variables that might affect fluoride in- timing and characteristics of fluoride uptake in
corporation into bones are genetic background both tissues are quite different, which has impli-
[40–43], renal function [44] and remodeling rate cations for the usefulness of these biomarkers.
[36]. Bone fluoride concentrations also depend In enamel, systemic fluoride uptake occurs only
on the type of bone analyzed. They are higher during tooth formation. As a consequence, bulk

62 Pessan · Buzalaf
enamel fluoride concentrations mainly reflect intake and the most appropriate indicator of to-
the level of systemic exposure to fluoride during tal fluoride body burden. This was confirmed in
tooth formation [32]. However, after tooth erup- a study that found a significant correlation be-
tion, enamel fluoride concentrations are subject to tween dentin fluoride concentration and dental
change. Enamel fluoride concentrations are high- fluorosis severity in unerupted third molars of in-
est at the surface and reduce progressively toward dividuals with different levels of exposure to fluo-
the dentin-enamel junction. Fluoride concentra- ride from the drinking water. However, the coef-
tions at the enamel surface tend to decrease with ficient of determination was very low (r2 = 0.1)
age in areas subjected to tooth wear [48–50]. On [52], suggesting that other factors – such as in-
the other hand, they tend to increase in areas that dividual genetic variation that affects the suscep-
accumulate plaque, since carbonated hydroxiapa- tibility to dental fluorosis [40] and the fluoride
tite is gradually replaced by fluoridated hydroxi- metabolism [41] – may account for dental fluo-
apatite during cariogenic challenges in the pres- rosis severity. This makes it difficult to establish
ence of topical fluoride [for details, see Buzalaf et ‘normal’ levels of fluoride in dentin, above which
al., this vol., pp. 97–114]. The possible changes in excessive exposure to fluoride would be expected
enamel fluoride concentrations caused by vari- to occur.
ables other than fluoride intake (such as wear or
sequential de- and remineralization reactions) re-
duce the ability of this biomarker to estimate flu- Conclusion
oride intake or to the predict risk of developing
dental fluorosis. Despite one study finding a sig- The knowledge that fluoride controls caries main-
nificant correlation between enamel fluoride con- ly due to its topical effect made it possible to ob-
tent and the degree of dental fluorosis [51], a more tain the maximum benefit of this element with a
recent investigation with a larger sample size eval- minimum of unwanted effects. Researchers across
uating unerupted human third molars from areas the globe then turned their attention towards con-
with different fluoride levels in the drinking water trolling the amount of fluoride intake. Since this
did not confirm such a correlation [52]. is a hard task due to the plethora of available fluo-
Dentin, in contrast, continues to form and to ride sources, the use of biomarkers of exposure to
accumulate fluoride throughout life. Additionally, this ion gained considerable attention.
it only contains fluoride that has been incorpo- Recent and historical biomarkers assess (chron-
rated through systemic ingestion and is protected ic or subchronic) exposure to fluoride in the me-
from topical fluoride exposure by the presence of dium- and long-term, respectively. Considering
the covering enamel. Dentin fluoride concentra- the recent biomarkers, from the available knowl-
tions, similarly to what has been found for bone edge it seems that nails are more suitable for mon-
and contrarily to enamel, increase with age due itoring fluoride exposure, since more information
to continuous fluoride uptake throughout life is available and their collection is less prone to
[53]. Moreover, they reduce progressively from questioning by the subjects. As for the historical
the pulpal surface to the dentin-enamel junction biomarkers that could indicate total fluoride body
[48]. The pattern of fluoride uptake in dentin burden, bone – despite being studied more – does
and, consequently, its fluoride concentrations re- not seem to be suitable since its collection is dif-
semble those found for bone [32]. Since dentin is ficult and invasive. Dentin is more appropriate in
more easily obtained than bone (especially from this regard.
third molars that are often extracted), it seems to It should be emphasized, however, that none
be the best biomarker to estimate chronic fluoride of the above-mentioned biomarkers could be

Biomarkers of Fluoride Exposure 63

used as predictors of dental fluorosis for indi- great variability in levels. Nevertheless, they seem
vidual subjects, since ‘usual’ concentrations of to have the potential to be used in epidemiologi-
fluoride have not yet been established due to the cal surveys.

1 McDonnell ST, O’Mullane D, Cronin M, 12 Levy FM, Bastos JR, Buzalaf MA: Nails 23 Krechniak J: Fluorides in drinking water
MacCormac C, Kirk J: Relevant factors as biomarkers of fluoride in children of and hair. Fluoride 1975;8:38–40.
when considering fingernail clippings as fluoridated communities. J Dent Child 24 Zakrzewska H, Brzezińska M, Orowicz
a fluoride biomarker. Community Dent (Chic) 2004;71:121–125. W, Samujło D, Wójcik A, Kolanus A:
Health 2004;21:19–24. 13 Fukushima R, Rigolizzo DS, Maia LP, Content of fluoride in hair and hoofs of
2 Whitford GM: Monitoring fluoride Sampaio FC, Lauris JR, Buzalaf MA: wild boars and deer from Western
exposure with fingernail clippings. Sch- Environmental and individual factors Pomerania as a bioindicator of environ-
weiz Monatsschr Zahnmed 2005;115: associated with nail fluoride concentra- mental pollution. Ann Acad Med Stetin
685–689. tion. Caries Res 2009;43:147–154. 2004;50(suppl 1):100–103.
3 Biological markers in environmental 14 Buzalaf MA, Vilhena FV, Iano FG, Grizzo 25 Schamschula RG, Duppenthaler JL, Sugar
health research. Committee on Biologi- L, Pessan JP, Sampaio FC, Oliveira RC: E, Un PS, Toth K, Barmes DE: Fluoride
cal Markers of the National Research The effect of different fluoride concen- intake and utilization by Hungarian chil-
Council. Environ Health Perspect 1987; trations and pH of dentifrices on plaque dren: associations and interrelationships.
74:3–9. and nail fluoride levels in young chil- Acta Physiol Hung 1988;72:253–261.
4 Grandjean P: Biomarkers in epidemiol- dren. Caries Res 2009;43:142–146. 26 Kono K, Yoshida Y, Watanabe M, Orita
ogy. Clin Chem 1995;41:1800–1803. 15 Buzalaf MAR, Rodrigues MHC, Pessan Y, Dote T, Bessho Y: Urine, serum and
5 World Health Organization: Fluorides JP, Leite AL, Arana A, Villena RS, Forte hair monitoring of hydrofluoric acid
and oral health. Report of a WHO expert FD, Sampaio FC: Biomarkers of fluoride workers. Int Arch Occup Environ Health
committee and oral health status and in children exposed to different sources 1993;65:S95–S98.
fluoride use. World Health Organ Tech of systemic fluoride. J Dent Res 2011;90: 27 Kono K, Yoshida Y, Watanabe M,
Rep Ser 1994;846:37. 215–219. Watanabe H, Inoue S, Murao M, Doi K:
6 Buzalaf MA, Caroselli EE, Cardoso de 16 Balazova G: Long-term effect of fluoride Elemental analysis of hair among hydro-
Oliveira R, Granjeiro JM, Whitford GM: emission upon children. Fluoride 1971; fluoric acid exposed workers. Int Arch
Nail and bone surface as biomarkers for 4:85–88. Occup Environ Health 1990;62:85–88.
acute fluoride exposure in rats. J Anal 17 Schamschula RG, Sugar E, Un PS, Toth 28 Mandinic Z, Curcic M, Antonijevic B,
Toxicol 2004;28:249–252. K, Barmes DE, Adkins BL: Physiological Carevic M, Mandic J, Djukic-Cosic D,
7 Whitford GM, Sampaio FC, Arneberg P, indicators of fluoride exposure and utili- Lekic CP: Fluoride in drinking water and
von der Fehr FR: Fingernail fluoride: a zation: an epidemiological study. Com- dental fluorosis. Sci Total Environ 2010;
method for monitoring fluoride expo- munity Dent Oral Epidemiol 1985;13: 408:3507–3512.
sure. Caries Res 1999;33:462–467. 104–107. 29 Ophaug R: Determination of fluorine in
8 Correa Rodrigues MH, de Magalhaes 18 Taves DR: Separation of fluoride by biological materials: reaction paper. Adv
Bastos JR, Rabelo Buzalaf MA: Finger- rapid diffusion using hexamethyldisilox- Dent Res 1994;8:87–91.
nails and toenails as biomarkers of sub- ane. Talanta 1968;15:969–974. 30 Balazova G, Lipkova V: Evaluation of
chronic exposure to fluoride from denti- 19 Whitford GM: The Metabolism and Tox- some health parameters in children in
frice in 2- to 3-year-old children. Caries icity of Fluoride, ed 2. Basel, Karger, the vicinity of an aluminium factory.
Res 2004;38:109–114. 1996. Fluoride 1974;7:88–93.
9 Buzalaf MA, Pessan JP, Alves KM: Influ- 20 Sampaio FC, Whitford GM, Arneberg P, 31 Villa A, Anabalon M, Zohouri V, Magu-
ence of growth rate and length on fluo- von der Fehr FR: Validation of fingernail ire A, Franco AM, Rugg-Gunn A: Rela-
ride detection in human nails. Caries fluoride as a biomarker for dental fluo- tionships between fluoride intake, uri-
Res 2006;40:231–238. rosis (abstract 72). Caries Res 2003;37: nary fluoride excretion and fluoride
10 Czarnowski W, Krechniak J: Fluoride in 291–292. retention in children and adults: an
the urine, hair, and nails of phosphate 21 de Almeida BS, da Silva Cardoso VE, analysis of available data. Caries Res
fertiliser workers. Br J Ind Med 1990;47: Buzalaf MA: Fluoride ingestion from 2010;44:60–68.
349–351. toothpaste and diet in 1- to 3-year-old 32 Whitford GM: Intake and metabolism of
11 Feskanich D, Owusu W, Hunter DJ, Wil- Brazilian children. Community Dent fluoride. Adv Dent Res 1994;8:5–14.
lett W, Ascherio A, Spiegelman D, Mor- Oral Epidemiol 2007;35:53–63. 33 Rao HV, Beliles RP, Whitford GM,
ris S, Spate VL, Colditz G: Use of toenail 22 Fukushima R, Sampaio FC, Buzalaf Turner CH: A physiologically based
fluoride levels as an indicator for the risk MAR: Risk assessment of external con- pharmacokinetic model for fluoride
of hip and forearm fractures in women. tamination of nails with fluoride uptake by bone. Regul Toxicol Pharma-
Epidemiology 1998;9:412–416. (abstract 146). Caries Res 2008;42:235. col 1995;22:30–42.

64 Pessan · Buzalaf
34 Turner CH, Boivin G, Meunier PJ: A 42 Mousny M, Banse X, Wise L, Everett ET, 52 Vieira AP, Hancock R, Limeback H, Maia
mathematical model for fluoride uptake Hancock R, Vieth R, Devogelaer JP, R, Grynpas MD: Is fluoride concentra-
by the skeleton. Calcif Tissue Int 1993; Grynpas MD: The genetic influence on tion in dentin and enamel a good indica-
52:130–138. bone susceptibility to fluoride. Bone tor of dental fluorosis? J Dent Res 2004;
35 Richards A, Mosekilde L, Sogaard CH: 2006;39:1283–1289. 83:76–80.
Normal age-related changes in fluoride 43 Dequeker J, Declerck K: Fluor in the 53 Nakagaki H, Koyama Y, Sakakibara Y,
content of vertebral trabecular bone – treatment of osteoporosis: an overview Weatherell JA, Robinson C: Distribution
relation to bone quality. Bone 1994;15: of thirty years clinical research. Schweiz of fluoride across human dental enamel,
21–26. Med Wochenschr 1993;123:2228–2234. dentine and cementum. Arch Oral Biol
36 Ishiguro K, Nakagaki H, Tsuboi S, et al.: 44 Ekstrand J, Spak CJ: Fluoride pharma- 1987;32:651–654.
Distribution of fluoride in cortical bone cokinetics: its implications in the fluo- 54 Lima-Arsati YB, Martins CC, Rocha LA,
of human rib. Calcif Tissue Int 1993;52: ride treatment of osteoporosis. J Bone Cury JA: Fingernail may not be a reliable
278–282. Miner Res 1990;5(suppl 1):S53–S61. biomarker of fluoride body burden from
37 Parkins FM, Tinanoff N, Moutinho M, 45 Arnala I, Alhava EM, Kauranen P: dentifrice. Braz Dent J 2010;21:91–97.
Anstey MB, Waziri MH: Relationships of Effects of fluoride on bone in Finland: 55 Pessan JP, Pin ML, Martinhon CC, de
human plasma fluoride and bone fluo- histomorphometry of cadaver bone from Silva SM, Granjeiro JM, Buzalaf MA:
ride to age. Calcif Tissue Res 1974;16: low and high fluoride areas. Acta Orthop Analysis of fingernails and urine as bio-
335–338. Scand 1985;56:161–166. markers of fluoride exposure from den-
38 Eble DM, Deaton TG, Wilson FC Jr, 46 Lan CF, Lin IF, Wang SJ: Fluoride in tifrice and varnish in 4- to 7-year-old
Bawden JW: Fluoride concentrations in drinking water and the bone mineral children. Caries Res 2005;39:363–370.
human and rat bone. J Public Health density of women in Taiwan. Int J Epide- 56 Spate VL, Morris JS, Baskett CK, et al.:
Dent 1992;52:288–291. miol 1995;24:1182–1187. Determination of fluoride in human
39 Suzuki Y: The normal levels of fluorine 47 Chachra D, Limeback H, Willett TL, nails via cyclic instrumental neutron
in the bone tissue of Japanese subjects. Grynpas MD: The long-term effects of activation analysis. J Radioanalyt Nucl
Tohoku J Exp Med 1979;129:327–336. water fluoridation on the human skel- Chem 1994;179:27–33.
40 Everett ET, McHenry MA, Reynolds N, eton. J Dent Res 2010;89:1219–1223. 57 Schmidt CW, Leuschke W: Fluoride con-
Eggertsson H, Sullivan J, Kantmann 48 Weatherell JA: Uptake and distribution tent in fingernails of individuals with
C, Martinez-Mier EA, Warrick JM, of fluoride in bones and teeth and the and without chronic fluoride exposure.
Stookey GK: Dental fluorosis: variability development of fluorosis; in Barltrop W, Fluoride 1990;23:79–82.
among different inbred mouse strains. Burland BL (eds): MIneral Metabolism 58 Clarkson J, Watt RG, Rugg-Gunn AJ, et
J Dent Res 2002;81:794–798. in Paediatrics. Oxford, Blackwell, 1969, al: Proceedings: 9th World Congress on
41 Carvalho JG, Leite AL, Yan D, Everett ET, pp 53–70. Preventive Dentistry (WCPD): ‘Commu-
Whitford GM, Buzalaf MA: Influence of 49 Weatherell JA, Hallsworth AS, Robinson nity Participation and Global Alliances
genetic background on fluoride metabo- C: Fluoride in the labial surface of per- for Lifelong Oral Health for All,’ Phuket,
lism in mice. J Dent Res 2009;88: manent anterior teeth. Caries Res 1972; Thailand, September 7–10, 2009. Adv
1054–1058. 6:67–68. Dent Res 2010;22:2–30.
50 Weatherell JA, Robinson C, Hallsworth
AS: Changes in the fluoride concentra-
tion of the labial enamel surface with
age. Caries Res 1972;6:312–324.
51 Richards A, Likimani S, Baelum V, Fejer-
skov O: Fluoride concentrations in
unerupted fluorotic human enamel. Car-
ies Res 1992;26:328–332.

Prof. Juliano Pelim Pessan

Department of Pediatric Dentistry and Public Health
Araçatuba Dental School, São Paulo State University
Rua José Bonifácio, 1193
16015–050 Araçatuba – SP (Brazil)
Tel. +55 18 3636 3274, E-Mail

Biomarkers of Fluoride Exposure 65

Fluoride Intake, Metabolism and Toxicity
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 66–80

Acute Toxicity of Ingested Fluoride

Gary Milton Whitford
Department of Oral Biology, Medical College of Georgia, Augusta, Ga., USA

Abstract cal course of an overexposure include the chemical com-

This chapter discusses the characteristics and treatment pound (e.g. NaF, MFP, etc.), the age and acid-base status
of acute fluoride toxicity as well as the most common of the individual, and the elapsed time between exposure
sources of overexposure, the doses that cause acute tox- and the initiation of treatment. While fluoride has well-
icity, and factors that can influence the clinical outcome. established beneficial dental effects and cases of serious
Cases of serious systemic toxicity and fatalities due to toxicity are now rare, the potential for toxicity requires
acute exposures are now rare, but overexposures causing that fluoride-containing materials be handled and stored
toxic signs and symptoms are not. The clinical course of with the respect they deserve.
systemic toxicity from ingested fluoride begins with gas- Copyright © 2011 S. Karger AG, Basel
tric signs and symptoms, and can develop with alarming
rapidity. Treatment involves minimizing absorption by As is true of virtually all substances to which
administering a solution containing calcium, monitoring humans are exposed, including water, oxygen
and managing plasma calcium and potassium concen- and table salt, exposure to high amounts of fluo-
trations, acid-base status, and supporting vital functions. ride can cause adverse effects. It is a toxicological
Approximately 30,000 calls to US poison control centers axiom that such effects are due to the level of ex-
concerning acute exposures in children are made each posure to the substance, not to the substance it-
year, most of which involve temporary gastrointestinal self. Compared to the first half of the 20th century,
effects, but others require medical treatment. The most cases of serious fluoride toxicity are uncommon
common sources of acute overexposures today are dental today. At that time, sodium fluoride was used as a
products – particularly dentifrices because of their rela- pesticide and rat poison and commonly found in
tively high fluoride concentrations, pleasant flavors, and homes, hospitals and elsewhere. Because sodium
their presence in non-secure locations in most homes. For fluoride powder resembles flour, powdered sugar,
example, ingestion of only 1.8 ounces of a standard fluori- baking powder, sodium bicarbonate and similar
dated dentifrice (900–1,100 mg/kg) by a 10-kg child deliv- products, there were many accidental poisonings.
ers enough fluoride to reach the ‘probably toxic dose’ (5 Sodium fluoride was also used in a large number
mg/kg body weight). Factors that may influence the clini- of suicides [1].
Table 1. Details of 3 deaths caused by ingestion of fluoride-containing dental products

Age Body weight, Sex Dose, Comment Reference

kg mg/kg

27 months not reported M 3.1–4.51 ingested <100 0.5-mg fluoride Dukes [8]
tablets; death occurred 5 days

3 years 12.5 M 16 ingested <200 1.0-mg fluoride Eichler et al. [7]

tablets; vomited; death occurred
7 h later

3 years not reported M 24–351 swallowed 4% SnF2 rinse solution; Church [6]
vomited; death occurred 3 h later

1 Calculated using the 3rd and 97th percentiles for body weight of 3-year-old boys.

One of the most remarkable accidental poi- Doses Causing Serious Toxicity: Rationale for
sonings occurred at the Oregon State Hospital the Probably Toxic Dose
[2]. About 10 gallons of scrambled eggs were
mistakenly prepared with 17 pounds of sodium Based on the sketchy information that could be
fluoride instead of powdered milk. There were gathered after the mass poisoning at the Oregon
263 cases of acute poisoning, of which 47 were State Hospital, Lidbeck et al. [2] thought the acute
fatal. It was not possible to estimate the amounts lethal dose of fluoride was over 100 mg/kg. Hodge
of fluoride that were ingested by those affected, and Smith [1] estimated that the ‘certainly lethal
but the well-known signs and symptoms devel- dose’ was between 32 and 64 mg/kg or 5–10 g of
oped rapidly. Extremely severe nausea, bloody sodium fluoride for a 70-kg person. Dreisbach [5]
vomiting and diarrhea occurred almost imme- estimated the lethal dose at 6–9 mg/kg. These dif-
diately. General collapse accompanied by pallor, ferent estimates probably were due in large part
weakness, shallow breathing, weak heart sounds, to uncertainty about the amounts of fluoride that
wet cold skin, cyanosis and equally dilated pu- were actually ingested by the victims.
pils soon followed. When these signs were pro- Church [6], Eichler et al. [7] and Dukes [8] did
nounced, death almost always occurred with- not attempt to estimate the acute lethal dose, but
in 2–4 h. When death was delayed for up to 20 did present dosages in their case reports (table
hours, muscle paralysis, carpopedal spasm and 1). The case described by Dukes [8] was unusual
spasm of the extremities occurred. More recent because of the small dose and the length of time
reports of serious acute toxicity have indicated from ingestion to death. A 27-month-old child in-
that muscular and cardiovascular problems are gested an unknown number (but fewer than 100)
related to electrolyte imbalances, particularly of 0.5-mg fluoride tablets and experienced respi-
severe hypocalcemia and hyperkalemia [3, 4]. ratory failure. With treatment, the boy’s condition
A progressive, mixed respiratory and metabolic improved, but he died 5 days after ingesting the
acidosis develops as kidney function and respi- tablets. The amount of fluoride ingested was ap-
ration fail. proximately 50 mg, and the dose was estimated

Acute Toxicity 67
at between 3.1 and 4.5 mg/kg. A 3-year-old boy 1% calcium chloride or calcium gluconate or, if
died in a hospital after ingestion of about 200 1.0- these solutions are not available, as much milk as
mg fluoride tablets [7]. This child vomited imme- can be tolerated. These actions should be taken
diately, seemed to recover completely, but then as soon as possible because fluoride is rapidly ab-
collapsed and died 7 h after the swallowing the sorbed from the stomach and intestines. At the
tablets. The ingested dose was approximately 16 same time, the hospital should be informed that
mg/kg. The case described by Church [6] was a a case of fluoride toxicity is in progress so that
3-year-old boy who swallowed 4% stannous fluo- appropriate therapeutic interventions are in place
ride rinse from a small cup in a dental clinic. The when the patient arrives. Expeditious treatment is
child vomited immediately, had a convulsive sei- essential because severe cases often progress rap-
zure and died 3 h later. The ingested dose was es- idly toward death.
timated at between 24 and 35 mg/kg, but the ab- In cases of life-threatening toxicity, which must
sorbed dose was lower because of vomiting. be judged by the clinical signs and symptoms be-
The report by Eichler et al. [7] also described cause the exact amount of fluoride ingested, i.e.
108 non-fatal cases of fluoride toxicity in chil- the dose, is almost never known, an airway and
dren, most of whom had ingested fluoride tablets. intravenous line should be established immedi-
As the ingested dose increased from less than 0.5 ately upon arrival at the hospital. Blood samples
mg/kg to ‘more than 5.0 mg/kg’, the percentage of should be obtained upon arrival and then hour-
patients with symptoms increased from 15 to 71. ly for the measurement of serum fluoride, blood
The symptoms included nausea, vomiting and fa- pH and gases, and serum chemistry – including
tigue. One child died as described above. Based on calcium and potassium in particular. Intravenous
this report and those of Dukes [8] and Bayless and administration of calcium gluconate to prevent
Tinanoff [9], it is concluded that the probably toxic hypocalcemia, glucose to reverse hyperkalemia,
dose (PTD) for fluoride is 5.0 mg/kg. The PTD is and sodium lactate or sodium bicarbonate to
defined as the minimum dose that could cause se- minimize acidosis and increase urinary flow and
rious or life-threatening systemic signs and symp- pH in order to increase the urinary excretion rate
toms and that should trigger immediate therapeu- of fluoride should be given as required. Oxygen
tic intervention and hospitalization. This does not therapy, artificial respiration, electrocardiac con-
mean that doses lower than 5.0 mg/kg should be version and hemodialysis may be required. These
regarded as innocuous. various measurements and treatments should
continue until the vital signs have stabilized and
the serum chemistry values have normalized for
Treatment at least 24–48 h.

The treatment for serious or potentially life-

threatening cases of acute fluoride toxicity must Sources of Fluoride
attempt to minimize absorption from the GI tract,
increase urinary excretion and maintain the vital As discussed above, the acute dose of fluoride
signs within levels compatible with life [1, 3, 4, that may cause serious systemic toxicity is 5 mg/
9]. If vomiting has not occurred, it should be in- kg (11 mg/kg of sodium fluoride). This is called
duced unless the patient is unconscious (to avoid the ‘probably toxic dose’ (PTD). It is obvious that
aspiration into the lungs). Because of the strong optimally fluoridated water (ca. 1.0 mg/l) cannot
affinity of calcium for fluoride, absorption can be cause acute toxicity since 5 liters of water would
slowed and reduced by the oral administration of have to be ingested for every kg of body weight.

68 Whitford
Accidental overfeeds resulting in water fluoride was estimated that he had consumed 17.9 mg/kg
concentrations sufficient to cause acute toxicity, during the course of the day.
however, have occurred. Tragic examples such as these are rare today,
One example involved kidney patients in a he- but less severe episodes are not. Thousands of
modialysis center [3]. Approximately 1,000 gal- calls involving suspected or actual fluoride over-
lons of hydrofluorosilicic acid accidentally leaked doses are made to US poison control centers each
into the public drinking water supply, which in- year and, up until 2005 when the publication of
creased the fluoride concentration to a peak of the annual reports ended, they were compiled and
30 ppm. Two days later, the concentrations were published in the American Journal of Emergency
lower but still elevated, and 8 patients under- Medicine [11]. Table 2 shows a summary of the
going hemodialysis developed gastrointestinal 2000–2003 data. Among the more than 30,000 re-
symptoms. During dialysis 1 patient developed ports for which a medical outcome was recorded
pressure-like chest pain, difficulty breathing, in each year, approximately 7,250 were classified
nausea, vomiting, sweating, diarrhea and numb- as ‘none’. A ‘minor’ medical outcome means that
ness in the right arm. The dialysis procedure was there were some symptoms, but they were mini-
interrupted, the symptoms gradually improved mal (nausea, vomiting, dizziness) or required no
and the patient went home. However, 14 h later treatment. A ‘moderate’ outcome means that the
the difficulty in breathing returned. On the way symptoms were more severe, more prolonged or
to the hospital the patient had cardiopulmonary more of a systemic nature and that some treat-
arrest, but was successfully resuscitated despite ment was usually required. A ‘major’ outcome
severe hyperkalemia (10.5 meq/l; normal 4.0–5.5 means that the patient survived, but the toxicity
meq/l) and a markedly elevated plasma fluoride was life-threatening and/or resulted in residual
concentration (0.4 mg/l; normal 0.02–0.04 mg/l). disability.
Another 1 of the 8 dialysis patients was found In each of the 4 years, approximately 80% of
dead at home. the reports made to poison control centers in-
Another example occurred in 1992 in Hooper volved toothpastes or mouthwashes (categories D
Bay, a small village in Alaska near the Bering Sea and E) and nearly 90% involved young children.
[10]. The water supply for the 470 residents was Between 394 and 471 were treated in a health care
a single tank from which water was obtained and facility each year. Approximately 1,400 cases were
carried home for domestic use. One weekend, the classified as ‘minor’ or ‘moderate’ and 1–4 cases as
water fluoridation equipment failed, resulting in ‘major’ each year, and there was 1 death, a suicide
a peak concentration of 150 ppm. Approximately reported in the 2002 publication. These figures
296 residents experienced several of the less se- indicate that the most commonly used fluoride-
vere symptoms listed above (nausea, vomiting, containing dental products are sources of poten-
abdominal cramps). One woman was evacuat- tial toxicity.
ed by air after 2 days of vomiting and diarrhea.
Upon arrival at the hospital her serum calcium
was 5.2 mg/dl (one half normal) and her serum Fluoride Exposure from Dental Products
fluoride was 9.1 mg/l (400 times normal). She re-
covered after several days of treatment, but her Table 3 shows the fluoride concentrations in sev-
brother was found dead at home after prolonged eral dental products, the amounts that are usually
vomiting and diarrhea. He had attempted to re- used, and the amounts that contain the PTD (5
main hydrated by continuing to drink the water. mg/kg). The data indicate that there is little or no
His postmortem serum calcium was 4.9 mg/dl. It danger of systemic toxicity when the products are

Acute Toxicity 69
Table 2. Fluoride-related reports made to US poison control centers (2000–2003)

Year Category Reports Treated in Medical outcome

health care
facility none minor moderate major death

2000 A 3,681 191 1,028 337 15 1 0

B 158 17 51 6 0 0 0
C 2,637 90 578 58 3 0 0
D 22,291 360 5,505 1,262 46 0 0
E 2,073 34 520 78 5 0 0
Total 30,840 692 7,682 1,741 69 1 0

2001 A 3,635 179 947 306 19 0 0

B 484 49 99 11 4 0 0
C 2,176 99 529 34 3 0 0
D 22,790 391 5,014 1,328 38 4 0
E 2,179 32 464 77 2 0 0
Total 3,1264 750 7,053 1,756 66 4 0

2002 A 3,730 169 911 274 7 0 0

B 354 20 61 12 1 1 0
C 2,364 89 429 45 1 0 0
D 24,089 411 4,852 1,218 40 1 1
E 2,557 60 532 93 2 0 0
Total 33,092 749 6,785 1,642 51 2 1

2003 A 3,541 139 809 233 11 0 0

B 250 37 70 14 2 0 0
C 2,437 80 483 42 5 0 0
D 24,812 405 5,413 1,337 44 1 0
E 3,401 43 751 72 2 0 0
Total 34,441 704 7,526 1,698 64 1 0

Categories: A = electrolytes and minerals; B = adult vitamins; C = pediatric vitamins; D = fluoride toothpaste; E = fluo-
ride mouthwash. See Watson et al. [11].

used in the usual amounts and as recommend- however, these products are sometimes ingest-
ed. Rulings by the US Consumer Product Safety ed in excessive amounts. Unsupervised children
Commission (CPSC) that require child-resistant may drink mouthwash from the bottle, eat tooth-
caps for fluoride mouthwashes and most pre- paste from the tube or tablets from the bottle. An
scriptions for dietary fluoride supplements have 18-ounce bottle of 0.05% NaF mouthwash (510
reduced the risk of systemic toxicity from these ml) contains 124 mg of fluoride, an amount 2.37
products. These child-resistant caps, however, are times more than the PTD for a 10-kg child. Thus,
not ‘child proof ’ and ‘should be regarded as your ingestion of 7.6 ounces (215 ml) could cause se-
last line of defense’ [12]. As indicated in table 3, rious toxicity. Although the mouthwash labels

70 Whitford
Table 3. Fluoride contents of dental products and their relationships to the PTD

Product Concentration of salt fluoride Amount of product Amount of product

and fluoride usually containing the PTD for
used child weighing

% % ppm product fluoride 10 kg 20 kg

NaF 0.05 0.023 230 10 ml 2.3 mg 215 ml 430 ml
NaF 0.20 0.091 910 10 ml 9.1 mg 55 ml 110 ml
SnF2 0.40 0.097 970 10 ml 9.7 mg 50 ml 100 ml

NaF 0.22 0.10 1,000 1g 1.0 mg 50 g 100 g
MFP 0.76 0.10 1,000 1g 1.0 mg 50 g 100 g

Topical gel
NaF (APF, tray) 2.72 1.23 12,300 5 ml 61.5 mg 4 ml 8 ml
SnF2 (brush) 0.40 0.097 970 1 ml 0.97 mg 50 ml 100 ml

NaF tablet
0.25 mg – – – 1/day 0.25 mg 200 tabs 400 tabs
0.50 mg – – – 1/day 0.50 mg 100 tabs 200 tabs
1.00 mg – – – 1/day 1.00 mg 50 tabs 100 tabs

PTD = 5 mg/kg i.e. the amount of ingested fluoride that could cause serious or life-threatening systemic effects and
that should trigger immediate therapeutic intervention and hospitalization; MFP = sodium monofluorophosphate;
APF = acidulated phosphate fluoride. The average body weights of 1-year-old and 6-year-old children are approxi-
mately 10 and 20 kg, respectively.

in the USA are required by the Food and Drug All of these products should be kept out of the
Administration to specify that children under the reach of small children and secured with child re-
age of 6 years should not use a fluoride mouth- sistant caps.
wash, they do have access to them in many homes. Topical acidulated phosphate fluoride (APF)
Dentifrices are available in tubes containing up to gels and foams contain fluoride at a concentra-
8.2 ounces (232 g) so a 1,000-ppm product con- tion of 12.3 mg/ml (12,300 ppm). APF treat-
tains 232 mg of fluoride. Ingestion of only 1.76 ments are rarely given to 1-year-old children, but
ounces (50 g) by a 10-kg child provides enough they may be given to 2-year-old children (aver-
fluoride to reach the PTD. As for fluoride tablets, age body weight 12.4 kg). If maxillary and man-
the American Dental Association guidelines state dibular stock trays are loaded with 5 ml of gel
that up to 480 0.25-mg tablets, 240 0.50-mg tab- in each tray, then 123 mg of fluoride would be
lets, and 120 1.0-mg tablets may be prescribed placed in the mouth, which exceeds the PTD for
per household [13]. These numbers of tablets a 2-year-old by a factor of 2, so swallowing one
and amounts of fluoride contained in them ex- half of the applied gel would reach the PTD. The
ceed the PTD for both 10-kg and 20-kg children. currently recommended procedure for APF gel

Acute Toxicity 71
treatments minimizes the amount of gel that is
likely to be swallowed and it should be followed. 100

The recommendations are: (1) use the minimum

amount of gel required to cover the teeth; (2) use 80
no more than 2 ml of gel in each stock tray; (3) if
custom-made trays are used, then use only 5–10 60

Percent HF
pK = 3.45
drops of gel in each tray; (4) seat the child in an
upright position with the head inclined slightly
forward to discourage swallowing; (5) use a sa-
liva ejector throughout the procedure; and (6) al-
low the child to expectorate for 30 s after the pro-
cedure. When this procedure is used, the risk of
even temporary stomach irritation due to swal- 0
1 2 3 4 5 6 7
lowing is minimal. It is also worth noting that,
Solution pH
while the APF foams have the same fluoride con-
centration as the gels, much less fluoride is placed
Fig. 1. Relationship between the pH of a solution and the
in the patient’s mouth because much of the vol-
percentage of fluoride that exists as HF.
ume is occupied by air [14].
Whitford et al. [15] reported that, when in an
acidic solution, the threshold fluoride concen-
tration that produces histological and functional
damage to the canine stomach mucosa is between but endoscopic examination 2 h after the gel
19 and 95 mg/l or 1.0 and 5.0 mmol/l. It should treatment revealed mucosal petechiae or ero-
be noted that the pH of APF products is approxi- sions. Biopsies of the mucosa showed histological
mately 3.5. The pK of hydrofluoric acid (HF) is changes, including dilation of the gastric pits, lo-
3.4, so nearly 50% of the fluoride in the gel or foam calized losses of surface epithelium and bleeding,
exists as undissociated HF (ca. 6,000 mg/l), a mol- in 9 of the 10 subjects.
ecule that is very irritating to the stomach mucosa
and at a concentration far above that known to
damage the stomach mucosa. Unless care is taken Effects on the Stomach
to reduce swallowing even small amounts, nausea
and vomiting may occur. Following the ingestion of a large amount of fluo-
Products intended for self-application at home ride, or a relatively small amount in a small volume
may also cause damage to the stomach. Spak et al. (i.e. a high concentration), the first organ to be af-
[16] examined the histological effects of a 0.42% fected is the stomach. The contents of the stomach
fluoride gel (4,200 mg/l) with a pH of 6.5. Ten have a distinctly acidic pH. Between meals the pH
subjects added 1.5 g of the gel to each custom- is usually between 2 and 4, while during and for
made maxillary and mandibular tray (a total of 1–2 h after meals it is between 1 and 2. Figure 1
12.6 mg of fluoride) for a 5-min topical applica- shows the relationship between the pH of a so-
tion. The average amount of fluoride not recov- lution and the percentage of the total amount of
ered from the mouth was 5.1 mg or 40% of the fluoride in the solution that is combined with hy-
amount applied which was due to using more drogen ions to form HF, a weak acid whose pKa
than the recommended volume for custom-made is 3.45. The Henderson-Hasselbalch equation is
trays. None of the subjects experienced nausea, used to calculate the relative concentrations of

72 Whitford
ionic fluoride and HF at different pH values. The dogs [21] (fig. 2). Through a midline abdominal
equation is: incision, a portion of the stomach from the great-
er curvature with its gastrosplenic blood supply
pH = pK + log([F–]/[HF]) intact was mounted in a two-compartment Lucite
chamber with the mucosal surface facing upwards
When the pH of the stomach contents is 4.0, 22% as described by Moody and Durbin [22]. The sep-
of the fluoride is in the form of HF. When the pH tum of the chamber divided the tissue into two
is lower than 2.0, more than 95% is in the form of halves of equal surface area (14.2 cm2) so that
HF. HF is a highly diffusible and permeating mol- control and test solutions could be placed side-
ecule that diffuses down its concentration gradi- by-side on the mucosa. This permitted a compari-
ent to cross cell membranes and epithelia includ- son of ion fluxes across the mucosa as well as di-
ing the stomach mucosa, a tissue that is relatively rect observation of any gross changes that might
impermeable to most other ingested substances occur.
[17]. Upon entering the gastric mucosa where The mucosa on one side of the chamber was
the pH is close to neutrality, HF dissociates im- exposed to a saline solution with or without 10
mediately to release fluoride and hydrogen ions. mmol/l sodium fluoride (190 ppm) at a pH of
In sufficiently high concentrations, these ions can 6.2 for 22 15-min collection periods. At this pH,
disrupt the structure and function of the stomach only 0.2% of the fluoride is in the form of HF. The
[18–20]. mucosa on the other side was exposed to a sa-
It is important to understand that the effects line solution acidified with 0.1 n HCl (pH 1.6)
of fluoride on the stomach are dependent on the also with or without 10 mmol/l sodium fluoride.
concentration of fluoride (actually the HF con- At this pH, 98.6% of the fluoride is in the form
centration as discussed below) in contact with of HF. The solutions without fluoride served as
the mucosa, not on the ingested dose (i.e. mg/kg). the negative control solutions. The fluxes of wa-
For example, if 0.5 liters of water containing 5 mg ter (determined by changes in the concentration
of fluoride (10 ppm or 0.5 mmol/l) were ingested, of 14C-inulin) and sodium, potassium and hydro-
it is unlikely that even the mildest of symptoms gen ions were not affected by 10 mmol/l fluoride
would be felt and there would be only minimal when the solution pH was 6.2 and the gross ap-
or no adverse effects on the stomach. If, however, pearance of mucosa remained normal through-
5 ml of water containing 5 mg of fluoride (1,000 out the 5.5-hour study.
ppm or 52.6 mmol/l) were ingested, nausea and In contrast, the water and ion fluxes increased
perhaps vomiting and dizziness would be expe- immediately upon exposure to the pH-1.6 solu-
rienced by many people. In each case the same tion containing 10 mmol/l fluoride. The water,
amount of fluoride would have been ingested, but sodium and potassium fluxes were positive, i.e.
the effects would be quite different. In fact, fol- they were directed from the mucosa into the test
lowing the systemic absorption of fluoride, the solution. The hydrogen ion fluxes, however, were
toxic effects on internal organs are also depen- negative, i.e. directed into the mucosa, which in-
dent on the tissue concentration of fluoride but, dicated that HF was diffusing down its concen-
because the concentrations in these organs are tration gradient from the solution in the cham-
almost never known, the dose for systemic ef- ber and thus carrying hydrogen and fluoride ions
fects is usually expressed in terms of body weight into the tissue. When the mucosa was repeated-
(mg/kg). ly exposed to the control solution (without fluo-
The effects of pH on the gastric effects of flu- ride) after the exposures to 10 mmol/l fluoride,
oride were tested using in situ experiments with the fluxes were reduced slightly but they did not

Acute Toxicity 73
2.0 pH 6.2
pH 1.6

Water (ml/15 min)




C1 C2 C3 C4 T1 T2 C5 C6 C7 C8 T3 T4 C9 C10 C11 C12 T5 T6 C13 C14 C15 C16

H (µmol/15 min)



C1 C2 C3 C4 T1 T2 C5 C6 C7 C8 T3 T4 C9 C10 C11 C12 T5 T6 C13 C14 C15 C16


K (µmol/15 min)

C1 C2 C3 C4 T1 T2 C5 C6 C7 C8 T3 T4 C9 C10 C11 C12 T5 T6 C13 C14 C15 C16


Na (µmol/15 min)



C1 C2 C3 C4 T1 T2 C5 C6 C7 C8 T3 T4 C9 C10 C11 C12 T5 T6 C13 C14 C15 C16

Fig. 2. Effects of solution pH (6.2 or 1.6) on water and ion fluxes from the gastric mucosa of the dog
in response to exposure to sodium fluoride. The control solutions (labeled C1–C16) contained no
fluoride. The test solutions (labeled T1–T6) contained 10 mmol/l sodium fluoride. Fresh solutions
were placed on the mucosa every 15 min.

74 Whitford
return to the baseline values. Further, there was Factors That Influence Toxic Effects
an obvious increase in the secretion of mucus fol-
lowed by swelling (edema) and localized areas Chemical Compound
of hemorrhage within the first few minutes after The compounds of fluoride vary greatly with
placing the pH-1.6 solution containing fluoride respect to their solubilities. Very insoluble
on the mucosa. These findings made it clear that compounds such as calcium fluoride, cryolite
changes in the structure and function of the gas- (Na3AlF6), hydroxyfluorapatite and fluorapatite
tric mucosa are caused by exposure to high con- are poorly absorbed from the GI tract. Because
centrations of HF, and that equally high concen- of this their LD50 values, as determined in stud-
trations of ionic fluoride are without such effects. ies with laboratory animals, are much higher than
Using the same model, experiments were done those of highly soluble compounds such as sodi-
to determine the threshold HF concentration for um fluoride, fluorosilicic acid (H2SiF6) and sodi-
gastric toxicity [15]. The solution used on the um fluorosilicate (Na2SiF6), the three compounds
control side contained 50 mmol/l sodium chlo- that are commonly used to fluoridate drinking
ride in 0.1 n HCl (pH 1.6). The same solution was water at low fluoride concentrations.
used on the test side but also contained fluoride The highest fluoride concentrations to which
in the form of HF at 1.0, 5.0 or 10.0 mmol/l. The most people are regularly exposed are found in
water and ion fluxes throughout the 4-hour study certain dental products, particularly dentifrices
(16 15-min collection periods) and the gross and which typically contain 1,000–1,500 mg /kg fluo-
histological appearances on the control side were ride. The compounds most often added to denti-
normal. On the test side, exposure to the solution frices are sodium fluoride and disodium mono-
containing 1.0 mmol/l fluoride as HF produced fluorophosphate or MFP (Na2PO3F). The fluoride
small but statistically non-significant increases in in MFP is covalently bonded to the phosphorus.
the fluxes and only minor changes in the appear- Its release from MFP is slow in water and denti-
ance of the mucosa. However, all fluxes increased frices, but rapid in the presence of phosphatases
significantly upon exposure to the solution con- found in the intestine, plasma and internal or-
taining 5.0 mmol/l HF. Mucus secretion increased gans [23]. This was demonstrated in an experi-
as did the redness and swelling of the mucosa. ment with 2 groups of rats that were given fluo-
Subsequent exposure to the 10.0 mmol/l HF so- ride intravenously (2.0 mg/kg) as sodium fluoride
lution caused these effects to increase. Upon mi- or MFP [24]. Three blood samples were collected
croscopic examination, the thickness of the sur- at 10, 30 and 60 min after administration of the
face mucus layer and the epithelium were greatly doses. The plasma fluoride concentrations in the
reduced. In some sections, evidence of surface cell 2 groups were virtually identical, which indicated
exfoliation was seen indicating cell degeneration the complete hydrolysis of fluoride from MFP pri-
and necrosis. It was concluded that the threshold or to the 10-min blood collections.
concentration for adverse effects of fluoride in a Based on the time courses of plasma concen-
strongly acidic solution, i.e. HF, is more than 1.0 trations, however, there is evidence that the ab-
mmol/l (19 ppm) but less than 5.0 mmol/l (95 sorption of orally administered fluoride when
ppm). This explains why swallowed APF gel is given as MFP is somewhat slower than that from
damaging to the gastric mucosa and should be sodium fluoride [23]. The delayed absorption and
avoided. The total fluoride concentration (i.e. lower peak plasma fluoride concentrations appear
ionic fluoride plus HF) is 1.23% or 647 mmol/l to be due to the limited amount of phosphatase
(12,300 ppm) and, at pH 3.5, the HF concentra- activity in the stomach compared to the intestine.
tion is 305 mmol/l (6,104 ppm). There are, however, no significant differences in

Acute Toxicity 75
the percentages of the doses that are ultimately that ‘. . .professional organizations and regulato-
absorbed systemically. The limited phosphatase ry agencies should not endorse the policy of add-
activity in the stomach was also indicated in a ing greater amounts of fluoride, as MFP, to den-
study with humans by Müller et al. [25]. The sub- tal products based on the concept that fluoride in
jects ingested sodium fluoride or monofluoro- the form of MFP is less hazardous than that in the
phosphate tablets for 1 week. The gastric mucosa form of NaF’.
was then examined with a gastroscope. No sig-
nificant damage was found in the MFP group, but Age
acute hemorrhages and free blood were found in Maynard et al. [30] and Mornstad [31] report-
the NaF group. ed that, compared to adult laboratory animals,
The relative absence of gastric phosphatase young laboratory animals are more resistant to
activity also explained the lack of functional and the acute toxic effects of fluoride. It is not known
structural effects of MFP on the canine mucosa whether this is true for humans but there is rea-
[21]. Using the same split-chamber method de- son to think that it is. As mentioned earlier, the
scribed above, the mucosa on one side of the systemic effects of acute exposures to high doses
chamber was exposed to 10 mmol/l F as NaF and of fluoride are directly related to the concentra-
to 10 mmol/l MFP on the other side for two 15- tions in plasma and the target organs. The rate
min periods. The immediate effects on the NaF of removal of fluoride from plasma and the tar-
side included large increases in the fluxes of wa- get organs depends almost entirely on the rates
ter, sodium and potassium, increased mucus se- of uptake by calcified tissues, which contain 99%
cretion and increased mucosal swelling and red- of the fluoride in the body, and excretion in the
ness. None of these effects occurred on the MFP urine. Therefore, any factor that increases these
side except for a slight and transient increase in rates should reduce the severity of the acute toxic
the potassium flux. effects.
Theoretically, the lower peak plasma fluoride Miller and Phillips [32] fed 3 groups of rats
concentrations could reduce the acute toxicity of a diet with the same fluoride concentration for
MFP compared to sodium fluoride. In their study 4.5 months. The rats in 1 group began consum-
with rats, Shourie et al. [26] reported that the 24- ing the diet when they were weaned (21 days of
h LD50 doses for these two compounds were 75 age) while 2 other groups started at 9 or 20 weeks
and 36 mg/kg, respectively. In their study with of age. At the end of the 4.5-month feeding pe-
mice, Lim et al. [27] reported LD50 values of 94 riods, the bone fluoride concentrations were in-
and 44 mg/kg, respectively. These findings were versely related to the age at which the rats entered
used to support the increase of the total fluoride the study – the younger rats had higher concen-
amount as MFP above the limit established by the trations. Similar results were reported by Zipkin
American Dental Association (260 mg total flu- and McClure [33] and Suttie and Philips [34] who
oride per tube of dentifrice). More recent stud- used rats and by Weidmann and Weatherell [35]
ies, however, could not confirm such differences. who used rabbits.
In their study with rats, Gruninger et al. [28] re- Whitford [21] used dogs of different ages (4
ported LD50 values of 102 and 98 mg/kg for MFP weeks, 6 months and several years) and infused
and sodium fluoride – with mice the values were isotonic solutions containing sodium fluoride
54 and 58 mg/kg, respectively. In their study with intravenously for 20 min and then the infusion
rats, Whitford et al. [29] reported LD50 values of pump was turned off. Blood samples were collect-
84.3 and 85.5 mg/kg for MFP and sodium fluo- ed 12 times over 6 h. Each dog received the same
ride. Based on these results, the authors stated dose in terms of body weight (5.0 mg/kg). The

76 Whitford
1,400 2.0


AUC (mmol • h • I–1

1,200 1.2


1,000 0.4

0 1 2 3 4 5 6
800 Hours
Plasma [F] (µM)

600 6 months
4 weeks



0 1 2 3 4 5 6
Times after starting F infusion of 5 mg/kg (h)

Fig. 3. Effect of age, or stage of skeletal development, on arterial plasma fluoride concentrations
in dogs. The cumulative areas under the plasma-time curves (AUC) are shown [21].

peak plasma fluoride concentrations were 305, excretion. The results appear to be explained by
1,004 and 1,367 μmol/l, respectively, and the areas the fact that the crystallites in developing bone are
under the time-plasma concentration curves were loosely organized and not compacted as in mature
markedly higher in the older animals as well (fig. bone; thus, providing a much greater surface area
3). Similar results were found in a study with rats for the rapid uptake of fluoride.
that were 23 days or 6.5 months of age [21].
These age-related differences in plasma fluo- Acid-Base Status
ride concentrations were due almost entirely to There is a considerable body of evidence show-
a greater rate of fluoride uptake by the bone of ing that the rates of fluoride absorption from the
younger animals, and not to differences in urinary GI tract and excretion in the urine, as well as the

Acute Toxicity 77
distribution of fluoride across the membranes of oral administration of ammonium chloride or so-
individual cells, are all dependent on pH gradi- dium bicarbonate, respectively. In the latter study
ents [21]. These observations are best explained alkalosis was established by the intravenous infu-
by the fact that wherever there is a pH gradient sion of sodium bicarbonate with or without aceta-
across an epithelium or cell membrane separating zolamide (Diamox®) during fluoride exposure. In
two adjacent fluid compartments, there will also each study, the alkalotic animals tolerated signifi-
be a difference in the HF concentrations. In cases cantly higher fluoride doses and survived twice
of distribution across cell membranes, HF, a high- as long while the fluoride infusions continued.
ly diffusible and permeating molecule, will rap- They also maintained higher blood pressures,
idly diffuse down its concentration gradient un- heart rates, glomerular filtration rates and renal
til the HF concentrations in the extracellular and clearances of fluoride at any given plasma fluo-
intracellular compartments are equal. The result ride concentration. They died with significantly
is that the concentration of ionic fluoride will be higher plasma fluoride concentrations and lower
higher in the more alkaline compartment which, tissue-to-plasma fluoride concentration ratios. It
in nearly all tissues, is the extracellular fluid. was concluded that metabolic alkalosis, whether
The magnitude of the pH gradient across cell present before fluoride exposure or imposed dur-
membranes can be increased by alkalinizing the ing the development of toxicity, favorably influ-
extracellular fluids which can be done, for ex- enced the clinical course and that establishing an
ample, by hyperventilating or the administration alkalosis and a more alkaline urinary pH should
of sodium bicarbonate or sodium lactate. These be added to the therapeutic regimen.
actions increase extracellular pH more than in-
tracellular pH [36]. Consequently the extracellu-
lar concentration of HF falls to a greater extent Conclusion
than that in the intracellular compartment which
causes HF to diffuse from cells into the extracel- As used in this chapter, acute toxicity means ad-
lular fluids. Thus the intracellular concentration verse effects that occur within a short period of
of fluoride is reduced, thereby lowering the effects time following the oral administration or inges-
of fluoride on intracellular enzymes and transport tion of a single dose of fluoride or multiple dos-
systems. Further, the rate of fluoride absorption es within a few hours. The stomach – where the
from the GI tract is inversely related to the pH of effects range from some degree of nausea to ab-
the stomach contents while the rate of urinary flu- dominal pain, bloody vomitus and diarrhea – is
oride excretion is directly related to the pH of the the first organ affected, with those latter effects
renal tubular fluid. For all these reasons, it would signaling impending systemic effects that should
be expected that the acute toxic effects of fluoride be regarded as potentially fatal. Serious systemic
would be reduced by increasing the pH of the ex- toxic effects may occur when the amount ingest-
tracellular fluids and urine. ed reaches the PTD of 5.0 mg/kg. It is difficult,
This expectation was confirmed in two studies however, to know the exact amount that was in-
with rats. The effects of pre-existing acid-base dis- gested, so estimations about the degree of toxic-
turbances on acute fluoride toxicity [37] and of al- ity and judgments about what actions and treat-
kalosis imposed during the development of acute ments should be taken typically depend on the
fluoride toxicity [38] were tested. In each study, early clinical signs and symptoms.
fluoride was infused intravenously until death oc- Today the most common sources of significant
curred. In the former study, acidosis or alkalosis amounts of ingested fluoride available to most
was established before fluoride exposure by the persons are fluoride-containing dental products

78 Whitford
(tables 2, 3). Compared to the frequency with warning labels on toothpaste boxes and tubes, and
which these dental products are used, cases of rational recommendations for the safe use of pro-
acute toxicity are exceedingly rare. In the US, for fessionally applied products have contributed to
example, it can be reasonably estimated that fluo- the safe use of these products.
ride dentifrices are used at least once each day by Nevertheless, ingestion of excessive amounts of
at least 200 million people and that fluoride mouth fluoride, whether accidental or intentional, does
rinses, dietary supplements and professionally ap- occur and moderate and major health outcomes
plied topical products are used thousands of times can follow. It is for this reason that healthcare per-
each day. By comparison, only about 700 persons, sonnel, as well as parents, should be familiar with
most of whom were young children, were treat- the characteristics of acute fluoride toxicity, the
ed in a healthcare facility each year from 2000 sources of potentially toxic doses and how to limit
through 2003, and fewer than 100 experienced access to them especially by children, the amounts
moderate or major health outcomes (table 2). The of ingested fluoride that can cause harmful effects,
use of child-resistant caps on fluoride mouth rinse and what to do in case of overexposures. This
bottles and most fluoride supplement containers, chapter is a source of such information.

1 Hodge HC, Smith FA: Biological effects 10 Gessner BD, Beller M, Middaugh JP, 17 Whitford GM, Pashley DH. Fluoride
of inorganic fluorides; in Simons JH Whitford GM: Acute fluoride poisoning absorption: the influence of gastric acid-
(ed): Fluorine Chemistry. New York, from a public water system. New Engl ity. Calc Tiss Int 1984;36:302–307.
Academic Press, 1965, pp 1–364. J Med 1994;330:95–99. 18 Easmann RP, Steflik DE, Pashley DH,
2 Lidbeck WL, Hill IB, Beeman JA: Acute 11 Watson WA, Litovitz TL, Klein-Schwartz McKinney RV, Whitford GM: Surface
sodium fluoride poisoning. J Am Med W, Rodgers GC, Youniss J, Reid N, Rouse changes in rat gastric mucosa induced
Assoc 1943;121:826–827. WG, Rembert RS, Borys D: 2003 annual by sodium fluoride: a scanning electron
3 McIvor M, Baltazar RF, Beltran J, Mower report of the American Association of microscopic study. J Oral Pathol 1984;13:
MM, Wenk R, Lustgarten J, Salomon J: Poison Control Centers Toxic Exposure 255–264.
Hyperkalemia and cardiac arrest from Surveillance System. Am J Emerg Med 19 Easmann RP, Pashley DH, Birdsong NL,
fluoride exposure during hemodialysis. 2004;22:335–404. McKinney RV, Whitford GM: Recovery
Am J Cardiol 1983;51:901–902. 12 US Consumer Product Safety Commis- of rat gastric mucosa following single
4 McIvor ME, Cummings CC, Mower MM, sion: Press release (March 26, 2009). fluoride dosing. J Oral Pathol 1985;14:
Baltazar RF, Wenk RE, Lustgarten JA, 2009. 779–792.
Salomon J: The manipulation of potas- 13 Burrell KH, Chan JT: Systemic and topi- 20 Pashley DH, Allison NB, Easmann R,
sium efflux during fluoride intoxication: cal fluorides; in Ciancio SG (ed): ADA McKinney RV, Horner JA, Whitford GM:
implications for therapy. Toxicology Guide to Dental Therapeutics, ed 2. Chi- The effects of fluoride on the gastric
1985;37:233–239. cago, ADA, 2000, p 233. mucosa of the rat. J Oral Pathol 1984;13:
5 Dreisbach RH: Handbook of Poisoning. 14 Whitford GM, Adair SM, McKnight 535–545.
Los Altos, Langer, 1980, pp 210–213. Hanes CM, Perdue EC, Russell CM: 21 Whitford GM: Gastric toxicity; in Myers
6 Church LI: Fluorides: use with caution. Enamel uptake and patient exposure to HM (ed): The Metabolism and Toxicity
Maryland Dent Assoc J 1976;19:106. fluoride: comparison of APF gel and of Fluoride. Basel, Karger, 1996.
7 Eichler HG, Lenz K, Fuhrmann M, foam. Pediat Dent 1995;17:199–203. 22 Moody FG, Durbin RP: Effects of glycine
Hruby K: Accidental ingestion of NaF 15 Whitford GM, Pashley DH, Garman RH: and other instillates on concentration of
tablets by children: report of a poison Effects of fluoride on structure and gastric acid. Am J Physiol 1965;209:
control center and one case. Int J Clin function of canine gastric mucosa. Dig 122–126.
Pharmacol Ther Toxicol 1982;20: Dis Sci 1997;42:2146–2155. 23 Ericsson Y: Monofluorophosphate physi-
334–338. 16 Spak C-J, Sjostedt S, Eleborg L, Veress B, ology: general considerations. Caries Res
8 Dukes MNG: Side Effects of Drugs. Perbeck L, Ekstrand J: Studies of human 1983;17(suppl 1):46–55.
Oxford, Excerpta Medica, 1980, p 354. gastric mucosa after application of 24 Whitford GM, Pashley DH, Allison NB:
9 Bayless JM, Tinanoff N: Diagnosis and 0.42% fluoride gel. J Dent Res 1990;69: Monofluorophosphate physiology: dis-
treatment of acute fluoride toxicity. J Am 426–429. cussion. Caries Res 1983;17(suppl 1):
Dent Assoc 1985;110:209–211. 69–76.

Acute Toxicity 79
25 Müller P, Schmid K, Warnecke G, Set- 29 Whitford GM, Birdsong-Whitford NL, 35 Weidmann SM, Weatherell JA: The
nikar I, Simon B: Sodium fluoride- Finidori C: Acute toxicity of sodium flu- uptake and distribution of fluorine in
induced gastric mucosal lesions: com- oride and monofluorophosphate sepa- bones. J Path Bact 1959;78:243–255.
parison with sodium rately or in combination in rats. Caries 36 Boron WF, Roos A: Comparison of
monofluorophosphate. Gastroenterology Res 1990;24:121–126. microelectrode, DMO and methylamine
1992;30:252–254. 30 Maynard EA, Downs WL, LeSher MF: methods of measuring intracellular pH.
26 Shourie KL, Hein JW, Hodge HC: Pre- University of Rochester Atomic Energy Am J Physiol 1976;231:799–809.
liminary studies on the caries inhibiting Project Quarterly Technical Report. UP- 37 Reynolds KE, Whitford GM, Pashley
potential and acute toxicity of sodium 164, 1951, p 73–77. DH: Acute fluoride toxicity: the influ-
monofluorophosphate. J Dent Res 31 Mornstad H: Acute sodium fluoride tox- ence of acid-base status. Toxicol Appl
1950;29:529–533. icity in rats in relation to age and sex. Pharmacol 1978;45:415–427.
27 Lim JK, Renaldo GJ, Chapman P: LD50 of Acta Pharmacol Toxicol 1975;37:425– 38 Whitford GM, Reynolds KE, Pashley
SnF2, NaF and Na2PO3F in the mouse 428. DH: Acute fluoride toxicity: influence of
compared to the rat. Caries Res 32 Miller RF, Phillips PH: The enhancement metabolic alkalosis. Toxicol Appl Phar-
1978;12:177–179. of the toxicity of sodium fluoride in the macol 1979;50:31–39.
28 Gruninger SE, Clayton R, Chang SB, rat by high dietary fat. J Nutr
Siew C: Acute oral toxicity of dentifrice 1955;56:447–454.
fluorides in rats and mice. J Dent Res 33 Zipkin I, McClure FJ: Deposition of fluo-
1988;67(special issue, abstr 1769):334. rine in the bones and teeth of the grow-
ing rat. J Nutr 1952;47:611–620.
34 Suttie JW, Phillips PH: The effect of age
on fluorine deposition in the femur of
the rat. Arch Biochem 1959;83:355–359.

Gary Milton Whitford, PhD, DMD

Department of Oral Biology, School of Dentistry, Medical College of Georgia
1120 15th Street
Augusta, GA 30912 (USA)
Tel. +1 706 721 0388, E-Mail

80 Whitford
Fluoride Intake, Metabolism and Toxicity
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 81–96

Chronic Fluoride Toxicity: Dental Fluorosis

Pamela DenBesten ⭈ Wu Li
Department of Orofacial Sciences, School of Dentistry, University of California, San Francisco, Calif., USA

Abstract followed by hypomineralized bands. Enhanced mineral

Dental fluorosis occurs as a result of excess fluoride precipitation with local increases in matrix acidity may
ingestion during tooth formation. Enamel fluorosis and affect maturation stage ameloblast modulation, poten-
primary dentin fluorosis can only occur when teeth are tially explaining the dose-related decrease in cycles of
forming, and therefore fluoride exposure (as it relates to ameloblast modulation from ruffle-ended to smooth-
dental fluorosis) occurs during childhood. In the perma- ended cells that occur with fluoride exposure in rodents.
nent dentition, this would begin with the lower incisors, Specific cellular effects of fluoride have been implicated,
which complete mineralization at approximately 2–3 but more research is needed to determine which of these
years of age, and end after mineralization of the third changes are relevant to the formation of fluorosed teeth.
molars. The white opaque appearance of fluorosed As further studies are done, we will better understand
enamel is caused by a hypomineralized enamel sub- the mechanisms responsible for dental fluorosis.
surface. With more severe dental fluorosis, pitting and Copyright © 2011 S. Karger AG, Basel
a loss of the enamel surface occurs, leading to second-
ary staining (appearing as a brown color). Many of the Excess fluoride ingestion results in dental fluoro-
changes caused by fluoride are related to cell/matrix sis. The mechanisms affected by long-term chron-
interactions as the teeth are forming. At the early matu- ic exposure to low levels of fluoride are likely to
ration stage, the relative quantity of amelogenin protein differ from those affected by acute exposures to
is increased in fluorosed enamel in a dose-related man- high levels of fluoride [1–3]. Some mechanisms
ner. This appears to result from a delay in the removal of affected by lower chronic fluoride levels, result-
amelogenins as the enamel matures. In vitro, when fluo- ing in enamel fluorosis, are likely to be specific
ride is incorporated into the mineral, more protein binds to this uniquely mineralizing tissue, while others
to the forming mineral, and protein removal by protei- may also affect other cells and tissues.
nases is delayed. This suggests that altered protein/min- Enamel fluorosis refers to fluoride-related al-
eral interactions are in part responsible for retention of terations in enamel, which occur during enam-
amelogenins and the resultant hypomineralization that el development. These alterations become more
occurs in fluorosed enamel. Fluoride also appears to severe with increasing fluoride intake, and time
enhance mineral precipitation in forming teeth, result- of exposure. The severity of fluorosis is related
ing in hypermineralized bands of enamel, which are then to the concentration of fluoride in the plasma,
a b c d

Fig. 1. Dental fluorosis. a Mild with slight accentuation of the perikymata. b Moderate, showing a white opaque ap-
pearance. c Moderate, white opaque enamel with some discoloration and pitting. d Severe.

considered to be in equilibrium with the tissue with much less known about the potential effects
fluid that bathes the enamel organ [4, 5]. Plasma of fluoride on dentin formation. Therefore, most
fluoride levels are influenced by many factors, in- of the focus will be on enamel fluorosis. The sec-
cluding total fluoride intake, type of intake (i.e. tions of this chapter comprise:
ingested vs. inhaled), renal function, rate of bone 1 Clinical manifestation, treatment and preven-
metabolism, metabolic activity, etc. [6]. In addi- tion of dental fluorosis;
tion to these variables, genetic factors have been 2 Etiology and prevalence of dental fluorosis;
shown to dictate the severity of enamel fluorosis 3 Pathology, pathogenesis and mechanism of
in mice [7]. dental fluorosis.
In humans, plasma fluoride concentrations
resulting from long-term ingestion of 1–10 ppm
fluoride in the drinking water range from 1 to 10 Clinical Manifestation, Treatment and
μmol/l. Fluorotic changes can be obtained in inci- Prevention of Dental Fluorosis
sors of rodents drinking water containing 25–100
ppm fluoride; these doses also elevate plasma flu- Clinical Manifestations of Dental Fluorosis
oride levels to 3–10 μmol/l, similar to those found Clinically, mild cases of dental fluorosis are char-
to cause fluorosis in humans. A complicating fac- acterized by a white opaque appearance of the
tor in assessing the exact dose, or determining enamel, caused by increased subsurface poros-
the stages of enamel formation most sensitive to ity (fig. 1). The earliest sign is a change in color,
fluoride, is that fluoride incorporated into bone showing many thin white horizontal lines running
is gradually released by continuous bone remod- across the surfaces of the teeth, with white opaci-
eling [5, 8]. Levels of plasma fluoride as low as ties at the newly erupted incisal end. The white
1.5 μmol/l (resulting from fluoride release from lines run along the ‘perikymata’, a term referring
bone) are still capable of inducing mild enamel to transverse ridges on the surface of the tooth,
fluorosis in the rat incisor after the initial expo- which correspond to the incremental lines in the
sure ends [4, 8]. enamel known as Striae of Retzius [10, 11].
The effects of chronic fluoride exposure have At higher levels of fluoride exposure, the white
also been linked to effects on other tissues and sys- lines in the enamel become more and more de-
tems [9]. However, in this chapter, we will focus fined and thicker. Some patchy cloudy areas and
primarily on the effects of fluoride on tooth devel- thick opaque bands also appear on the involved
opment. The largest body of research has investi- teeth. With increased dental fluorosis, the entire
gated the effects of fluoride on enamel formation, tooth can be chalky white and lose transparency

82 DenBesten · Li
Table 1. Fluorosis index of H.T. Dean (1942)

Score Criteria

Normal (0) The enamel represents the usual translucent semivitriform type of structure. The surface is smooth,
glossy, and usually of a pale creamy white color.
Questionable The enamel discloses slight aberrations from the translucency of normal enamel, ranging from a
(0.5) few white flecks to occasional white spots. This classification is utilized in those instances where a
definite diagnosis of the mildest form of fluorosis is not warranted and a classification of ‘normal’ is
not justified.
Very mild (1) Small opaque, paper white areas scattered irregularly over the tooth but not involving as much as
25% of the tooth surface. Frequently included in this classification are teeth showing no more than
about 1–2 mm of white opacity at the tip of the summit of the cusps of the bicuspids or second
Mild (2) The white opaque areas in the enamel of the teeth are more extensive but do not involve as much
as 50% of the tooth.
Moderate (3) All enamel surfaces of the teeth are affected, and the surfaces subject to attrition show wear.
Brown stain is frequently a disfiguring feature.
Severe (4) Includes teeth formerly classified as ‘moderately severe and severe.’ All enamel surfaces are
affected and hypoplasia is so marked that the general form of the tooth may be affected. The
major diagnostic sign of this classification is discrete or confluent pitting. Brown stains are
widespread and teeth often present a corroded-like appearance.

As reproduced in National Academy of Sciences [p.169, 16].

[10, 12]. With higher fluoride doses or prolonged decay, most likely because of the uneven surface
exposure, deeper layers of enamel are affected; the or loss of the outer protective layer [15].
enamel becomes less well mineralized. Damage
to the enamel surface occurs in patients with Fluorosis Indices
moderate-to-severe degrees of enamel fluorosis. In 1942, H.T. Dean developed an index to de-
Teeth can erupt with pits, with additional pitting scribe and diagnosis enamel fluorosis [16, 17].
occurring with posteruptive enamel fracture. He scored the fluorotic teeth into 6 categories ac-
In the individuals with moderate dental fluo- cording to their clinical manifestations, includ-
rosis, yellow to light brown staining is observed in ing normal teeth, which were given a score of 0
the areas of enamel damage. In very severe cases, (table 1). Using this index, Dean [17, 18] deter-
the enamel is porous, poorly mineralized, stains mined the ‘optimal’ concentration of fluoride in
brown, and contains relatively less mineral and drinking water (1 ppm), where caries incidence
more proteins than sound enamel. Severely fluo- decreased and with a minimal level of dental
rosed enamel can easily chip posteruptively dur- fluorosis.
ing normal mechanical use [13, 14]. Although This classification is still the ‘gold standard’,
teeth with mild dental fluorosis may be more re- though other indices have been developed – in-
sistant to dental decay because of the higher lev- cluding the widely used Thylstrup and Fejerskov
els of fluoride contained in the enamel surface, Fluorosis Index (TFI) [19], which has an ex-
severely fluorosed teeth are more susceptible to panded range for the more severe forms of dental

Dental Fluorosis 83
Table 2. Dietary reference intakes for fluoride

Age groups Reference weight, Adequate intake, Tolerable upper intake,

kg (lb) mg/day mg/day

Infants 0–6 months 7 (16) 0.01 0.7

Infants 7–12 months 9 (20) 0.50 0.9
Children 1–3 years 13 (29) 0.70 1.3
Children 4–8 years 22 (48) 1.00 2.2
Children 9–13 years 40 (88) 2.00 10.0
Boys 14–18 years 64 (142) 3.00 10.0
Girls 14–18 years 57 (125) 3.00 10.0
Males ≥19 years 76 (166) 4.00 10.0
Females ≥19 years 61 (133) 3.00 10.0

Source: US National Academy of Sciences. Institute of Medicine. Food and Nutrition Board [p. 288, 21].

fluorosis. This index is a 10-point classification dose suggested by USEPA is 0.06 mg fluoride/kg/
system to characterize dental fluorosis affecting day, which is the estimate of daily exposure that is
bucal/lingual and occlusal surfaces and correlates likely to be without any appreciable risk of delete-
visual assessment with polarized and light micro- rious effects (any degrees of dental fluorosis) dur-
scopic analysis [19]. Dean’s index is expanded to ing a lifetime [20].
include: mild (TFI = 1–3), moderate (TFI = 4–5) Specific guidelines for different ages (table 2)
and severe (TFI = 6–9) [19]. were published by the US Food and Nutrition
Board of the Institute of Medicine in 1997, rec-
Treatment of Dental Fluorosis ommending total daily fluoride intakes [21]. In
The treatments for fluorotic teeth are limited. For this guideline, the suggested total daily exposure
the mildest forms of fluorosis (TFI 1, 2) bleaching, dosage for infants younger than 6 months of age
to make the color of the tooth surface uniform, of 0.01 mg fluoride/day in all drinks and food is
can be recommended. Treatments for moderate lower than the USEPA recommended reference
dental fluorosis include microabrasion, where the dose. These guidelines suggest greater attention
outer affected layer of enamel is abraded from the should be given to the total fluoride intake of in-
tooth surface in an acidic environment. Composite fants from water used to dilute infant formulas,
restorations combined with microabrasion or ap- foods and other supplement sources.
plication of aesthetic veneers can be used for the
patients with TFI ≥5, while for the cases with TFI
8–9, prosthetic crowns may be necessary [19]. Etiology and Prevalence of Dental Fluorosis

Prevention of Dental Fluorosis There are multiple sources of fluoride and all have
Dental fluorosis can be limited or prevented by the potential to cause dental fluorosis – includ-
following the ‘recommended limits for fluo- ing natural fluoride, artificial or added fluoride
ride exposure’, suggested by US Environmental in drinking water and dental products, as well as
Protection Agency (USEPA) [20]. The reference occupation-related exposures [22].

84 DenBesten · Li
10 100

Caries experience per child in DMF teeth

Percent of population affected

8 80

by dental fluorosis
6 mild Mild 60

4 40
Fig. 2. Concentrations of fluoride in
2 20
drinking water are related to caries
incidence in children and severity Severe
of dental fluorosis. Adapted from a 0 0
report of the Department of Health 0 1 2 3 4 5 6
and Human Services of US (1991) ppm fluoride

Natural Sources of Fluoride Causing Dental fluoride, 1.4 million drank water with fluoride
Fluorosis between 2.0 and 3.9 mg/l and 200,000 people in-
Dental fluorosis resulting from high fluoride lev- gested water with fluoride concentrations ≥4.0
els in underground water is an issue in specific re- mg/l [16]. Some areas have extremely high con-
gions of the world. Fluoride can exist in an ionized centrations of fluoride in drinking water – such as
form in ground waters, and in areas where the soil in Colorado (11.2 mg/l), Oklahoma (12.0 mg/l),
lacks calcium – such as occurs in areas with high New Mexico (13.0 mg/l) and Idaho (15.9 mg/l)
levels of granite or gneiss – relatively high fluoride [9] – though water with levels higher that those
levels are detected in groundwater. When the level recommended by the USEPA are monitored and
of fluoride is above 1.5 mg/l (1.5 ppm) in drinking are not used for human consumption.
water, dental fluorosis can occur. In some parts of
Africa, China, the Middle East and southern Asia Additional Sources of Fluoride Associated with
(India, Sri Lanka), as well as some areas in the Dental Fluorosis
Americas and Japan, high concentrations of ionic Two primary sources have been identified as
fluoride have been found in ground waters, veg- being potentially responsible for the prevalence
etables, fruit, tea and other crops, although drink- of dental fluorosis: fluoride in drinking water
ing water is usually the major source of the daily and fluoride-containing dental products. Since
fluoride intake [23]. The atmosphere in these ar- 1945, fluoride has been used as a supplement
eas may have high levels of fluoride from dust in in many public drinking water systems to con-
areas with fluoride-containing soils and gas, re- trol dental decay [24]. In 2000, approximate-
leased from industries, underground coal fires ly 162 million people (65.8% of the population
and volcanic activities [23]. served by public water systems) received wa-
In the USA, approximately 10 million people ter that contained fluoride ranging from 0.7 to
are exposed to naturally fluoridated public water. 1.2 mg/l (usually 1 mg/l), depending on the lo-
In 1993, it was reported that 6.7 million people cal climate. The level of fluoridation is lower in
drank water with fluoride concentrations ≤1.2 high-temperature areas as people usually drink
mg/l, 1.4 million drank water with 1.3–1.9 mg/l more water. The fluoridation of public drinking

Dental Fluorosis 85
60 NIDR, 1986–1987
Fig. 3. Change in dental fluorosis NHANES, 1999–2004
prevalence among children aged 39.6
12–15 years participating in 2 na- 40
tional surveys in the USA (1986– 30.2

1987 and 1999–2004). Dental fluoro-
sis (based on Dean’s fluorosis index) 19.7 17.2
is defined as: very mild, mild, mod-
erate or severe. Percentages do not
sum to 100 due to rounding. Error 8.6
bars = 95% CI. Sources: National 4.1 3.6
Health and Nutrition Examination 1.3
Survey (1999–2004) [27] and Unaffected Questionable Very mild Mild Moderate
National Survey of Oral Health in and severe
U.S. School Children (1986–1987) Dental fluorosis

water has significantly decreased the incidence or greater enamel fluorosis, up from 22.6% in the
of dental decay at a relatively low cost. In the 1986–1987 study (fig. 3) [27].
studies by Dean and colleagues completed in the The incidence of very mild and greater fluoro-
1930s, the risk of dental fluorosis at 1 ppm fluo- sis in persons aged 6–39 years was 19.79% in white
ride in drinking water was extremely low, par- non-Hispanics, 32.88% in black non-Hispanics,
ticularly in relation to the impact of fluoride on and 25.8% in Hispanics (table 3). The increased
dental caries (fig. 2) [25]. Following these stud- prevalence of fluorosis in black non-Hispanics
ies, water fluoridation was considered by the may suggest a genetic influence on fluorosis
US Centers for Disease Control to be 1 of the susceptibility.
10 great public health achievements in the 20th
century [26].
However, as fluoride has become more widely Pathology, Pathogenesis and Mechanism of
used in dental products (toothpastes, mouth rins- Dental Fluorosis
es, fluoride supplements) and been incorporated
into food sources (via fluoridated water), multi- The primary pathological finding of fluorosed
ple sources of fluoride exposure are now related enamel is a subsurface porosity, along with hy-
to the reported increase in the incidence of den- per- and hypomineralized bands within the form-
tal fluorosis. Even a small ‘pea-sized’ amount of ing enamel (fig. 4) [28–34]. Fluoride can also re-
toothpaste containing 1,450 ppm fluoride, would sult in mineralization-related effects on dentin
contain approximately 0.36–0.72 mg fluoride, formation.
which if consumed twice a day could contribute Severely fluorosed human dentin is charac-
to fluoride levels that would increase the risk of terized by a highly mineralized sclerotic back-
dental fluorosis in children [21]. In the USA, the ground pattern, scattered with hypomineralized
prevalence of dental fluorosis appears to be in- porous lesions primarily in the subsurface area.
creasing. In children aged 15–17 years, the 1999– Scanning electron microscope images show den-
2004 National Health and Nutrition Examination tin tubules with an irregular distribution and
Survey (NHANES) found 40.6% had very mild narrow and disrupted lumina, rather than the

86 DenBesten · Li
Table 3. Enamel fluorosis among persons aged 6–39 years by selected characteristics

Unaffected Questionable Very mild Mild Moderate/


% SE % SE % SE % SE % SE
Age group (years)
6–11 59.81 4.07 11.80 2.50 19.85 2.12 5.83 0.73 2.71 0.59
12–15 51.46 3.51 11.96 1.84 25.33 1.98 7.68 0.93 3.56 0.59
16–19 58.32 3.30 10.21 1.70 20.79 1.78 6.65 0.67 4.03 0.77
20–39 74.86 2.28 8.83 1.23 11.15 1.22 3.34 0.58 1.81 0.39

Male 67.65 2.63 9.99 1.45 15.65 1.52 4.58 0.54 2.12 0.39
Female 66.97 2.84 9.83 1.34 15.58 1.36 4.84 0.61 2.78 0.49

White, non-Hispanic 69.69 3.13 10.43 1.62 14.09 1.56 3.87 0.60 1.92 0.48
Black, non-Hispanic 56.72 3.30 10.40 2.16 21.21 2.16 8.24 0.82 3.43 0.54
Mexican-American 65.25 3.89 8.95 1.29 15.93 2.24 5.05 0.72 4.822 1.81

Poverty Status3
<100% FPL 68.02 3.21 10.67 1.64 14.28 1.73 4.07 0.69 2.97 0.66
100–199% FPL 66.92 2.91 9.11 1.79 16.11 1.46 5.21 0.78 2.65 0.56
≥200% FPL 66.88 2.75 10.73 1.33 15.56 1.56 4.83 0.50 2.00 0.37

Total 67.40 2.65 9.91 1.35 15.55 1.37 4.69 0.49 2.45 0.40

Data from National Health and Nutrition Examination Survey (1999–2002) [27] and calculated using Dean’s index. All
estimates are adjusted by age (single years) and sex to the USA 2000 standard population, except sex, which is
adjusted only by age.
1 Calculated using ‘other race/ethnicity’ and ‘other Hispanic’ in the denominator.
2 Unreliable estimate: the standard error is 30% the value of the point estimate, or greater.
3 Percentage of the federal poverty level (FPL), which varies by income and number of persons living in the


regular-appearing lumina seen in normal dentin the blood stream, possibly reducing the risk of
[35]. dental fluorosis by lowering serum fluoride levels
The pathogenesis of dental fluorosis is relat- [8, 39]. Nutrition is also important for controlling
ed to physiological conditions, including body the serum level of fluoride, as ions such as cal-
weight, rate of skeletal growth and remodeling, cium, magnesium and aluminum can reduce the
nutrition, and renal function [36–38]. Bone is a bioavailability of fluoride. A deficiency in these
reservoir of fluoride, as fluoride is incorporated ions in food can also affect (enhance) fluoride up-
in the forming apatite crystals, and this ion can take [40].
also be released from these crystals as bone re- Genetic background appears to have role in the
models. Therefore, rapid bone growth, as occurs pathogenesis of dental fluorosis. This may be the
in the growing child, will remove fluoride from reason why in human populations, individuals

Dental Fluorosis 87
fluoride susceptibility. Most of the studies of the
mechanisms of fluoride in forming fluorosed
enamel have used the rodent incisor or molars as a
model, as it is not possible to do similar studies us-
ing human teeth. The rodent incisor is a continu-
ously erupting tooth, with all stages of enamel for-
mation present in each tooth, whereas the molar
is a rooted tooth, which begins formation in utero.
As previously mentioned, though rodents require
the ingestion of much higher levels of fluoride in
the drinking water (10–20 times) as compared
to humans, the serum levels at which fluorosis is
formed in rodents and humans is similar.
Pre-secretory ameloblasts differentiate into se-
Fig. 4. Microradiograph of fluorosed enamel from cretory ameloblasts after the dentin matrix begins
Colorado Springs. Note the radiolucent outer third of the to mineralize. The pre-secretory ameloblasts and
enamel with a well-calcified surface layer. From Newbrun overlying cells of the enamel organ, including the
[96], reprinted with permission.
enamel knot, are thought to influence the tooth
morphogenesis. However, there is no evidence
that exposure of developing teeth to physiologi-
cal levels of fluoride in vivo [42] and in organ cul-
drinking water with similar fluoride contents have ture [43–46] affects tooth morphogenesis. Even
a wide range of severity of dental fluorosis (fig. 2). in teeth with severe fluorosis, the size and form of
Evidence for a genetic component to fluoride sus- the teeth are not changed [47].
ceptibility comes from work by Everett et al. [7], As the pre-ameloblasts differentiate to secreto-
which tested 12 different in-bred mouse strains ry ameloblasts, they begin to secrete enamel ma-
to compare their susceptibility to fluoride. Mouse trix proteins, and lay down a thin layer of apris-
teeth have been found to be an excellent model matic enamel deposited against mantle dentin. As
for human tooth formation, and in Everett’s study, the secretory ameloblast Tomes’ processes form,
they found that some mouse strains were high- the inner enamel layer, which constitutes the bulk
ly susceptible to fluoride-related dental fluorosis, of enamel, begins to be laid down. This enamel
while other strains were highly fluorosis resistant. matrix consists of prismatic enamel with rod (or
They concluded that there is a genetic component prisms) and interrod structures (interprismatic
to dental fluorosis susceptibility [41]. enamel) formed by the Tomes’ processes of fully
differentiated secretory ameloblasts. These cells
Stages of Tooth Formation and Stage-Specific secrete matrix protein (predominantly amelogen-
Effects of Chronic Fluoride Exposure ins) into the enamel space through which thin but
Fluoride is a single highly electronegative ion that long enamel crystals grow preferentially in length
interacts with the cells and matrix at the differ- in the wake of the retreating cells.
ent stages of enamel formation in relation to flu- Secretory stage ameloblasts exposed to high
oride dose and time of exposure. Tooth enamel chronic levels of fluoride have a somewhat dis-
development can be divided into 4 major stages: rupted morphology and increased numbers of
pre-secretory, secretory, transition and matura- vacuoles at the apical border. Chronic exposure to
tion stages, all with unique properties that affect fluoride in drinking water or repeated injections

88 DenBesten · Li
of moderate fluoride doses reduces the thickness Maturation ameloblasts of adult rat incisors
of enamel by about 10% [42, 48]. Although this [42] are shorter, and fluorotic enamel organs have
suggests that chronic exposure to fluoride reduces a disrupted maturation ameloblast modulation
biosynthesis of matrix by secretory ameloblasts, [42, 62, 63]. The first modulation bands that dis-
there is no evidence to support this [1, 49, 50]. appear during fluoride exposure are the most in-
Instead, the small reduction in enamel thickness cisal smooth-ended ameloblasts. At prolonged ex-
may be attributed to a limited disruption of vesic- posure other smooth-ended bands disappear one
ular transport in fluorotic secretory ameloblasts by one in an incisal to apical direction [62]. In ad-
and subsequent intracellular degradation of a mi- dition to changes in modulation, fluoride also re-
nor portion of the matrix by the lysosomal system duces the cyclic uptake of 45Ca labeling in a similar
[51–53]. Alternatively, the reduction in enamel pattern [62]. When fluoride exposure is discontin-
thickness may be related to an effect of fluoride ued, smooth-ended bands reappear starting from
on crystal elongation in the secretory stage. the youngest most apical part towards older more
At the end of secretion, the ameloblasts lose incisal bands. This suggests that the fluoride ef-
their Tomes’ process and deposit a final layer of fects on ameloblast modulation are reversible, and
aprismatic enamel with small crystals. The cells that the young modulating cells recover more rap-
transform via a short transitional stage, where idly than older ameloblasts. After eruption, the
enamel matrix proteins undergo rapid proteoly- enamel is exposed to mineral ions of the oral flu-
sis, leaving the porous enamel matrix characteris- ids, including fluoride, which can influence the
tic of this transition stage. composition of the outer layers of enamel.
Late secretory-transitional cell stage ame-
loblasts appear to be more sensitive to fluoride Direct Effects of Fluoride on Ameloblasts
than early and fully secretory ameloblasts. In Ameloblasts and tooth organs exposed to high
hamster molar tooth germs, a dose of 4.5 mg/kg (millimolar) levels of fluoride in vitro, which would
fluoride induces the late secretory to transitional be much greater than the micromolar levels of flu-
cells, but not early secretory ameloblasts to detach oride found in the plasma carrying fluoride ions
occasionally from the surface and form subam- to tooth organs in vivo, show many alterations.
eloblastic cysts. The enamel below the cysts under These include changes in the structure of early
late secretory ameloblasts will give rise to the shal- secretory ameloblasts, reduced protein synthesis,
low occlusal pits, often seen in severely fluorosed altered cell proliferation, apoptosis, stress-related
teeth in various species [47, 54–60]. This stage of protein upregulation and elevation of F-actin [64–
development is likely also to be associated with 67]. However, some of these same changes are not
the formation of accentuated perikymata that is readily apparent in vivo, and therefore, the effects
clinically the first sign of enamel fluorosis. of fluoride when examined in culture, must be
In the maturation stage, the ameloblasts mod- carefully analyzed for biological relevance.
ulate cyclically from cells with a smooth-ended However, there are in vitro data indicating that
to a ruffle-ended distal membrane, the latter ameloblasts can be sensitive to low levels of flu-
with characteristics of resorbing cells. During oride. Human primary enamel organ epithelial
this modulation, matrix proteins continue to be cells grown in culture show that exposure to fluo-
removed from the extracellular space, and min- ride levels as low as 5 μmol/l results in reduced
eralization increases to form a fully mineralized expression of the secretory stage matrix metallo-
enamel matrix. Amelogenin proteins are retained proteinase 20 (MMP-20) [68], mediated by JNK/
in the fluorosed rat enamel matrix at this stage of c-Jun signaling [69]. These results suggest that
enamel formation [50, 61]. fluoride may have specific effects on ameloblast

Dental Fluorosis 89
Secretory enamel showed no Transition/early maturation-stage
difference in proteins from enamel shows more proteins with
animals ingesting different ingestion of increasing amounts of
amounts of fluoride fluoride

Fig. 5. SDS PAGE separation of pro-

31,000 31,000
teins in secretory and maturation
stages of enamel matrix of fluoride-
treated and untreated rat tooth. A = 21,500 21,500
Standard; B = 0 ppm; C = 10 ppm; D
= 25 ppm; E = 50 ppm; F = 100 ppm. 14,000 14,000
From DenBesten [50], reprinted with A B C D E F A B C D E F

differentiation mediated through MAP-kinase protein constituting 90–95% of total proteins in

signaling. the enamel protein matrix [71]. Amelogenin and
Rodent studies have shown that ingestion of the other matrix proteins are hydrolyzed by ma-
fluoride alters the number of bands of smooth trix proteinases as enamel forms, allowing replace-
ended ameloblasts and their rate of modulation ment of the protein matrix with an organized hy-
in the maturation stage ameloblasts [42, 62]. droxyapatite structure. MMP-20 is the proteinase
However, there is currently no evidence to de- primarily responsible for the initial hydrolysis of
termine whether these changes in maturation amelogenins in the secretory enamel matrix, while
stage ameloblast modulation are a direct effect kallikrein 4 (KLK4) is the predominant proteinase
of fluoride, or more likely, in response to matrix- in the transition/maturation stage [72, 73].
mediated alterations related to fluoride exposure An analysis of proteolytic activity in enamel
to the developing enamel matrix. matrix, isolated from secretory and maturation
At extremely high levels of ingested fluoride stage rat enamel, showed a significantly reduced
(150 ppm) in the drinking water, ameloblasts have activity in early maturation stage enamel isolated
been shown to exhibit apoptosis and endoplasmic from rats ingesting 100 ppm fluoride (5–10 μm
reticulum stress responses [65]; however, at low- serum fluoride), as compared to control matura-
er levels (75 ppm) these effects were not noted. tion enamel [74]. This effect of fluoride ingestion
Further studies at lower fluoride levels will need in decreasing matrix proteinase activity correlates
to be done to determine whether this is a potential to an increased retention of amelogenin proteins
mechanism relevant to chronic fluoride toxicity in maturation stage fluorosed enamel in a dose-
in humans. dependent manner (fig. 5). Matrix proteins disap-
pear from nonfluorosed enamel in the maturation
Fluoride-Related Alterations of the Forming stage, but are retained in fluorosed enamel, with
Enamel Matrix May Indirectly Affect Ameloblast increased retention at higher levels of ingested
Function fluoride [48, 50].
The extracellular enamel matrix proteins include This retention of amelogenin proteins could
amelogenins, ameloblastin and enamelin, all of delay final mineralization of the enamel matrix,
which support and modulate enamel crystal for- contributing to subsurface hypomineralization
mation [70]. Amelogenin is the chief structural characteristic of fluorosed enamel. The reason for

90 DenBesten · Li
this retention of amelogenins is most likely relat-
ed to altered proteolytic activity in the fluorosed 100 ***
enamel matrix. **
80 **

Amelogenins (%)
Reduced Proteolytic Activity May Be due to the 60
Effects of Fluoride Incorporation into Growing
Enamel Crystals
Crystals in sound enamel are long, and the dy- 20
namics of enamel crystal growth, size of the crys-
tals and their shape are well controlled by matrix 0
0 100 1,000 2,000
proteins during enamel formation [75–77]. Some Fluoride concentrations (ppm)
studies report that crystals isolated from fluo-
rosed enamel have a significantly greater diameter
Fig. 6. Degradation of amelogenin adsorbed on apa-
than crystals in sound enamel, as determined by tite crystals by MMP-20. Amelogenins were pre-bound
high-resolution electron microscopy [78], X- ray to carbonated hydroxyapatite crystals containing differ-
diffraction of powdered enamel samples [79] or ent amounts of fluoride (X-axis) and then degraded by
scanning microscopy of fractured inner enamel MMP-20. Y-axis indicates the percentages of amelogenins
degraded by MMP-20 from apatite crystals as compared
specimens [80]. Some organ culture studies have
to the amount of amelogenin initially bound. Note the
shown large flattened hexagonal crystals mixed decreased degradation of amelogenin from the apatite
with many small irregularly shaped crystals in crystal surface as the concentration of fluoride in the apa-
hypermineralized areas [81, 82]. However, other tite increases (unpublished data).
studies reported no differences between fluorotic
and normal human crystals [28, 83].
There is, however, no doubt that the fluoride
content of crystals in fluorosed enamel is great- In further investigation of the role of fluoride
er than that of normal enamel. Fluoride substi- incorporation into apatite on amelogenin process-
tutes for hydroxyl groups in enamel carbonated ing, we characterized hydrolysis of amelogenins
hydroxyapatite crystals, altering the crystalline bound to fluoride-containing apatites by recom-
structures and surface characteristics. To deter- binant MMP-20 or KLK-4. When fluoride was
mine whether an increased fluoride content of the in solution, amelogenin hydrolysis by MMP-20
apatite crystals could affect matrix/proteinase in- was reduced only at 1,000 ppm (52 mm, which
teractions, we measured the binding of recombi- is far higher than physiological levels of fluoride
nant human amelogenin to synthetic carbonated in enamel fluids). However, incorporation of flu-
hydroxyapatite crystals. oride into apatite significantly delayed MMP-20
The initial rate of amelogenin binding and the hydrolysis of the adsorbed amelogenin in a dose-
total amount of amelogenin bound to fluoride- dependent manner (fig. 6) even at the lowest level
containing carbonated hydroxyapatite was greater of fluoride-containing apatite (100 ppm F). This
than that in the control carbonated hydroxyapa- same effect of reduced amelogenin hydrolysis was
tite [84]. These results suggest that fluoride incor- found when amelogenins were hydrolyzed from
poration into the crystal lattice alters the crystal fluoride-containing apatites with recombinant
surface to enhance amelogenin binding, poten- KLK-4 (unpublished results).
tially contributing to the increased amount of The levels of fluoride incorporated into the
amelogenin and the inhibition of crystal growth apatite crystals in these in vitro studies are bio-
in fluorosed enamel. logically relevant. Although the enamel fluid

Dental Fluorosis 91
surrounding the ameloblasts is likely to contain Abundant amelogenins generated by
no more than 10 μm (0.19 ppm) fluoride, fluo- secretory ameloblasts may be a potent contrib-
ride is incorporated into the growing crystals in utor to controlling pH at the secretory stage,
concentrations ranging from 10 ppm near the where the pH is maintained at neutral [76, 93].
dental-enamel junction to several thousand ppm At the end of the secretory stage, enamel ma-
at the enamel surface [85]. Fluoride-containing trix proteinases are activated, and at the transi-
apatite with fluoride concentrations of 100 ppm tion stage, enamel matrix proteins are rapidly
are found in the inner enamel (300 μm from the lost. At this stage, the cell junctions between the
surface) of human teeth with minimal (mild) flu- ameloblasts are open, allowing fluoride to read-
orosis [85]. The higher fluoride-containing apa- ily move from the serum to the enamel matrix.
tite (approximately 2,000 ppm F) is similar to that The presence of increased amounts of fluoride
found in the mid-layer of enamel (150 μm from in the transition stage may make this stage high-
the surface) of severely fluorosed human teeth. ly susceptible to the effects of fluoride on enamel
Therefore, these studies indicate that the reduced formation.
hydrolysis of amelogenin found in fluorosed mat- At the maturation stage, the pH in the enamel
uration stage enamel [1, 51] may be due to the re- matrix changes periodically between acidic
duction in the rate of hydrolysis of amelogenins (pH 5.8) and neutral (pH 7.2) as ameloblasts
bound to fluoride-containing enamel crystals. modulate [94, 95]. If we assume that the acidifi-
These effects of fluoride incorporation on hy- cation of the enamel matrix has a role in amelo-
drolysis of apatite-bound amelogenins is consis- blast modulation from ruffle-ended to smooth-
tent with the observation that fluoride-induced ended ameloblasts, in dental fluorosis, changes
subsurface hypomineralization can independent- in matrix pH secondary to fluoride-enhanced
ly occur in the maturation stage only [59, 62, 86]. mineral deposition could contribute to the delay
Mineralization defects in fluorosed rat incisor in the transition from ruffle-ended to smooth-
maturation stage enamel are characterized by the ended ameloblasts. This delay in ameloblast
development of a generalized hypomineralized modulation (which is a characteristic of fluo-
porous subsurface area along the entire crown rosed maturation ameloblasts) could possibly
enamel [4, 87–90]. This type of defect correlates contribute to the delay in removal of amelogen-
to the porous white opacities seen clinically. ins which occur in fluorosed enamel.
Particularly at this final stage of enam-
Potential Effects of Matrix pH on Fluoride-Related el mineralization, Bronckers et al. [93] have
Changes in Enamel Formation hypothesized that fluoride in the enamel ma-
Matrix protein removal may also be influenced by trix may enhance mineralization resulting in
fluoride-mediated changes in pH during apatite localized hypermineralization, requiring the
crystal formation. Formation of apatite results in ameloblasts to pump additional bicarbonate
the formation of a substantial number of protons into the extracellular enamel matrix. This hy-
[10Ca2+ + 6 HPO42- + 2H2O → Ca10(PO4)6(OH)2 permineralization would deplete the local res-
+ 8H+] that need to be neutralized. Amelogenins ervoir or free calcium ions, resulting in a sub-
bind as many as 12 protons per molecule [91]. sequent band of hypomineralized enamel. This
However, if this amelogenin buffering system is hypothesis is supported by a recent study show-
either not available, or is saturated, it is conceiv- ing an upregulation of mRNA for the pH reg-
able that a fluoride-induced pH drop could alter ulator NBCe1 in fluorosed maturation stage
the amelogenin tertiary structure and affect its ameloblasts as compared to control maturation
function [92]. ameloblasts [92].

92 DenBesten · Li
In summary, the mechanisms by which fluo- (6) In late maturation, when amelogenins
ride alters enamel maturation are multi-factorial. are finally removed (or in mild dental fluorosis
We propose a multi-stage model for the formation with minimal amelogenin retention), fluoride-
of fluorosed enamel, as follows: mediated hypermineralization may increase the
(1) Crystals forming in the secretory stage of local acidification affecting ameloblast function,
enamel have an increased fluoride content and such as ion transport activities. Although po-
therefore bind more amelogenin. rous subsurface enamel is the major phenotype
(2) Hydrolysis of amelogenins by proteinases of fluorosed enamel, successive layers of hypom-
is delayed by altered amelogenin interactions with ineralized and hypermineralized enamel are also
the fluoride-containing hydroxyapatite crystals. a characteristic of the fluorosed enamel matrix [4,
(3) At the transition stage, fluoride is rapidly 89].
deposited into the porous enamel matrix between It is likely that there are additional effects of
the open cell junctions, resulting in increased for- fluoride, including other indirect effects on cells
mation of fluoride-containing apatite, and a delay at different stages of formation, and that in the
in protein hydrolysis secondary to altered miner- course of our and others’ studies this model and
al/matrix interactions. our understanding of the mechanisms (includ-
(4) The net result of these fluoride-related ef- ing more potential direct cellular effects) will be
fects in the secretory and transition stages is reten- expanded.
tion of amelogenins in the maturation stage. This
delay in removal of amelogenins increases the rela-
tive pH in the maturation stage under ruffle-ended Acknowledgements
ameloblasts as amelogenins buffer the increased
protons resulting from mineral formation. The authors would thank Antonius Bronkers and
Donacian Lyaruu from ACTA University, Amsterdam for
(5) The reduced acidification of the matrix un-
their insightful discussions, as well as funding support
der ruffle-ended ameloblasts further delays mod- from NIH grant # R01DE013508.
ulation to smooth-ended ameloblasts, resulting in
fewer bands of modulating ameloblasts.

1 Aoba T, Moreno EC, Tanabe T, Fukae M: 5 Angmar-Månsson B, Whitford GM: 9 National Research Council: Fluoride in
Effects of fluoride on matrix proteins Enamel fluorosis related to plasma F Drinking Water: A Scientific Review of
and their properties in rat secretory levels in the rat. Caries Res 1984;18: EPA’s Standards. Washington, National
enamel. J Dent Res 1990;69:1248–1250. 25–32. Academies Press, 2006, p 530.
2 Richards A: Nature and mechanisms of 6 Angmar-Månsson B, Lindh U, Whitford 10 Moller IJ: Fluorides and dental fluorosis.
dental fluorosis in animals. J Dent Res GM: Enamel and dentin fluoride levels Int Dent J 1982;32:135–147.
1990;69(spec No):701–705, discussion and fluorosis following single fluoride 11 Kroncke A: Perikymata (in German).
721. doses: a nuclear microprobe study. Car- Dtsch Zahnarztl Z 1966;21:1397–1401.
3 Giambro NJ, Prostak K, DenBesten PK: ies Res 1990;24:258–262. 12 Smith GE: Fluoride, teeth and bone. Med
Characterization of fluorosed human 7 Everett ET, McHenry MA, Reynolds N, J Aust 1985;143:283–286.
enamel by color reflectance, ultrastruc- et al: Dental fluorosis: variability among 13 McKay FS: The study of mottled enamel
ture, and elemental composition. Caries different inbred mouse strains. J Dent (dental fluorosis). J Am Dent Assoc
Res 1995;29:251–257. Res 2002;81:794–798. 1952;44:133–137.
4 Angmar-Månsson B, Ericsson Y, Ekberg 8 Angmar-Månsson B, Whitford GM: 14 Mottled enamel. Am J Public Health
O: Plasma fluoride and enamel fluorosis. Environmental and physiological factors Nations Health 1933;23:47–48.
Calcif Tissue Res 1976;22:77–84. affecting dental fluorosis. J Dent Res 15 Ockerse T, Wasserstein B: Stain in mot-
1990:706–713. tled enamel. J Am Dent Assoc

Dental Fluorosis 93
16 Subcommittee on Health Effects of 29 Fejerskov O, Larsen MJ, Josephsen K, 41 Everett ET, Yan D, Weaver M, Liu L,
Ingested Fluoride (National Research Thylstrup A: Effect of long-term admin- Foroud T, Martinez-Mier EA: Detection
Council): Health Effects of Ingested Flu- istration of fluoride on plasma fluoride of dental fluorosis-associated quantita-
oride. Washington, National Academy of and calcium in relation to forming tive trait loci on mouse chromosomes 2
Sciences, 1993;169. enamel and dentin in rats. Scand J Dent and 11. Cells Tissues Organs 2009;189:
17 Dean HT: Fluorine in the control of Res 1979;87:98–104. 212–218.
dental caries. J Am Dent Assoc 1956;52: 30 Fejerskov O, Silverstone LM, Melsen B, 42 Smith CE, Nanci A, Denbesten PK:
1–8. Moller IJ: Histological features of fluo- Effects of chronic fluoride exposure on
18 Dean HT: Endemic fluorosis and its rela- rosed human dental enamel. Caries Res morphometric parameters defining the
tion to dental caries: 1938. Public Health 1975;9:190–210. stages of amelogenesis and ameloblast
Rep 2006;121(suppl 1):213–219, 31 Fejerskov O, Thylstrup A, Larsen MJ: modulation in rat incisors. Anat Rec
discussion 212. Clinical and structural features and pos- 1993;237:243–258.
19 Thylstrup A, Fejerskov O, Mosha HJ: A sible pathogenic mechanisms of dental 43 Bronckers AL, Jansen LL, Wöltgens JH:
polarized light and microradiographic fluorosis. Scand J Dent Res 1977;85: A histological study of the short-term
study of enamel in human primary teeth 510–534. effects of fluoride on enamel and den-
from a high fluoride area. Arch Oral Biol 32 Fejerskov O, Yanagisawa T, Tohda H, et tine formation in hamster tooth-germs
1978;23:373–380. al: Posteruptive changes in human den- in organ culture in vitro. Arch Oral Biol
20 USEPA: Fluorine (soluble fluoride) tal fluorosis – a histological and ultra- 1984;29:803–810.
(CASRN 7782-41-4). Washington, Inte- structural study. Proc Finn Dent Soc 44 Bronckers AL, Jansen LL, Wöltgens JH:
grated Risk Information System, 1987. 1991;87:607–619. Long-term (8 days) effects of exposure
21 Institute of Medicine: Dietary reference 33 Kidd EA, Thylstrup A, Fejerskov O: The to low concentrations of fluoride on
intakes: for calcium, phosphorus, mag- histopathology of enamel caries in fluo- enamel formation in hamster tooth-
nesium, vitamin D, and fluoride. Wash- rosed deciduous teeth. Caries Res germs in organ culture in vitro. Arch
ington, National Academies Press, 1997. 1981;15:346–352. Oral Biol 1984;29:811–819.
22 Levy SM, Guha-Chowdhury N: Total 34 Kierdorf U, Kierdorf H, Fejerskov O: 45 Kerley MA, Kollar EJ: Regeneration of
fluoride intake and implications for Fluoride-induced developmental tooth development in vitro following
dietary fluoride supplementation. J Pub- changes in enamel and dentine of Euro- sodium fluoride treatment. Am J Anat
lic Health Dent 1999;59:211–223. pean roe deer (Capreolus capreolus L.) 1977;149:181–195.
23 World Health Organization: Water for as a result of environmental pollution. 46 Levenson GE: The effect of fluoride on
Life: Making It Happen. Geneva, WHO, Arch Oral Biol 1993;38:1071–1081. ameloblasts of mouse molar tooth germs
2005. 35 Rojas-Sanchez F, Alaminos M, Campos ‘in vitro’. J Biol Buccale 1980;8:255–263.
24 Lennon MA: One in a million: the first A, Rivera H, Sanchez-Quevedo MC: 47 Kierdorf H, Kierdorf U: Disturbances of
community trial of water fluoridation. Dentin in severe fluorosis: a quantitative the secretory stage of amelogenesis in
Washington, National Institutes of histochemical study. J Dent Res fluorosed deer teeth: a scanning
Health, 2006. 2007;86:857–861. electron-microscopic study. Cell Tissue
25 Service PH: Review of Fluoride Benefits 36 Ekstrand J, Spak CJ, Ehrnebo M: Renal Res 1997;289:125–135.
and Risks, Report of the ad hoc Sub- clearance of fluoride in a steady state 48 Zhou R, Zaki AE, Eisenmann DR: Mor-
mommittee on Fluoride of the Commit- condition in man: influence of urinary phometry and autoradiography of
tee to Coordinate Environmental Health flow and pH changes by diet. Acta Phar- altered rat enamel protein processing
and Related Programs. Washington, macol Toxicol (Copenh) 1982;50: due to chronic exposure to fluoride.
Department of Health and Human Ser- 321–325. Arch Oral Biol 1996;41:739–747.
vices, 1991. 37 Jarnberg PO, Ekstrand J, Irestedt L, 49 Bronckers ALJJ, Wöltgens JHM: Short
26 Centers for Disease Control: Ten Great Santesson J: Renal fluoride excretion term effects of fluoride on biosynthesis
Public Health Achievements in the 20th during and after enflurane anaesthesia: of enamel-matrix proteins and dentin
Century. Washington, CDC, 2008. dependency on spontaneous urinary collagens and on mineralization during
27 Beltrán-Aguilar ED, Barker LK, Canto pH-variations. Acta Anaesthesiol Scand hamster tooth-germ development in
MT, et al: Surveillance for dental caries, 1980;24:129–134. organ culture. Archs Oral Biol 1985;39:
dental sealants, tooth retention, edentu- 38 Spak CJ, Berg U, Ekstrand J: Renal clear- 181–185.
lism, and enamel fluorosis – United ance of fluoride in children and adoles- 50 DenBesten PK: Effects of fluoride on
States, 1988–1994 and 1999–2002. cents. Pediatrics 1985;75:575–579. protein secretion and removal during
MMWR Surveill Summ 2005;54:1–43. 39 Pendrys DG, Stamm JW: Relationship of enamel development in the rat. J Dent
28 Fejerskov O, Johnson NW, Silverstone total fluoride intake to beneficial effects Res 1986;65:1272–1277.
LM: The ultrastructure of fluorosed and enamel fluorosis. J Dent Res 51 Bronckers AL, Lyaruu DM, Bervoets TJ,
human dental enamel. Scand J Dent Res 1990;69(spec No):529–538, discussion Wöltgens JH: Fluoride enhances intrac-
1974;82:357–372. 556–527. ellular degradation of amelogenins dur-
40 Taves DR: Dietary intake of fluoride ing secretory phase of amelogenesis of
ashed (total fluoride) v. unashed (inor- hamster teeth in organ culture. Connect
ganic fluoride) analysis of individual Tissue Res 2002;43:456–465.
foods. Br J Nutr 1983;49:295–301.

94 DenBesten · Li
52 Matsuo S, Inai T, Kurisu K, Kiyomiya K, 63 Nishikawa S, Josephsen K: Cyclic local- 76 Simmer JP, Fincham AG: Molecular
Kurebe M: Influence of fluoride on ization of actin and its relationship to mechanisms of dental enamel forma-
secretory pathway of the secretory junctional complexes in maturation tion. Crit Rev Oral Biol Med 1995;6:
ameloblast in rat incisor tooth germs ameloblasts of the rat incisor. Anat Rec 84–108.
exposed to sodium fluoride. Arch Toxi- 1987;219:21–31. 77 Smith CE, Nanci A: Protein dynamics of
col 1996;70:420–429. 64 Bartlett JD, Dwyer SE, Beniash E, Skobe amelogenesis. Anat Rec 1996;245:
53 Monsour P, Harbrow J, Warshawsky H: Z, Payne-Ferreira TL: Fluorosis: a new 186–207.
Effects of acute doses of sodium fluoride model and new insights. J Dent Res 78 Kerebel B, Daculsi G: Ultrastructural
on the morphology and the detectable 2005;84:832–836. and crystallographic study of human
calcium associated with secretory 65 Kubota K, Lee DH, Tsuchiya M, et al: enamel in endemic fluorosis (in French).
ameloblasts in rat incisors. J Histchem Fluoride induces endoplasmic reticulum J Biol Buccale 1976;4:143–154.
Cytochem 1989;37:463–471. stress in ameloblasts responsible for 79 Vieira A, Hancock R, Dumitriu M, et al:
54 Fejerskov O, Larsen MJ, Richards A, Bae- dental enamel formation. J Biol Chem How does fluoride affect dentin micro-
lum V: Dental tissue effects of fluoride. 2005;280:23194–23202. hardness and mineralization? J Dent Res
Adv Dent Res 1994;8:15–31. 66 Li Y, Decker S, Yuan ZA, et al: Effects of 2005;84:951–957.
55 Kardos TB, Hunter AR, Hubbard MJ: sodium fluoride on the actin cytoskel- 80 Sundstrom B, Jongebloed WL, Arends J:
Scanning electron microscopy of eton of murine ameloblasts. Arch Oral Fluorosed human enamel. A SEM inves-
trypsin-treated enamel from fluorosed Biol 2005;50:681–688. tigation of the anatomical surface and
rat molars. Adv Dent Res 1989;3: 67 Yan Q, Zhang Y, Li W, Denbesten PK: outer and inner regions of mildly fluo-
183–187. Micromolar fluoride alters ameloblast rosed enamel. Caries Res 1978;12:329–
56 Milhaud GE, Charles E, Loubiere ML, lineage cells in vitro. J Dent Res 2007;86: 338.
Kolf-Clauw M, Joubert C: Effects of fluo- 336–340. 81 Yanagisawa T, Takuma S, Fejerskov O:
ride on secretory and postsecretory 68 Zhang Y, Yan Q, Li W, DenBesten PK: Ultrastructure and composition of
phases of enamel formation in sheep Fluoride down-regulates the expression enamel in human dental fluorosis. Adv
molars. Am J Vet Res 1992;53: of matrix metalloproteinase-20 in Dent Res 1989;3:203–210.
1241–1247. human fetal tooth ameloblast-lineage 82 Yanagisawa T, Takuma S, Tohda H,
57 Nelson DG, Coote GE, Vickridge IC, cells in vitro. Eur J Oral Sci Fejerskov O, Fearnhead RW: High reso-
Suckling G: Proton microprobe determi- 2006;114(suppl 1):105–110, discussion lution electron microscopy of enamel
nation of fluorine profiles in the enamel 127–109, 380. crystals in cases of human dental fluoro-
and dentine of erupting incisors from 69 Zhang Y, Li W, Chi HS, Chen J, Den- sis. J Electron Microsc (Tokyo) 1989;38:
sheep given low and high daily doses of besten PK: JNK/c-Jun signaling pathway 441–448.
fluoride. Arch Oral Biol 1989;34: mediates the fluoride-induced down- 83 Robinson C, Yamamoto K, Connell SD,
419–429. regulation of MMP-20 in vitro. Matrix et al: The effects of fluoride on the nano-
58 Richards A, Kragstrup J, Josephsen K, Biol 2007;26:633–641. structure and surface pK of enamel crys-
Fejerskov O: Dental fluorosis developed 70 Robinson C, Brookes SJ, Shore RC, tals: an atomic force microscopy study of
in post-secretory enamel. J Dent Res Kirkham J: The developing enamel human and rat enamel. Eur J Oral Sci
1986;65:1406–1409. matrix: nature and function. Eur J Oral 2006;114(suppl 1):99–104, discussion
59 Suckling G, Thurley DC, Nelson DG: The Sci 1998;106(suppl 1):282–291. 127–109, 380.
macroscopic and scanning electron- 71 Fincham AG, Moradian-Oldak J: Recent 84 Tanimoto K, Le T, Zhu L, et al: Effects of
microscopic appearance and microhard- advances in amelogenin biochemistry. fluoride on the interactions between
ness of the enamel, and the related histo- Connect Tissue Res 1995;32:119–124. amelogenin and apatite crystals. J Dent
logical changes in the enamel organ of 72 Hu JC, Ryu OH, Chen JJ, et al: Localiza- Res 2008;87:39–44.
erupting sheep incisors resulting from a tion of EMSP1 expression during tooth 85 Richards A, Fejerskov O, Baelum V:
prolonged low daily dose of fluoride. formation and cloning of mouse cDNA. Enamel fluoride in relation to severity of
Arch Oral Biol 1988;33:361–373. J Dent Res 2000;79:70–76. human dental fluorosis. Adv Dent Res
60 Susheela AK, Bhatnagar M: Fluoride 73 Hu JC, Sun X, Zhang C, et al: Enamelysin 1989;3:147–153.
toxicity: a biochemical and scanning and kallikrein-4 mRNA expression in 86 Richards A, Kragstrup J, Nielsen-Kudsk
electron microscopic study of enamel developing mouse molars. Eur J Oral Sci F: Pharmacokinetics of chronic fluoride
surface of rabbit teeth. Arch Toxicol 2002;110:307–315. ingestion in growing pigs. J Dent Res
1993;67:573–579. 74 DenBesten PK, Yan Y, Featherstone JD, 1985;64:425–430.
61 DenBesten PK, Crenshaw MA: The et al: Effects of fluoride on rat dental 87 Angmar-Månsson B, Whitford GM:
effects of chronic high fluoride levels on enamel matrix proteinases. Arch Oral Plasma fluoride levels and enamel fluo-
forming enamel in the rat. Arch Oral Biol 2002;47:763–770. rosis in the rat. Caries Res 1982;16:
Biol 1984;29:675–679. 75 Moradian-Oldak J, Jimenez I, Maltby D, 334–339.
62 DenBesten PK, Crenshaw MA, Wilson Fincham AG: Controlled proteolysis of 88 Kierdorf H, Kierdorf U, Richards A,
MH: Changes in the fluoride-induced amelogenins reveals exposure of both Josephsen K: Fluoride-induced altera-
modulation of maturation stage amelo- carboxy- and amino-terminal regions. tions of enamel structure: an experimen-
blasts of rats. J Dent Res 1985;64:1365– Biopolymers 2001;58:606–616. tal study in the miniature pig. Anat
1370. Embryol (Berl) 2004;207:463–474.

Dental Fluorosis 95
89 Richards A, Likimani S, Baelum V, Fejer- 91 Ryu OH, Hu CC, Simmer JP: Biochemi- 94 Sasaki S, Takagi T, Suzuki M: Cyclical
skov O: Fluoride concentrations in cal characterization of recombinant changes in pH in bovine developing
unerupted fluorotic human enamel. Car- mouse amelogenins: protein quantita- enamel as sequential bands. Arch Oral
ies Res 1992;26:328–332. tion, proton absorption, and relative Biol 1991;36:227–231.
90 Shinoda H: Effect of long-term adminis- affinity for enamel crystals. Connect 95 Smith CE: Cellular and chemical events
tration of fluoride on physico-chemical Tissue Res 1998;38:207–214, discussion during enamel maturation. Crit Rev Oral
properties of the rat incisor enamel. 241–206. Biol Med 1998;9:128–161.
Calcif Tissue Res 1975;18:91–100. 92 Zheng L, Zhang Y, He P, Kim J, Sch- 96 Newbrun E, Brudevold F: Studies on the
neider R, Bronckers AL, Lyaruu DM, physical properties of fluorosed enamel-
DenBesten PK: NBCe1 in Mouse and I: Microradiography. Arch Oral Biol
Human Ameloblasts may be Indirectly 1960;2:15–20.
Regulated by Fluoride. J Dent Res
93 Bronckers AL, Lyaruu DM, DenBesten
PK: The impact of fluoride on amelo-
blasts and the mechanisms of enamel
fluorosis. J Dent Res 2009;88:877–893.

Pamela DenBesten
Department of Orofacial Sciences, School of Dentistry, University of California, San Francisco
513 Parnassus Avenue
San Francisco, CA 94143 (USA)
Tel. +1 415 502 7828, E-Mail

96 DenBesten · Li
Impact of Fluoride in the Prevention of Caries and Erosion
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 97–114

Mechanisms of Action of Fluoride for

Caries Control
Marília Afonso Rabelo Buzalafa ⭈ Juliano Pelim Pessanc ⭈
Heitor Marques Honóriob ⭈ Jacob Martien ten Cated
aDepartment of Biological Sciences and bDepartment of Pediatric Dentistry, Orthodontics and Public Health, Bauru
Dental School, University of São Paulo, Bauru, and cDepartment of Pediatric Dentistry and Public Health, Araçatuba Dental
School, São Paulo State University, Araçatuba, Brazil; dDepartment of Cariology, Endodontology and Pedodontology,
Academic Center for Dentistry Amsterdam, Amsterdam, The Netherlands

Abstract plaque have biological activity on critical virulence factors

Fluoride was introduced into dentistry over 70 years ago, of S. mutans in vitro, such as acid production and glucan
and it is now recognized as the main factor responsible for synthesis, but the in vivo implications of this are still not
the dramatic decline in caries prevalence that has been clear. Evidence also supports fluoride’s systemic mecha-
observed worldwide. However, excessive fluoride intake nism of caries inhibition in pit and fissure surfaces of per-
during the period of tooth development can cause dental manent first molars when it is incorporated into these
fluorosis. In order that the maximum benefits of fluoride teeth pre-eruptively.
for caries control can be achieved with the minimum risk Copyright © 2011 S. Karger AG, Basel
of side effects, it is necessary to have a profound under-
standing of the mechanisms by which fluoride promotes The multifactorial disease dental caries is caused
caries control. In the 1980s, it was established that fluo- by the simultaneous interplay of different factors –
ride controls caries mainly through its topical effect. Fluo- dietary sugars, dental biofilm and the host – within
ride present in low, sustained concentrations (sub-ppm the context of the oral environment. The complex
range) in the oral fluids during an acidic challenge is able and long-lasting interactions of these factors and
to absorb to the surface of the apatite crystals, inhibiting how they lead to caries was already described half
demineralization. When the pH is re-established, traces a century ago [1]. With our current understanding,
of fluoride in solution will make it highly supersaturated the most obvious way to fight caries is to control the
with respect to fluorhydroxyapatite, which will speed causal agents by removing the dental biofilm and
up the process of remineralization. The mineral formed reducing sugar consumption. These approaches
under the nucleating action of the partially dissolved min- form the basis of comprehensive protocols to con-
erals will then preferentially include fluoride and exclude trol the disease, but have been proven insufficient
carbonate, rendering the enamel more resistant to future to lead to a desired level of prevention because they
acidic challenges. Topical fluoride can also provide antimi- strongly rely on patient compliance. Even before a
crobial action. Fluoride concentrations as found in dental complete understanding of the etiology of dental
Fig. 1. General composition of dental enamel and dentin.

caries was reached, fluoride had emerged as a piv- the tooth surface, with demineralization being
otal adjunct to combat the disease [2]. Fluoride is caused by the production of acids by oral bacte-
currently recognized as the main factor responsible ria after sugar consumption [9]. To understand
for the significant decline in caries prevalence that how the acids can attack the dental tissues, it
has been observed worldwide [3]. On the other is fundamental to know their biochemical
hand, excessive fluoride intake during the period properties.
of tooth development may cause dental fluorosis,
the only proven side effect of the use of fluoride of Composition of Enamel and Dentin
dental relevance [4]. An increase in the prevalence Despite the presence of common constituents,
of dental fluorosis has been reported concomitant- enamel and dentin have different structures that
ly with the decrease in caries [5–7]. Although most will affect caries progression within these tissues
of this fluorosis is mild or very mild, and has little as well as the reactivity of fluoride with them.
or no impact on quality of life of affected people Permanent enamel is an acellular tissue com-
[8], a judicious use of fluoride to avoid moderate posed chiefly of minerals (calcium-deficient
and severe fluorosis is needed. Thus, in order that carbonated hydroxyapatite, 85% in volume).
the maximum benefit of fluoride for caries control Hydroxyapatite molecules are arranged in long
can be achieved with a minimum risk of side ef- and thin apatite crystals, which in turn are or-
fects, it is necessary to have a comprehensive un- ganized into the resulting enamel prisms (fig. 1).
derstanding of the mechanisms by which fluoride Despite the high mineral content, the space be-
promotes caries control. tween the crystals is occupied by water (12% by
volume) and organic material (3% by volume)
[10, 11]. It is in this space filled with the enamel
Biochemistry of Caries Development fluid that the de- and remineralization reactions
take place. In brief, upon a cariogenic challenge,
Dental caries is the net result of consecutive hydroxyapatite crystals are dissolved from the
cycles of de- and remineralization of dental tis- subsurface, while fluorapatite crystals are depos-
sues at the interface between the biofilm and ited at the surface, thus resulting in a subsurface

98 Buzalaf · Pessan · Honório · ten Cate

lesion. The dissolution process of enamel is there- normally available in saliva are compared with
fore a chemical event. the concentrations that are necessary to reach
On the other hand, permanent dentin con- saturation and form this mineral. The solu-
tains (by volume) 47% apatite, 33% organic bility product of enamel (KSPenamel) which is
components and 20% water (fig. 1). The mineral related to the concentrations of Ca+2, PO4–3 and
phase is also hydroxyapatite, similar to enamel, OH– required for the formation of enamel crys-
but the crystallites have much smaller dimen- tals, has been calculated at 5.5 × 10–55 mol9/l9 at
sions than those found in enamel. As a conse- 37°C, slightly higher than that required to form
quence, the ratio surface area/crystallite volume hydroxyapatite (KSPHA 7.41 × 10–60 mol9/l9).
is larger, which makes the mineral phase more Under physiological conditions (pH 7.0), based
reactive. As a result, dentin surfaces are more on the salivary concentrations of free Ca+2,
susceptible to caries attack than enamel surfaces. PO–43 and OH– that are available to form enamel
The organic matrix is mainly composed of colla- crystals, the ion activity product of hydroxyap-
gen (90%), but there are many non-collagenous atite (IAPHA) has been calculated at 6.1 × 10–48
components that determine the properties of the mol9/l9 [11]. Therefore, if the IAPHA in saliva un-
matrix and interfere with de- and remineraliza- der physiological conditions is higher than the
tion reactions. Collagen forms the backbone of concentrations required to form enamel crystals
dentin and serves as a template for the deposi- (KSPenamel) this implies that enamel mineral does
tion of apatite crystallites within the collagen not dissolve in saliva (fig. 2a). Contrarily, enamel
helix. This kind of structure promotes a syner- crystals would be expected to grow or new crys-
gism between matrix and apatite: the mineral tals would be expected to form at the biofilm-free
phase cannot be completely dissolved during tooth surfaces. This does not happen because sa-
an acid attack and the matrix does not undergo liva contains proteins that inhibit hydroxyapa-
enzymatic degradation while its surface is still tite crystal growth, including statherin and many
protected by apatite [11]. Dentin caries is thus proline-rich proteins [20].
a biochemical process characterized initially by When a biofilm is covering the enamel sur-
the dissolution of the mineral, which in turn ex- face, it reduces the access of saliva to the tooth.
poses the organic matrix to breakdown [12–15] The relevant fluid phase in this case is the bio-
by bacterial-derived enzymes as well as by host- film fluid which, under resting conditions,
derived enzymes such as matrix metalloprotei- is also supersaturated with respect to enamel
nases present in dentin and saliva [16, 17]. It is (IAPHA 1.4 × 10–47). This would favor reminer-
also important to highlight that dentin is a cel- alization of previously demineralized enamel or
lular tissue and that upon exogenous challenges promote the formation of supragingival calculus
the pulpo-dentinal organ responds with mineral (fig. 2b).
deposition [18]. This process, combined with the The characteristics of the plaque fluid mi-
flow of dentinal fluid from the pulp, reduces the croenvironment change considerably upon a
rate of lesion progression in dentin in vivo [19]. sugar challenge. In this case, bacteria produce
lactic acid that makes the plaque fluid pH fall
Dental Mineral Dynamics (typically between 4.5 and 5.5). The driving force
The reason why caries progresses slowly is due to is then shifted to mineral dissolution. But why
the high supersaturation of saliva with respect to does this happen if saliva is continuously secreted
enamel mineral under physiological conditions. with relatively stable Ca+2 and PO–43 concentra-
This can be easily understood when the concentra- tions, which would apparently maintain IAPHAP
tions of free ions required to form hydroxyapatite unaltered?

Mechanisms of Action of Fluoride 99

Fig. 2. Dynamics of minerals in saliva and enamel under neutral (a, b) and acidic conditions (c, d).

The pH fall has a profound effect on the solu- and OH– are reduced, thus decreasing the IAPHAP
bility of hydroxyapatite and other calcium phos- and turning the solution undersaturated with re-
phates. In general, the solubility of apatite in- spect to enamel (IAPHA< KSPenamel), promoting
creases 10 times with a decrease of 1 pH unit. enamel dissolution (fig. 2c–d) [11]. The dissolu-
This happens because H+ combines with PO4–3 tion can be avoided by increasing the concentra-
and OH– to form H2PO4–3 and H2O (Eq. 1). As tions of Ca+2 and/or PO4–3 in the fluid. Therefore,
a consequence, the concentrations of free PO4–3 the lower the pH, the higher the concentrations

100 Buzalaf · Pessan · Honório · ten Cate


Ca (mmol/l)

Fig. 3. Solubility of apatite as a 2

function of pH, expressed in terms
of calcium concentrations. Blue 1
line indicates salivary calcium
concentrations. The critical pH for
dissolution of hydroxyapatite (HA)
and fluorhydroxyapatite (FA) is 5.5
and 4.5, respectively.

of Ca+2 and PO4–3 required to reach saturation level, dissolution takes place (demineralization)
in respect to hydroxyapatite. This relationship is (fig. 3).
shown in figure 3.
Carious Lesion Formation
The existence of mineral phases with different sol-
ubilities in the dental tissues explains the patterns
(1) of demineralization found in caries. Under nor-
mal conditions (pH around 7.0), the oral fluids
are supersaturated with respect to both hydroxy-
apatite and fluorhydroxyapatite. Thus, there is a
When the pH is gradually lowered from 7.0 to tendency towards formation of these two miner-
5.0, the value of pH for which the fluid becomes als (formation of calculus and remineralization of
saturated with respect to the mineral in question demineralized areas).
(IAP = KSP) is the so-called ‘critical pH’. At those When bacteria metabolize sugars producing
conditions, equilibrium exists (no mineral dis- lactic acid, pH decreases in saliva and biofilm flu-
solution and no mineral precipitation). For hy- id (4.5<pH<5.5) rendering these fluids undersatu-
droxyapatite, the critical pH is around 5.5, while rated with respect to hydroxyapatite while still su-
it is approximately 4.5 for fluorhydroxyapatite. persaturated with respect to fluorhydroxyapatite.
When the pH is above the critical level for the Consequently, hydroxyapatite dissolves from the
formation of a respective mineral phase, precipi- subsurface and fluorhydroxyapatite forms in the
tation of this phase occurs (remineralization). surface layers. Saliva, in turn, has a strong buff-
Contrarily, when the pH is below the critical ering capacity, and this property together with

Mechanisms of Action of Fluoride 101


Mineral loss and gain

Mineral loss and gain


Fig. 4. Cyclic nature of de- and remineralization reactions. Source: Buzalaf et al. [68].

outward diffusion of acids makes the biofilm pH Mechanisms by Which Fluoride Controls
rise within a few minutes. When the pH becomes Caries
greater than 5.5, the condition of supersaturation
of the oral fluids with respect to hydroxyapatite is Supplementation of public water supplies with con-
restored; the partially demineralized crystals then trolled levels of fluoride was the first approach in-
undergo remineralization. The net result of suc- volving the use of fluoride for caries control. The
cessive de- and remineralization cycles with the encouraging results coming from this measure lat-
preponderance of the former over the latter leads er prompted the recommendation for the use of
to caries (fig. 4). fluoride supplements by pregnant women in order
The supersaturation of the oral fluids with to prevent caries in their offspring. Since the first
respect to fluorhydroxyapatite during cariogen- cariostatic benefits of fluoride were observed when
ic challenges is responsible for the maintenance this element was ingested from ‘systemic’ sources,
of the surface layer of carious lesions (fig. 2d). from the 1940s to the 1970s it was originally be-
With time, formation of fluorhydroxyapatite at lieved that the cariostatic mechanism of fluoride
the expense of hydroxyapatite further increases relied mainly on its uptake in the forming enamel.
the concentration of fluorhydroxyapatite in the This would lead to the formation of fluorhydroxy-
surface layer. This layer has a protective role, apatite, a mineral phase more resistant to future dis-
slowing the diffusion of demineralizing agents solution. For this purpose, ingestion of fluoride was
into the lesion. On the other hand, it also ren- considered unavoidable and the occurrence of den-
ders remineralization of the lesion body more tal fluorosis was regarded as a necessary risk in or-
difficult [11]. der to achieve the cariostatic benefits of fluoride.

102 Buzalaf · Pessan · Honório · ten Cate

2,000 ppm

3,000 ppm

Ca, mmol/l (saturation)

3 2

0 6 8 10

22,500 ppm 38,000 ppm


Fig. 5. Calculated solubility of

fluorhydroxyapatite at 37°C in 0.1 0 10 20 30 40 50 60 70 80 90 100
mol/l acetate buffer at initial pH
HAP % of substitution by F (HAP for FAP) FAP
5.0 as a function of the degree of
replacement of OH– by F–.

However, something seemed to be missing. It [22, 23]. Elegant in situ studies conducted in
was observed that fluoride concentrations typi- Scandinavia greatly contributed to the consoli-
cally found in enamel were unable to confer sig- dation of this concept. In one of the studies, the
nificant protection against caries. The highest authors placed human and shark enamel slabs
fluoride concentrations in enamel are found in in removable appliances and covered them with
the surface. They are usually around 2,000 ppm orthodontic bands to allow plaque accumulation.
(6% replacement of OH– by F– in hydroxyapa- Shark enamel was used because it is composed
tite) in non-fluoridated areas and 3,000 ppm (8% almost of pure fluorapatite (around 30,000 ppm
replacement of OH– by F– in hydroxyapatite) in fluoride). Microradiographic analyses revealed
fluoridated areas. However, these concentrations that carious lesions formed in both substrates,
dramatically fall after the outer first 10–20 μm although they were less severe in shark enamel.
of enamel to around 50 ppm in non-fluoridated The authors compared these data with data from
areas and hundreds of ppm in fluoridated areas previous studies with human enamel when dai-
[21]. These levels are far below those able to con- ly mouthrinsing with 0.2% NaF was used. They
fer expressive reduction on the solubility of hy- observed that the mineral loss in human enamel
droxyapatite (fig. 5). treated with fluoride rinse was lower than that of
In the 1980s, the concept that fluoride controls shark enamel without any additional treatment.
caries lesion development primarily through its The lesion depths of these substrates were similar
topical effect on de- and remineralization pro- (fig. 6) [24]. These studies proved that structur-
cesses taking place at the interface between the ally bound fluoride (shark enamel) was not very
tooth surface and the oral fluids was established effective in inhibiting demineralization, while

Mechanisms of Action of Fluoride 103

Lesion depth (µm)

⌬Z (vol% µm)
90 800 1680
400 965
20 36 39 607
a b

Human Shark Human (daily rinse 0.2% NaF)

Fig. 6. Lesion depth (a) and mineral loss (ΔZ; b) in human and shark enamel after 4 weeks in situ as evaluated by mi-
croradiography. Groups human and shark refer to human and shark enamel slabs, respectively, which did not receive
any additional treatment. Group human (daily rinse 0.2% NaF) refers to human enamel slabs that received daily rinses
of 0.2% NaF. Bars indicate SD (n = 6). Original data from Øgaard et al. [24].

fluoride in solution (NaF rinse) led to a high de- These pools can be didactically divided into 5 cat-
gree of protection. This provided evidence that egories [25] (fig. 7):
the primary action of fluoride is topical due to its 1 FO: outer fluoride, present outside enamel (in
presence in the fluid phases of the oral environ- the biofilm or saliva);
ment. It is important to stress out that the con- 2 FS: fluoride present in the solid phase,
centrations of fluoride found in shark enamel are incorporated in the structure of the crystals,
many times higher than those typically present also known as fluorhydroxyapatite;
in human enamel, but even so they were unable 3 FL: fluoride present at the enamel fluid;
to completely inhibit enamel dissolution. On the 4 FA: fluoride adsorbed to the crystal surface,
other hand, fluoride concentrations as little as 1 also known as loosely-bound;
ppm present in an acid solution can reduce the 5 CaF2: ‘CaF2-like’ material; globules deposited
solubility of carbonated hydroxyapatite to that on enamel and biofilm after application of
equivalent to hydroxyapatite. Higher concentra- highly concentrated fluoride products; acts as a
tions of fluoride in solution decrease the solubility pH-controlled fluoride and calcium reservoir.
following a logarithmic pattern [23].
Thus, to interfere in the dynamics of dental car- Fluoride Mechanisms of Action
ies formation, fluoride must be constantly present Inhibition of Demineralization
in the oral environment at low concentrations. In If fluoride is present in plaque fluid (FL) when bac-
order that the mechanisms involved in this pro- teria produce acids, it will penetrate along with
cess can be more easily understood it is helpful the acids at the subsurface, adsorb to the crystal
initially to consider the different ‘pools’ of fluo- surface (FA) and protect crystals from dissolution
ride that can be found in the oral environment. [26]. When the entire crystal surface is covered

104 Buzalaf · Pessan · Honório · ten Cate

Fig. 7. Schematic representation of the different ‘pools’ of fluoride in the oral environment. Modified from Arends and
Christoffersen [25].

by FA (100% coverage), it will not dissolve upon already able to substantially inhibit acid dissolu-
a pH fall caused by bacterial-derived acids, since tion of tooth minerals [23, 27].
this type of coating makes the characteristics of Calcium fluoride (CaF2) is an important source
the crystal similar to those of fluorapatite. On the of fluoride to the oral fluids (FL). It is known as
other hand, when the coating of FA is partial, the pH-controlled fluoride and calcium reservoir.
uncoated parts of the crystal will undergo dissolu- This compound forms when the fluoride concen-
tion (fig. 8) [25]. trations in the solution bathing enamel are higher
While FA is the ‘pool’ of fluoride that effective- than 100 ppm. The formation of CaF2 is a two-
ly protects the crystals from dissolution, the role stage reaction. Initially, a slight dissolution of the
of fluoride present in solution (FL) is equally im- enamel surface must occur to release Ca+2 that in
portant, since the higher the concentration of FL, a second stage will react with fluoride that is ap-
the higher the probability that it adsorbs (FA) and plied, thereby forming CaF2 globules. These glob-
protects the crystals. However, very low fluoride ules precipitate not only on sound enamel sur-
concentrations (sub-ppm range) in solution are faces but also and more importantly on biofilm,

Mechanisms of Action of Fluoride 105

Fig. 8. Events taking place at the subsurface of enamel upon a cariogenic acidic challenge.
Fluoride (FL) penetrates at the subsurface along with the acids, adsorbs to the surface of the crys-
tal and protects it from dissolution (left chart). When coverage is partial, uncovered portions of the
crystal will dissolve (right chart). Modified from Arends and Christoffersen [25].

106 Buzalaf · Pessan · Honório · ten Cate

Fig. 9. Schematic representation of remineralization occurring in the presence of fluoride. Fluoride speeds up the
process of remineralization and leads to the precipitation of a coat poor in carbonate and rich in fluoride on the par-
tially demineralized original crystallite. This renders the tooth structure more resistant to subsequent acidic challenges.
Modified from Featherstone [26].

pellicle and enamel porosities. The dissolution coating will be less soluble due to the exclusion
rate of CaF2 globules is limited by the adsorption of carbonate and incorporation of fluoride, ren-
of HPO4–2 that is lost under acidic pH, thus allow- dering the enamel more resistant to future acidic
ing CaF2 to dissolve and fluoride and calcium to challenges (fig. 9). After repeated cycles of disso-
be released. This fluoride will add to the ‘pool’ of lution and reprecipitation, enamel crystals may be
FL [11, 28]. completely different from their original state [11,
Enhancement of Remineralization
After an acidic challenge, salivary flow buffers the Role of ‘Systemic’ Fluoride
acids produced by the bacteria. When the pH is As mentioned above, the main mechanisms of
higher than 5.5, remineralization will naturally action of fluoride rely on its topical use since low,
occur (fig. 3) since saliva is supersaturated with sustained levels of fluoride in the oral fluids can
respect to the dental mineral. Traces of fluoride significantly control caries progression and re-
in solution during dissolution of hydroxyapatite versal. However, this concept does not invalidate
will make the solution highly supersaturated with the use of ‘systemic’ methods such as fluoridated
respect to fluorhydroxyapatite. This will speed up water. More than 60 years of intensive research
the process of remineralization. Fluoride will ad- attest to the safety and effectiveness of this mea-
sorb to the surface of the partially demineralized sure to control caries [4]. In this case, however,
crystals and attract calcium ions. Since carbonate- it should be emphasized that despite being clas-
free or low-carbonate apatite is less soluble, these sified as a ‘systemic’ method of fluoride delivery
phases will tend to form preferentially instead of (as it involves ingestion of fluoride), the mech-
the original mineral, under the nucleating ac- anism of action of fluoridated water to control
tion of the partially dissolved minerals. This new caries is mainly through its topical contact with

Mechanisms of Action of Fluoride 107

the teeth while in the oral cavity or when redis- since then many reports have been published on
tributed to the oral environment by means of sa- direct and indirect effects of fluoride on the ener-
liva. Since fluoridated water is consumed many gy and biosynthesis of streptococci [34]. Bacterial
times a day, the high frequency of contact of fluo- metabolism can be affected by fluoride through
ride present in the water with the tooth structure several complex mechanisms that are beyond the
or intraoral fluoride reservoirs helps to explain scope of the present chapter and therefore will be
why water fluoridation is so effective in control- presented only briefly.
ling caries, despite having fluoride concentra- Fluoride exerts its effects on oral bacteria by a
tions much lower than fluoride toothpastes, for direct inhibition of cellular enzymes (directly or
example [29]. This general concept can be ap- in combination with metals) or enhancing proton
plied to all methods of fluoride use traditionally permeability of cell membranes in the form of hy-
classified as ‘systemic’. In the light of the current drogen fluoride (HF) [33, 35]. The biological ef-
knowledge regarding the mechanisms by which fects and mechanisms of action of fluoride on oral
fluoride control caries, this system of classifica- bacteria are summarized in table 1.
According to the reaction H+ + F– HF, HF

tion is in fact misleading.
One point that deserves attention regarding is formed more easily under acidic conditions
the mechanism by which fluoridated water leads (pKa = 3.15) and enters the cell due to a higher
to caries control is that even recent studies have permeability of HF to bacterial cell membranes.
shown a beneficial pre-eruptive effect of water HF then dissociates in H+ and F– in the cytoplasm,
fluoride on caries control. Well-designed cohort which is more alkaline than the exterior
studies have reported that pre-eruption exposure environment [34]. This intracellular F– inhibits
to fluoride is important for caries prevention, es- glycolytic enzymes, resulting in a decrease in
pecially in pit and fissure surfaces of permanent acid production from glycolysis. F– in the cyto-
first molars. This could be due to the difficult ac- plasm also lowers cytoplasmatic pH (which de-
cess of topical fluoride to these areas. The anti- creases the entire glycolytic activity), affecting
caries protection may occur due to pre-eruption both the acid production and acid-tolerance of
fluoride uptake in the crystalline structure (FS) of S. mutans [33]. Cell membrane-associated H+-
the developing enamel, its adsorption on the crys- ATPases are also inhibited by F– because excret-
tal surface (FA) or its presence in the enamel flu- ed protons are brought back into the cell, there-
id (FL). Upon post-eruption acidic challenge, FS fore decreasing excretion of H+ from the cell
would be released to the fluid phase (FL), thus in- (fig. 10) [35, 36].
hibiting demineralization and enhancing remin- It is known that fluoride concentrations in
eralization [30, 31]. plaque can be increased for several hours after ex-
posure to a fluoridated dentifrice [37–40]. Lynch
Effects in Oral Bacteria et al. [41] concluded that low levels of plaque and
Although the main action of fluoride on the dy- salivary fluoride resulting from the use of 1,500
namics of dental caries is on de- and remineraliza- ppm fluoride toothpastes are insufficient to have
tion processes that occur on dental hard tissues, it a significant antimicrobial effect on plaque bac-
has also been proposed that the fluoride ion can teria. A recent review, however, concluded that
affect the physiology of microbial cells, includ- fluoride concentrations as found in dental plaque
ing cariogenic streptococci, which can thus in- have biological activity on critical virulence fac-
directly affect demineralization [32, 33]. The in- tors of S. mutans in vitro, such as acid production
hibitory effect of fluoride in pure cultures of oral and glucan synthesis, but the in vivo implications
streptococci was described over 70 years ago, and are still not clear [33].

108 Buzalaf · Pessan · Honório · ten Cate

Fig. 10. Fluoride accumulation, distribution and efflux from bacterial cells. BF=Bound fluoride.
Modified from Hamilton and Bowden [69].

Table 1. Biological effects and mechanisms of action of fluoride on oral bacteria

Biological activity Examples Mechanism

Enzyme inhibition (at sub- enolase, urease, P-ATPase, phosphatases, direct binding of F− or HF
millimolar levels of fluoride) heme catalase, heme peroxidase

F-ATPase, nitrogenase , RecA, CheY binding of metal-F complex

Dissipation of proton gradient/ acidification of cytoplasm action as transmembrane

motive force (at micromolar (inhibition of glycolysis, PTS system, and IPS proton carrier
levels of fluoride) formation)

inhibition of macromolecular synthesis and


PTS system = Phosphotransferase sugar transport system; IPS formation = intracellular polysaccharide formation.
Source: Koo [33].

Mechanisms of Action of Fluoride 109

As most of the evidence of antimicrobial effects important role on the mechanism of action of the
of fluoride on oral bacteria comes from in vitro ion. It is known that plaque and salivary fluoride
studies, caution must be taken when interpreting levels decrease rapidly after the application of a
these results. Clinical studies addressing the sub- fluoride vehicle, following a bi-phasic exponen-
ject, however, seem to indicate that fluoride does tial pattern [53]. These levels, however, are sig-
have an antimicrobial effect, and that this effect is nificantly elevated for many hours after the expo-
dependent on factors such as the fluoride concen- sure to the fluoridated agent when compared to
tration applied and associated antibacterial compo- baseline levels, indicating that fluoride is bound
nents. With regard to fluoride concentration, stud- to intraoral reservoirs and subsequently released
ies with different research protocols have shown to saliva over time [29, 37–40].
significantly lower plaque scores in subjects using a Fluoride can be deposited on dental hard tissues
5,000 ppm fluoride toothpaste, in comparison with as CaF2 (as discussed above), bound to the oral mu-
formulations containing 500, 1,100 and 1,500 ppm cosa and retained by dental plaque components.
fluoride [42, 43]. Concerning other components Oral mucosa has been shown to be an important
with inhibitory effects on plaque growth, it was fluoride reservoir, mainly due to its large surface
also demonstrated that the combination of high area, releasing fluoride to saliva over time [54].
levels of fluoride (5,000 ppm) and sodium lauryl Although all fluoride reservoirs contribute to the
sulphate reduces de novo plaque formation in sub- maintenance of the ion in the oral cavity, fluoride
jects using slurries of dentifrices with different flu- retained in dental plaque is likely more relevant
oride concentrations [43]. Also, the association of from a clinical perspective [for details, see Vogel,
fluoride with other ions in formulations contain- this vol., pp. 146–157], as it is the site where de- and
ing stannous fluoride or amine fluoride has been remineralization processes take place. Considering
shown to be effective in promoting lower plaque that most subjects do not completely remove dental
formation and acid production, either alone or in plaque after toothbrushing, the amount of fluoride
combination [44–46]. The use of a stabilized stan- retained in plaque can help determine the fate of
nous fluoride/sodium hexametaphosphate denti- the enamel underneath it [37–39].
frice [47, 48] as well as a stannous-containing so- Fluoride has a strong affinity to both organ-
dium fluoride dentifrice [49] have also proven to ic and inorganic components of plaque, and can
be effective in reducing plaque formation. be found as ionic, ionizable and strongly bound
Fluoride-releasing materials have also been forms. Although the amount of fluoride in the
shown to provide antimicrobial effects. Results ionizable fraction is considerably larger than in
from in vitro and in situ studies indicate that flu- the ionic pool, it adds to the amount of ionic fluo-
oride released from glass ionomer cements has an ride in plaque fluid, which is responsible for the
inhibitory effect on the pH fall and the acid pro- cariostatic action of fluoride [29]. The clinical rel-
duction rate of S. mutans and S. sanguinis [36]. evance of fluoride retained in plaque is that it can
Reduced S. mutans growth and lower pH fall on be released under acidic conditions during car-
plaque formed on glass ionomer cements has also iogenic challenges. In other words, fluoride is re-
been shown to occur when compared with com- leased when it is most needed to reduce demin-
posite resin [50–52]. eralization, to enhance remineralization of early
lesions, or both. Clinical studies support the con-
Fluoride in Intraoral Reservoirs cept that the amount of fluoride in oral reservoirs
Besides interfering in de- and remineralization is of paramount importance in its cariostatic ef-
processes, along with effects in oral bacteria, fectiveness, as caries incidence and activity have
fluoride retained in intraoral reservoirs plays an been shown to be inversely related to fluoride

110 Buzalaf · Pessan · Honório · ten Cate

Percentage inhibition of demineralization
100 Enamel
80 Dentin

0.01 0.1 1 10
Fluoride concentration (ppm)

Fig. 11. Inhibition of demineralization of enamel and dentin at different concentrations of fluo-
ride in solution. Data are expressed as percentage of demineralization at 0 ppm fluoride. Modified
from ten Cate et al. [11].

concentrations in saliva and/or dental plaque ppm fluoride dentifrices to arrest root carious le-
[55–57]. sions [42, 64].
(4) Dentin seems to benefit from a higher daily
frequency of exposure to fluoride [65] and also
Dentin De- and Remineralization and the from the combination of methods of fluoride use
Protective Effect of Fluoride [66] which is not necessarily the case for enamel.
(5) Dentin contact area with cariogenic acids
The essence of de- and remineralization process- is larger than that of enamel. For this reason, den-
es, as well as the interactions with fluoride that tin is apparently much more permeable to acids,
were described above for enamel, also apply to with demineralization taking place at a relatively
dentin. The main differences of both substrates large depth, while mineral deposition is restrict-
are: ed to the outer layers. If the crystallites surround-
(1) Dentin is more susceptible to caries attack ing the diffusion channels (tubules) are coated
than enamel, with a critical pH more than 1 pH with a fluoride-rich mineral, the acids will bypass
unit higher than that for enamel [58]. these relatively resistant minerals, while mineral
(2) Dentin demineralizes faster and reminer- and fluoride ions will readily be deposited. Thus,
alizes slower than enamel under the same experi- the lesion front in dentin moves deeper, while the
mental conditions [59, 60]. surface layer becomes broader. In enamel, on the
(3) More concentrated fluoride is needed to in- other hand, diffusion is much slower and allows
hibit demineralization [61, 62] (fig. 11) and to en- acids to ‘sidestep’ into smaller intraprismatic po-
hance remineralization [63] of dentin when com- rosities and dissolve crystallites that are still un-
pared with enamel. In fact, clinical trials show a affected by either acid or fluoride. Thus, mineral
beneficial effect of 5,000 ppm fluoride over 1,100 uptake and loss occur at similar depths for enamel

Mechanisms of Action of Fluoride 111

lesions, while for dentin lesions mineral uptake is enamel more resistant to future acidic challeng-
predominant at the surface and mineral loss at the es. Topical fluoride can also present antimicro-
lesion front [60]. bial action. Fluoride concentrations as found in
It has also been recently shown that very deep dental plaque have biological activity on critical
lesions extending through enamel into dentin can virulence factors of S. mutans in vitro, such as
be remineralized. Although this process is slow, it acid production and glucan synthesis, but the in
indicates that remineralization might be used to vivo implications of this are still not clear.
treat deep lesions [67]. Evidence from cohort studies also supports
fluoride’s systemic mechanism of caries inhibi-
tion in pit and fissure surfaces of permanent first
Conclusion molars when it is incorporated into these teeth
pre-eruptively. In this case, upon post-eruption
Knowledge of the mechanisms by which fluo- acidic challenge, FS would be released to the fluid
ride promotes caries control is essential for the phase (FL), thus inhibiting demineralization and
achievement of the maximum benefits of this enhancing remineralization. Additionally, ingest-
element with minimum risk of side effects. The ed fluoride can exert a topical mechanism of ac-
main action of fluoride for caries control occurs tion when it recirculates in the oral environment
through its topical effect. Fluoride present in through saliva.
low, sustained concentrations (sub-ppm range) The essence of de- and remineralization pro-
in the oral fluids during an acidic challenge is cesses, as well as the interactions with fluoride that
able to absorb to the surface of the apatite crys- were described for enamel, also apply to dentin.
tals, inhibiting demineralization. When the pH The main differences are that dentin is more sus-
is reestablished, traces of fluoride in solution ceptible to caries attack than enamel, with a criti-
will make it highly supersaturated with respect cal pH more than 1 pH unit higher. Consequently,
to fluorhydroxyapatite, which will speed up the dentin demineralizes faster and remineralizes
process of remineralization. The mineral formed slower, requiring higher fluoride concentrations
under the nucleating action of the partially dis- and frequencies of application when compared
solved minerals will then preferentially include with enamel.
fluoride and exclude carbonate, rendering the

1 Keyes PH: The infectious and transmis- 4 McDonagh MS, Whiting PF, Wilson PM, 7 Whelton HP, Ketley CE, McSweeney F,
sible nature of experimental dental car- Sutton AJ, Chestnutt I, Cooper J, Misso O’Mullane DM: A review of fluorosis in
ies: findings and implications. Arch Oral K, Bradley M, Treasure E, Kleijnen J: the European Union: prevalence, risk
Biol 1960;1:304–320. Systematic review of water fluoridation. factors and aesthetic issues. Community
2 Dean HT, Arnold FA, Elvolve E: Addi- BMJ 2000;321:855–859. Dent Oral Epidemiol 2004;32(suppl 1):
tional studies of the relation of fluoride 5 Clark DC: Trends in prevalence of dental 9–18.
domestic waters to dental caries experi- fluorosis in North America. Community 8 Chankanka O, Levy SM, Warren JJ, Chal-
ence in 4,425 white children aged 12–14 Dent Oral Epidemiol 1994;22:148–152. mers JM: A literature review of aesthetic
years in 13 cities in 4 states. Public 6 Khan A, Moola MH, Cleaton-Jones P: perceptions of dental fluorosis and rela-
Health Rep 1942;57:1155–1179. Global trends in dental fluorosis from tionships with psychosocial aspects/oral
3 Bratthall D, Hansel-Petersson G, Sund- 1980 to 2000: a systematic review. SADJ health-related quality of life. Commu-
berg H: Reasons for the caries decline: 2005;60:418–421. nity Dent Oral Epidemiol 2010;38:
what do the experts believe? Eur J Oral 97–109.
Sci 1996;104:416–422; discussion
423–425, 430–432.

112 Buzalaf · Pessan · Honório · ten Cate

9 Fejerskov O, Kidd EA, Nyvad B, Baelum 21 Weatherell JA, Deutsch D, Robinson C, 35 Marquis RE, Clock SA, Mota-Meira M:
V: Defining the disease: an introduction; Hallsworth AS: Assimilation of fluoride Fluoride and organic weak acids as mod-
in Fejerskov O, Kidd E (eds): Dental Car- by enamel throughout the life of the ulators of microbial physiology. FEMS
ies The Disease and its Clinical Manage- tooth. Caries Res 1977;11(suppl 1): Microbiol Rev 2003;26:493–510.
ment, ed 2. Oxford, Blackwell Munks- 85–115. 36 Nakajo K, Imazato S, Takahashi Y, Kiba
gaard, 2008, pp 3–6. 22 ten Cate JM: In vitro studies on the W, Ebisu S, Takahashi N: Fluoride
10 Featherstone JD, Lussi A: Understanding effects of fluoride on de- and remineral- released from glass-ionomer cement is
the chemistry of dental erosion. Monogr ization. J Dent Res 1990;69 Spec responsible to inhibit the acid produc-
Oral Sci 2006;20:66–76. No:614–9; discussion 34–6. tion of caries-related oral streptococci.
11 ten Cate JM, Larsen MJ, Pearce EIF, 23 Featherstone JD, Glena R, Shariati M, Dent Mater 2009;25:703–708.
Fejerskov O: Chemical interactions Shields CP: Dependence of in vitro 37 Pessan JP, Alves KM, Ramires I, et al.:
between the tooth and oral fluids; in demineralization of apatite and reminer- Effects of regular and low-fluoride den-
Fejerskov O, Kidd E (eds): Dental Caries alization of dental enamel on fluoride tifrices on plaque fluoride. J Dent Res
the Disease and Its Clinical Manage- concentration. J Dent Res 1990;69 Spec 2010;89:1106–1110.
ment. Oxford, Blackwell Munksgaard, No:620–625, discussion 634–636. 38 Pessan JP, Sicca CM, de Souza TS, da
2008, pp 209–231. 24 Øgaard B, RØlla G, Ruben J, Dijkman T, Silva SM, Whitford GM, Buzalaf MA:
12 Klont B, ten Cate JM: Remineralization Arends J: Microradiographic study of Fluoride concentrations in dental plaque
of bovine incisor root lesions in vitro: demineralization of shark enamel in a and saliva after the use of a fluoride den-
the role of the collagenous matrix. Caries human caries model. Scand J Dent Res tifrice preceded by a calcium lactate
Res 1991;25:39–45. 1988;96:209–211. rinse. Eur J Oral Sci 2006;114:489–493.
13 Kleter GA, Damen JJ, Everts V, Niehof J, 25 Arends J, Christoffersen J: Nature and 39 Pessan JP, Silva SM, Lauris JR, Sampaio
ten Cate JM: The influence of the organic role of loosely bound fluoride in dental FC, Whitford GM, Buzalaf MA: Fluoride
matrix on demineralization of bovine caries. J Dent Res 1990;69 Spec No: uptake by plaque from water and from
root dentin in vitro. J Dent Res 1994;73: 601–605, discussion 634–636. dentifrice. J Dent Res 2008;87:461–465.
1523–1529. 26 Featherstone JD: Prevention and reversal 40 Whitford GM, Buzalaf MA, Bijella MF,
14 Nyvad B, Fejerskov O: An ultrastructural of dental caries: role of low level fluo- Waller JL: Plaque fluoride concentra-
study of bacterial invasion and tissue ride. Community Dent Oral Epidemiol tions in a community without water flu-
breakdown in human experimental 1999;27:31–40. oridation: effects of calcium and use of a
root-surface caries. J Dent Res 1990;69: 27 ten Cate JM, Featherstone JD: Mechanis- fluoride or placebo dentifrice. Caries Res
1118–1125. tic aspects of the interactions between 2005;39:100–107.
15 Schupbach P, Guggenheim B, Lutz F: fluoride and dental enamel. Crit Rev 41 Lynch RJ, Navada R, Walia R: Low-levels
Human root caries: histopathology of Oral Biol Med 1991;2:283–296. of fluoride in plaque and saliva and their
initial lesions in cementum and dentin. 28 RØlla G: On the role of calcium fluoride effects on the demineralisation and
J Oral Pathol Med 1989;18:146–156. in the cariostatic mechanism of fluoride. remineralisation of enamel; role of fluo-
16 Tjäderhane L, Larjava H, Sorsa T, Uitto Acta Odontol Scand 1988;46:341–345. ride toothpastes. Int Dent J 2004;54:
VJ, Larmas M, Salo T: The activation 29 Whitford GM, Wasdin JL, Schafer TE, 304–309.
and function of host matrix metallopro- Adair SM: Plaque fluoride concentra- 42 Baysan A, Lynch E, Ellwood R, Davies R,
teinases in dentin matrix breakdown in tions are dependent on plaque calcium Petersson L, Borsboom P: Reversal of
caries lesions. J Dent Res 1998;77: concentrations. Caries Res 2002;36: primary root caries using dentifrices
1622–1629. 256–265. containing 5,000 and 1,100 ppm fluo-
17 Chaussain-Miller C, Fioretti F, Goldberg 30 Singh KA, Spencer AJ: Relative effects of ride. Caries Res 2001;35:41–46.
M, Menashi S: The role of matrix metal- pre- and post-eruption water fluoride on 43 Nordstrom A, Mystikos C, Ramberg P,
loproteinases (MMPs) in human caries. caries experience by surface type of per- Birkhed D: Effect on de novo plaque
J Dent Res 2006;85:22–32. manent first molars. Community Dent formation of rinsing with toothpaste
18 Frank RM, Voegel JC: Ultrastructure of Oral Epidemiol 2004;32:435–446. slurries and water solutions with a high
the human odontoblast process and its 31 Singh KA, Spencer AJ, Brennan DS: fluoride concentration (5,000 ppm). Eur
mineralisation during dental caries. Car- Effects of water fluoride exposure at J Oral Sci 2009;117:563–567.
ies Res 1980;14:367–380. crown completion and maturation on 44 Madlena M, Dombi C, Gintner Z, Ban-
19 Shellis RP: Effects of a supersaturated caries of permanent first molars. Caries oczy J: Effect of amine fluoride/stannous
pulpal fluid on the formation of caries- Res 2007;41:34–42. fluoride toothpaste and mouthrinse on
like lesions on the roots of human teeth. 32 Marquis RE: Antimicrobial actions of dental plaque accumulation and gingival
Caries Res 1994;28:14–20. fluoride for oral bacteria. Can J Micro- health. Oral Dis 2004;10:294–297.
20 Moreno EC, Varughese K, Hay DI: Effect biol 1995;41:955–964. 45 Gerardu VA, van Loveren C, Heijnsbroek
of human salivary proteins on the pre- 33 Koo H: Strategies to enhance the biologi- M, Buijs MJ, van der Weijden GA, ten
cipitation kinetics of calcium phosphate. cal effects of fluoride on dental biofilms. Cate JM: Effects of various rinsing proto-
Calcif Tissue Int 1979;28:7–16. Adv Dent Res 2008;20:17–21. cols after the use of amine fluoride/stan-
34 ten Cate JM, van Loveren C: Fluoride nous fluoride toothpaste on the acid pro-
mechanisms. Dent Clin North Am duction of dental plaque and tongue
1999;43:713–42, vii. flora. Caries Res 2006;40:245–250.

Mechanisms of Action of Fluoride 113

46 Paraskevas S, van der Weijden GA: A 53 ten Cate JM: Current concepts on the 62 ten Cate JM, Damen JJ, Buijs MJ: Inhibi-
review of the effects of stannous fluoride theories of the mechanism of action of tion of dentin demineralization by fluo-
on gingivitis. J Clin Periodontol 2006;33: fluoride. Acta Odontol Scand 1999;57: ride in vitro. Caries Res 1998;32:
1–13. 325–329. 141–147.
47 Bellamy PG, Jhaj R, Mussett AJ, Barker 54 Zero DT, Raubertas RF, Pedersen AM, 63 Herkstroter FM, Witjes M, Arends J:
ML, Klukowska M, White DJ: Compari- Fu J, Hayes AL, Featherstone JD: Studies Demineralization of human dentine
son of a stabilized stannous fluoride/ of fluoride retention by oral soft tissues compared with enamel in a pH-cycling
sodium hexametaphosphate dentifrice after the application of home-use topical apparatus with a constant composition
and a zinc citrate dentifrice on plaque fluorides. J Dent Res 1992;71: during de- and remineralization periods.
formation measured by digital plaque 1546–1552. Caries Res 1991;25:317–322.
imaging (DPIA) with white light illumi- 55 Gaugler RW, Bruton WF: Fluoride con- 64 Lynch E, Baysan A, Ellwood R, Davies R,
nation. J Clin Dent 2008;19:48–54. centration in dental plaque of naval Petersson L, Borsboom P: Effectiveness
48 Bellamy PG, Khera N, Day TN, Barker recruits with and without caries. Arch of two fluoride dentifrices to arrest root
ML, Mussett AJ: A randomized clinical Oral Biol 1982;27:269–272. carious lesions. Am J Dent 2000;13:
trial to compare plaque inhibition of a 56 Schamschula RG, Sugar E, Un PS, Toth 218–220.
sodium fluoride/potassium nitrate den- K, Barmes DE, Adkins BL: Physiological 65 Laheij AM, van Strijp AJ, van Loveren C:
tifrice versus a stabilized stannous fluo- indicators of fluoride exposure and utili- In situ remineralisation of enamel and
ride/sodium hexametaphosphate denti- zation: an epidemiological study. Com- dentin after the use of an amine fluoride
frice. J Contemp Dent Pract 2009;10:1–9. munity Dent Oral Epidemiol 1985;13: mouthrinse in addition to twice daily
49 He T, Sun L, Li S, Ji N: The anti-plaque 104–107. brushings with amine fluoride tooth-
efficacy of a novel stannous-containing 57 Nobre dos Santos M, Melo dos Santos L, paste. Caries Res 2010;44:260–266.
sodium fluoride dentifrice: a random- Francisco SB, Cury JA: Relationship 66 Vale GC, Tabchoury CP, Del Bel Cury
ized and controlled clinical trial. among dental plaque composition, daily AA, Tenuta LM, Ten Cate JM, Cury JA:
Am J Dent 2010;23:11–16. sugar exposure and caries in the primary APF and dentifrice effect on root den-
50 Benelli EM, Serra MC, Rodrigues AL Jr, dentition. Caries Res 2002;36:347–352. tine demineralization and biofilm.
Cury JA: In situ anticariogenic potential 58 Hoppenbrouwers PM, Driessens FC, J Dent Res 2011;90:77–81.
of glass ionomer cement. Caries Res Borggreven JM: The demineralization of 67 ten Cate JM: Remineralization of deep
1993;27:280–284. human dental roots in the presence of enamel dentine caries lesions. Aust Dent
51 Seppa L, Korhonen A, Nuutinen A: fluoride. J Dent Res 1987;66:1370–1374. J 2008;53:281–285.
Inhibitory effect on S. mutans by 59 Arends J, Christoffersen J, Buskes JA, 68 Buzalaf MA, Hannas AR, Magalhaes AC,
fluoride-treated conventional and resin- Ruben J: Effects of fluoride and meth- Rios D, Honorio HM, Delbem AC: pH-
reinforced glass ionomer cements. Eur J anehydroxydiphosphate on enamel and cycling models for in vitro evaluation of
Oral Sci 1995;103:182–185. on dentine demineralization. Caries Res the efficacy of fluoridated dentifrices for
52 Seppa L, Torppa-Saarinen E, Luoma H: 1992;26:409–417. caries control: strengths and limitations.
Effect of different glass ionomers on the 60 ten Cate JM, Buijs MJ, Damen JJ: pH- J Appl Oral Sci 2010;18:316–334.
acid production and electrolyte metabo- cycling of enamel and dentin lesions in 69 Hamilton IR, Bowden GHW: Fluoride
lism of Streptococcus mutans Ingbritt. the presence of low concentrations of effects on oral bacteria; in Fejerskov O,
Caries Res 1992;26:434–438. fluoride. Eur J Oral Sci 1995;103: Ekstrand J, Burt BA (eds): Fluoride in
362–367. Dentistry, ed 2. Copenhagen, Muks-
61 ten Cate JM, Duijsters PP: Influence of gaard, 1996, pp 230–251.
fluoride in solution on tooth demineral-
ization. I. Chemical data. Caries Res

Marília Afonso Rabelo Buzalaf

Department of Biological Sciences
Bauru Dental School, University of São Paulo
Al. Octávio Pinheiro Brisolla, 9–75
Bauru-SP, 17012–901 (Brazil)
Tel. +55 14 3235 8346, E-Mail

114 Buzalaf · Pessan · Honório · ten Cate

Impact of Fluoride in the Prevention of Caries and Erosion
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 115–132

Topical Use of Fluorides for Caries Control

Juliano Pelim Pessana ⭈ Kyriacos Jack Toumbac ⭈
Marília Afonso Rabelo Buzalafb
aDepartment of Pediatric Dentistry and Public Health, Araçatuba Dental School, São Paulo State University, Araçatuba, and
bDepartment of Biological Sciences, Bauru Dental School, University of São Paulo, Bauru, Brazil; cDivision of Child Dental Health,
Leeds Dental Institute, University of Leeds, Leeds, UK

of all ages, and additional fluoride therapy should also be
Since the early findings on the protective effects of flu-
targeted towards individuals at high caries risk.
oride present in drinking water upon caries incidence
Copyright © 2011 S. Karger AG, Basel
and prevalence, intensive research has been conducted
in order to determine the benefits, safety, as well as the
cost-effectiveness of other modalities of fluoride deliv- Since the early findings of the protective effects
ery. The present chapter reviews the various forms of top- of fluoride present in drinking water upon car-
ical fluoride use – professionally and self-applied – with ies incidence and prevalence over 70 years ago,
special emphasis on clinical efficacy and possible side intensive research has been conducted in order to
effects. The most widely used forms of fluoride delivery determine the benefits, safety, as well as the cost-
have been subject of several systematic reviews, provid- effectiveness of modalities of fluoride delivery
ing strong evidence supporting the use of dentifrices, other than water. These include both topical and
gels, varnishes and mouth rinses for the control of car- systemic methods of fluoride delivery, which dif-
ies progression. Dentifrices with fluoride concentrations fer widely regarding fluoride concentration and
of 1,000 ppm and above have been shown to be clini- mode of application. The current scientific con-
cally effective in caries prevention when compared to a sensus regards a constant supply of low levels of
placebo treatment, but the evidence regarding formula- fluoride, especially at the biofilm/saliva/dental
tions with 450–550 ppm is still subject of debate. There- interface (topical effect), as being the most ben-
fore, the recommendation for low-fluoride dentifrice use eficial in preventing dental caries [1]. Therefore,
must take into account both risks and benefits. The evi- the classification of the methods of fluoride de-
dence for the combined use of two modalities of fluo- livery into topical and systemic has been recently
ride application in comparison to a single modality is still questioned. Ellwood et al. [2] proposed that the
inconsistent, implying that more studies with adequate methods should be classified according to their
methodology are needed to determine the real benefits mode of application, as follows:
of each method. Considering the currently available evi- – Community Methods: introduced on a
dence and risk-benefit aspects, it seems justifiable to rec- population basis (water, milk and salt
ommend the use of fluoridated dentifrices to individuals fluoridation);
– Self-Applied Methods: used at home choice of a certain fluoride vehicle – or com-
(toothpastes, mouth rinses, tablets, drops, bination of modalities – on scientific knowl-
lozenges and chewing gums); edge rather than just personal preference. It is
– Professional Methods: delivered by healthcare worth mentioning that systematic reviews usu-
professionals (solutions, gels, foams, varnishes, ally adopt the prevented fraction as the primary
slow-release fluoride devices and fluoride- estimate of effect of the treatment, which refers
releasing dental materials). to the caries increment in the treatment group
However, for didactical reasons, the present expressed as a percentage of the control group
chapter will use the term topical fluorides to de- [6]. This estimate has also been adopted in the
scribe the methods that provide fluoride to ex- present review.
posed surfaces of the dentition at elevated con-
centrations for a local protective effect, and which
are therefore not intended for ingestion [3]. These Fluoride Compounds
include both self-applied and professional meth-
ods of fluoride delivery. Community methods are There is a diversity of fluoride compounds used
be discussed in the paper by Sampaio and Levy in fluoride agents available to the public and to
[this vol., pp. 133–145]. healthcare professionals. These differ greatly ac-
It is important to highlight that most of the cording to their chemical structures, which ulti-
scientific knowledge on the effectiveness of topi- mately have implications on the mode of action
cal fluoride methods was obtained when caries for each compound. According to Axelsson [7],
incidence and prevalence were still very high, the three main categories are:
so even modest interventions led to significant 1 inorganic compounds: readily soluble salts
reductions in caries levels [4]. As a substantial that provide free fluoride;
decrease in caries prevalence has been observed 2 monofluorophosphate-containing compounds:
over the last decades, authorities have questioned fluoride is covalently bound to PO3–2 ions and
the real benefit of topical fluoride applications. requires hydrolysis to release fluoride ions;
A large number of clinical trials have been con- 3 organic fluorides: fluoride bound to organic
ducted in order to assess the clinical effective- compounds.
ness of the various modalities of topical fluo- The main fluoride compounds, along with
ride administration, using different protocols their main characteristics and vehicles in which
that are not always appropriate and do not allow they are frequently used, are briefly summarized
direct comparisons between results. In order to in table 1. It is worth mentioning that some com-
avoid misinterpretation of the information from pounds can be used in combination in different
those trials due to methodological issues, meta- vehicles, such as toothpastes (NaF/MFP, SnF2/
analytical approaches have been used in system- AmF), prophylaxis pastes (NaF/MFP, NaF/AmF)
atic reviews aiming to provide real estimates of and mouth rinses (SnF2/AmF) [7].
the effectiveness and safety of the various forms
of topical fluoride administration. In the present
chapter, the main conclusions of the most recent Solutions
systematic reviews conducted by the Cochrane
Central Register of Controlled Trials (the most Fluoridated solutions were the first vehicles of
comprehensive source of reports of trials avail- professionally applied fluorides aiming at a re-
able) [5] is presented whenever evidence is avail- duction in caries levels [8]. These included neu-
able, so professionals and patients can base their tral NaF solutions, as well as SnF2, acidulated

116 Pessan · Toumba · Buzalaf

Table 1. Main fluoride compounds used in topical formulations

Compounds Characteristics Vehicles in which the

compound is used

Inorganic compounds
Sodium fluoride (NaF) the most commonly used fluoride compound (both self- dentifrices
application and professional use); mouth rinses
when in solution, NaF salt readily releases fluoride into chewing-gums
saliva, dental plaque, pellicle and enamel crystallites solutions
prophylaxis pastes
slow-release devices
Stannous fluoride (SnF2) releases both F– and Sn+2 ions into the oral environment, dentifrices
which have cariostatic and antimicrobial properties, mouth rinses
respectively; solutions
tooth staining and instability are the main disadvantages gels
prophylaxis pastes
Ammonium fluoride (NH4F) although investigated intensively some decades ago, it is solutions
currently unused – mainly due to its unpleasant taste and
lack of superiority in clinical performance over NaF
Titanium tetrafluoride (TiF4) able to significantly reduce enamel solubility (as solution), solutions
due to the formation of a glaze on enamel and dentine; varnishes
currently being tested in solutions/varnishes as
preventative for caries and erosion1

Organic compounds
Amine fluoride associated with a reduction in plaque adhesiveness dentifrices
due to the greater affinity of hydrophilic counter-ions to gels
the enamel; mouth rinses
also associated with complexed store of fluoride ions, prophylaxis pastes
which may enhance diffusion through carious enamel
Silane fluoride associated with complexed store of fluoride (similarly to varnishes
amine fluoride);
unlike NaF, MFP and SnF2 (which dissolve in water and
release fluoride ions), this compound is insoluble, releasing
HF after contact with saliva, which diffuses into enamel
more efficiently than fluoride ions

Monofluorophosphate-containing compounds
Sodium can be used in both neutral and acidic vehicles; dentifrices
monofluorophosphate fluoride is covalently bound in Na2FPO3 and requires (neutral pH)
(Na2FPO3) hydrolysis in order to release fluoride ions; gels
both F– and PO3F–2 ions can diffuse into plaque and (neutral and acidic pH)
enamel, mainly in acidic pH, but the role of PO3F–2 ion
is not well established;
one of the main advantages is its compatibility with chalk-
based abrasives

Sources: Ellwood et al. [2], Axelsson [7] and Pessan et al. [47].
1See Magalhães et al. [this vol., pp. 158–170].

Topical Fluorides for Caries Control 117

phosphate fluoride (APF), AmF, NH4F and TiF4. application, suction devices must be used dur-
Most of the studies described in the literature, ing application, and the patient must be instruct-
however, used solutions of NaF, SnF2 and APF. ed to spit out repeated times after application.
These have differed regarding the mode of appli- Application time is 4 min for both neutral and
cation and frequency of use. NaF solutions were acidic products [11].
applied to all teeth and then allowed to dry for There is consistent evidence for the benefits
3 min (4 weekly applications at 3, 7, 10 and 13 of gels in caries prevention [12, 13]. A Cochrane
years of age), while SnF2 and APF solutions had review included data from 23 trials (7,747 chil-
to be constantly applied to the teeth for 3 min dren and adolescents), in which fluoride gels
(twice/year). Ripa [9] reviewed 35 clinical stud- were compared to a placebo treatment or no
ies evaluating the effectiveness of fluoridated so- treatment (table 2) [14]. The effect of fluori-
lutions, and concluded that NaF, SnF2 and APF dated gels in the permanent dentition was sig-
solutions presented similar clinical effectiveness nificantly influenced by the frequency of ap-
(around 30%), despite differences in fluoride plication, as well as the intensity of application
concentration, mode of application, frequency of (frequency × concentration) and self-application
use and other characteristics. With the introduc- (which can be associated with a higher frequency
tion of fluoride gels, which are much easier and of use). Therefore, fluoridated gels must be ap-
safer in clinical practice, fluoridated solutions plied 2–4 times per year (table 3), depending on
are no longer used and, therefore, will not be de- caries risk consideration, in order to achieve the
scribed in detail in the present chapter [10]. expected benefits in caries prevention. No con-
clusion could be drawn for the deciduous denti-
tion. The effect of fluoride gels obtained in the
Gels and Foams most recent meta-analysis (28% pooled DMFS
prevented fraction) is not substantially different
In contrast to solutions, fluoride gels have from that obtained more recently, when four new
been extensively used in both self-applied studies (published after the Cochrane review)
and professionally applied modalities, as gels were included [15].
are much more viscous than solutions. This Fluoridated foams became available more
property makes it possible to treat an entire arch recently, with compositions similar to APF gels.
at the same time, which reduces both the time However, as an amount of fluoride 4–5 times
of application and the risk of excessive ingestion lower is used during application of foams (due
of fluoride. When professionally used, gels can to the reduced density of these products), they
be applied with brushes or cotton pellets, but the can be considered as a safer option regarding the
use of trays (stock or custom made) minimizes risk of excessive fluoride intake. There are only a
the risk of excessive fluoride intake. Increased few studies comparing the effectiveness of foams
penetration of gels between the teeth can be and gels, and there seems to be little or no dif-
achieved by using thixotropic products (which ference between clinical efficacy of both vehicles
flow under pressure). If self-applied, gels are [16–18]. However, more studies with appropri-
usually used in trays or with a toothbrush. ate research protocols are still needed to address
Due to the high fluoride concentration in this issue before foams can be recommended as a
gels (0.5% F– in SnF2, 0.9% F– in neutral NaF substitute to gels.
and 1.23% F– in APF), care must be taken when
using these products in order to avoid side ef-
fects. The patient must remain seated during

118 Pessan · Toumba · Buzalaf

Table 2. Results of the Cochrane reviews on the effectiveness of topical fluoride methods in caries prevention

Year of Method Trials Children DMFS dmfs Main findings Main conclusions
publi- evaluated included included pooled pooled
cation in review in meta- PF with PF with
and meta- analysis 95% CI 95% CI %
analysis %

2002 gel 25 (23) 7,747 28 not effect varied according clear evidence of a
(19–37) available to type of control caries-inhibiting effect of
group used (DMFS PF fluoride gel;
19% higher in non- best estimate of the
placebo controlled magnitude of this effect,
trials); based on the 14
no significant placebo-controlled trials,
association between is a 21% reduction (95%
DMFS PF and baseline CI 14–28%) in DMFS PF
caries severity or
background fluoride
effect influenced by
increased frequency or
intensity of application
and self-application;
only 2 trials reported
on adverse events

2002 varnish 9 (7) 2,709 46 33 (19–48) no significant substantial caries-

(30–63) association between inhibiting effect in both
estimates of DMFS PF the permanent and the
and baseline caries deciduous dentitions;
severity or background little information
exposure to fluorides; concerning acceptability
power was limited due of treatment or possible
to the inclusion of few side effects

2003 toothpaste 74 (70) 42,300 24 not effect increased with benefits of fluoride
(2–28) available higher baseline levels toothpastes in caries
of DMFS, higher prevention firmly based
fluoride concentration, on trials of relatively
higher frequency of high quality
use, and supervised
brushing, but was not
influenced by exposure
to water fluoridation;
little information
concerning the
deciduous dentition or
adverse effects

Topical Fluorides for Caries Control 119

Table 2. Continued

Year of Method Trials Children DMFS dmfs Main findings Main conclusions
publi- evaluated included included pooled pooled
cation in review in meta- PF with PF with
and meta- analysis 95% CI 95% CI %
analysis %
2003 mouthrinse 36 (34) 14,600 26 not no significant clear reduction in caries
(23–30) available association between increment in children
DMFS pooled PF and who regularly use
baseline caries severity, fluoride mouth rinses at
background exposure two main strengths (230
to fluorides, rinsing or 900 ppm) and rinsing
frequency and fluoride frequencies (daily or
concentration weekly/fortnightly)

DMFS = Decayed/missing/filled surface (permanent teeth); dmfs = decayed/missing/filled surface (deciduous teeth);
PF = prevented fraction: (mean increment in the control group – mean increment in the intervention group)/mean increment
in the control group.

Varnishes There is evidence attesting to a substantial

caries-inhibiting effect in both the permanent
Fluoridated varnishes were introduced into the (46% reduction in pooled DMFS prevented frac-
market in the 1960s, and are intended for pro- tion) and deciduous (33% reduction in pooled
fessional application only. The main advantag- dmfs prevented fraction) dentitions, based on a
es of varnishes are the prolonged contact time Cochrane review including data from 2,709 chil-
between fluoride and the tooth surfaces (in- dren (table 2) [19]. As for rinses and gels, no sig-
creases fluoride uptake by dental hard tissues, nificant association between estimates of DMFS
as well as the formation of CaF2 reservoirs), prevented fractions and baseline caries severity or
and the possibility of using very small amounts background exposure to fluorides was observed.
of the product (a thin layer), which minimizes The results of that systematic review, however,
the risk of excessive fluoride ingestion. These must be interpreted with caution, due to the low
products are much more concentrated than gels, number of trials included, which limits the pow-
with typical concentrations of 22,600 ppm fluo- er of the analyses. A recent review found six tri-
ride (in NaF varnishes), 7,000 ppm fluoride (in als not included in the Cocrhane review, and the
difluorosilane varnishes) or 56,300 ppm flu- pooled DMFS prevented fractions ranged from 34
oride (in 6% NaF + 6% CaF2 varnishes) [19]. to 57% [15].
Duraphat (Inpharma, Germany – NaF) and Despite having higher fluoride concentrations,
Fluor Protector (Vivadent, Liechtenstein – dif- varnishes can be regarded as a safer option when
luorosilane) are the most used and studied prod- compared to gels, due to the small amount used
ucts. In order to achieve the maximum benefits during application (table 4). Fluoride concentra-
for caries prevention, varnishes must be applied tions in plasma and urine of children were report-
2–4 times/year (table 3), depending on caries- ed to be lower than toxic levels after the applica-
risk considerations. tion of a fluoride varnish [20, 21].

120 Pessan · Toumba · Buzalaf

Table 3. Evidence-based clinical recommendations for professionally applied topical fluoride

Age category of recall patients

<6 years 6–18 years >18 years

recommendation grade of strength recommendation grade of strength recommendation grade of strength

evidence of recom- evidence of recom- evidence of recom-
mendation mendation mendation

Low may not receive Ia B may not receive Ia B may not receive IV D
risk additional additional benefit additional benefit
benefit from from professional from professional
professional topical fluoride topical fluoride
topical fluoride application1 application1

Mode- varnish Ia A varnish Ia A varnish IV D2

rate application at application at application at
risk 6-month 6-month 6-month
intervals intervals intervals
fluoride gel Ia A fluoride gel IV D3
application at application at
6-month intervals 6-month intervals

High varnish Ia A varnish Ia A varnish IV D2

risk application at application at application at
6-month 6-month 6-month
intervals intervals intervals
varnish Ia D4 varnish Ia A varnish IV D2
application at application application
3-month at 3-month at 3-month
intervals intervals intervals
fluoride gel Ia A fluoride gel IV D3
application at application at
6-month intervals 6-month intervals
fluoride gel IV D3 fluoride gel IV D3
application at application at
3-month intervals 3-month intervals

Evidence from systematic reviews of randomized controlled trials (Ia) or expert committee reports or opinions or clinical experience of
respected authorities (IV).
Laboratory data have demonstrated the equivalence of foam to gels in terms of fluoride release; however, only 2 clinical trials have been
published evaluating its effectiveness. Because of this, the recommendations for use of fluoride varnish and gels have not been extrapo-
lated to foams.
Because there is insufficient evidence to address whether or not there is a difference in the efficacy of sodium fluoride vs. acidulated phos-
phate fluoride gels, the clinical recommendations do not distinguish between these formulations. Application time for fluoride gel and
foam should be 4 min. A 1-min fluoride application is not endorsed.
Source: American Dental Association [56].
1 Fluoridated water and toothpastes may already provide adequate caries prevention. Decisions on whether to apply topical fluoride

should balance this consideration with professional judgment and patient preferences.
2 Although there are no clinical trials, there is reason to believe that fluoride varnish would work similarly in this age group.
3 Although there are no clinical trials, there is reason to believe that fluoride gels would work similarly in this age group.
4 Emerging evidence indicates that applications more frequent than twice per year may be more effective in preventing caries.

Topical Fluorides for Caries Control 121

Table 4. Comparison of professionally applied fluoride methods (varnish, gel and foam) regarding their effectiveness,
clinical use, toxicity, cost and patient acceptance.

Caries Clinical Fluoride ingestion Cost Acceptability

prevention application

Varnish effective in easy; lowest risk; most preferred by

high-risk application moisture can be expensive patients and
children time varies better controlled than hygienists over gel
(permanent gel or foam

Gel effective in easy; % retained can be low cost well-tolerated by

high-risk 4-min substantial; most patients, but
children application procedure must be varnish is preferred
(permanent time followed to reduce
teeth) risk

Foam not clinically easy; risk of over-ingestion low cost not formally
tested; 4-min is less compared with assessed;
likely to be application gel likely to be to
similar to gel time similar to gel

Source: Hawkins et al. [59] (modified).

Rinses children; table 2) [22]. Such findings apply to both

daily and weekly/fortnightly rinses with 230- and
Fluoride mouth rinses have been successful- 900-ppm fluoride solutions, respectively, indicat-
ly used in dentistry for about 6 decades, either ing that the mode of application will depend on
as self-application or community-based meth- personal preferences (when used at home) and on
ods. NaF solutions are the most widely used, al- the availability of personnel to supervise the use
though formulations containing other fluoride of the solutions (in school-based programs). No
compounds are also available. Typically, solutions significant association was found between DMFS
containing 230 ppm fluoride are intended for dai- pooled prevented fraction and baseline caries se-
ly use at home, while a higher concentration (900 verity, background exposure to fluorides, rinsing
ppm fluoride) is used in community-based pro- frequency and fluoride concentration. No conclu-
grams at weekly/fortnightly intervals. The main sion could be drawn for the deciduous dentition.
advantages of the method include effectiveness, The benefits of mouthrinsing were shown to be
simplicity of use and the possibility of application affected by subsequent discontinuation [23].
by a non-dental professional, which ultimately af-
fects cost.
A clear reduction (26% in pooled DMFS pre- Dentifrices
vented fraction) in caries increments in the per-
manent dentition of children who regularly Fluoridated dentifrices are by far the most wide-
use fluoride rinses was found by a recent meta- spread form of fluoride delivery and are currently
analysis of 34 clinical trials (involving 14,600 used by over 500 million people worldwide [24].

122 Pessan · Toumba · Buzalaf

Table 5. Ingredients commonly used in fluoridated dentifrices

Active agents Other compounds

fluoride – 1 compound or 2 (in combination) abrasive particles

agents for enhancement of the fluoride effect detergents, foaming agents
chemical plaque control agents flavoring agents, preservatives, and coloring agents
anti-calculus agents thickeners, agents to regulate viscosity
buffering systems water

Source: Axelsson [7] (modified).

Considering the multifactorial etiology of dental dental plaque). In order to achieve these goals, the
caries, toothbrushing with a fluoridated denti- various formulations available may contain the
frice can be regarded as the best method of fluo- components listed in table 5. Dentifrices can vary
ride use, as it combines the mechanical removal widely in their composition, depending on the
or disruption of dental plaque (which is also ben- benefits that each formulation intends to provide
eficial in periodontal health maintenance) with (i.e. anti-plaque, anti-calculus, whitening). Due to
the caries-protective effect of fluoride [25]. It has the scope of this chapter, however, only factors re-
been regarded as the method of choice by public lated to the anti-caries properties of fluoridated
health authorities, as it is convenient, inexpen- dentifrices will be discussed.
sive, culturally accepted and widespread [26]. The abrasive system is an important compo-
Besides the therapeutic properties of dentifrices, nent that can affect fluoride availability, which
their cosmetic benefits (related to cleanliness, re- will ultimately interfere with its clinical perfor-
moval of stains, whiteness and protection against mance. It is known that the first formulations
oral malodor) constitute additional reasons for of toothpastes failed to show a significant effect
the wide acceptance of this method [2]. on the reduction in caries levels, due to the use
Dentifrices are available as various formula- of incompatible fluoride compounds and abra-
tions of gels and pastes, and may contain bleach- sive systems [27]. The compatibility of MFP, NaF
ing, anti-plaque and desensitizing ingredients, and other formulations with the abrasive systems
with labeling and flavoring characteristics di- most commonly used are [2, 7, 28]:
rected to adults and children. They must not be – MFP:
confounded with prophylaxis pastes (which have Alumina trihydrate (Al2O3•3H2O)
higher fluoride content, are more abrasive and are Anhydrous dicalcium phosphate (CaHPO4)
used less frequently) or gels (which do not have Dicalcium phosphate dihydrate (CaHPO4•
abrasive particles, are much more concentrated 2H2O)
and used at a lower frequency) [4] Calcium carbonate (CaCO3)
– MFP, NaF and other formulations
Composition Calcium phosphate (Ca2P2O7)
As previously mentioned, the greatest advantages Hydrated silica (SiO2)
of dentifrices are the removal or disruption of den- Sodium bicarbonate (NaHCO3)
tal plaque associated with fluoride delivery to den- Insoluble sodium metaphosphate (NaPO3)x
tal hard tissues and intraoral reservoirs (especially Acrylic polymer

Topical Fluorides for Caries Control 123

Clinical Efficacy fluoride concentration, an additional 6% reduc-
Marinho et al. [4] evaluated the effect of fluori- tion in caries is obtained [27, 30]. As it could be
dated dentifrices in caries increments through a wrongly assumed from those results that tooth-
meta-analysis involving data from 42,300 chil- pastes with higher fluoride concentrations should
dren and adolescents (table 2). Due to the large always be preferred in order to achieve the maxi-
number of trials (n = 70) of relatively high qual- mum benefits of these products, this topic needs
ity, there is strong evidence for the clinical effec- to be further explored, mainly in terms of risks/
tiveness of toothpastes in preventing caries in that benefits.
age group (pooled DMFS prevented fraction was Typically, low-fluoride toothpastes (usual-
24%). Unlike for rinses, gels and varnishes, the ly containing 500–550 ppm) have been recom-
effect of fluoridated toothpastes increased with mended to children under 7 years of age, in order
higher baseline levels of DMFS, higher fluoride to minimize fluoride ingestion from this source.
concentration, higher frequency of use and super- Toothpastes with fluoride concentrations in the
vised brushing. Background exposure to fluori- range of 1,000–1,500 ppm are usually indicated
dated water did not influence the effects of tooth- for children older than 7 years, adolescents and
pastes in caries reduction. Even considering the adults, and higher concentrations would be indi-
large number of trials and subjects evaluated, lit- cated for patients at high caries risk or to prevent
tle information was found concerning the decidu- root caries [26, 31].
ous dentition or adverse effects, such as fluoro- The results of a recent systematic review of
sis. After the publication of that Cochrane review studies comparing the clinical efficacy of tooth-
[4], ten new studies addressing the same topic pastes with different fluoride concentrations are
were published (between 2002 and 2008) and the summarized in table 6. When compared to pla-
pooled DMFS prevented fraction was around 25% cebo, significant differences in caries increments
[29], not substantially different from that found were only seen for formulations containing 1,000
by Marinho et al. [4]. ppm fluoride or above [6], suggesting that low-
fluoride toothpastes are less effective for caries
Factors Affecting Clinical Efficacy control when compared to conventional formula-
Among the factors that affect the clinical efficacy tions. Care must be taken when interpreting these
of toothpaste formulations, fluoride concentration results, in order to avoid misleading the reader.
and pH are the most relevant and, therefore, spe- First of all, the conclusion that the efficacy of
cial emphasis will be given to these. Other factors dentifrices containing 450–550 ppm is not signifi-
include frequency of brushing, amount of tooth- cantly different from placebo was based only on
paste used, rinsing behavior after brushing, tim- 2 trials, while the number of studies comparing
ing and duration of brushing, the fluoride com- placebo with conventional formulations (1,000–
pound in the toothpaste and age when brushing 1,500 ppm) was substantially higher (58 trials).
commenced. These will be briefly summarized. In addition, no conclusion could be taken when
comparing the clinical efficacy of low-fluoride
Fluoride Concentration and conventional toothpastes, as only one trial
It has been suggested that the fluoride concen- met the inclusion criteria of that review, clearly
tration of toothpastes is one of the main deter- indicating that additional randomized controlled
minants of their efficacy. Results of large clinical trials are still needed to fully address this issue.
trials comparing the effectiveness of toothpastes There is reason to be believe that the differenc-
in the range of 1,000–2,500 ppm fluoride indicate es between low-fluoride and conventional tooth-
that for every 500-ppm increase in toothpaste pastes may not be as large as might be expected.

124 Pessan · Toumba · Buzalaf

Table 6. Comparison of the clinical effectiveness of toothpastes with different fluoride concentrations

Comparisons Studies included in the Pooled DMFS prevented p value

meta-analysis, n fraction with 95% CI

Placebo vs.
250 ppm 3 8.9 (–1.6 to 19.4) 0.097
440/500/550 ppm 2 7.9 (–6.1 to 21.9) 0.27
1,000/1,055/1,100/1,250 ppm 54 22.2 (18.7 to 25.7) <0.00001
1,450/1,500 ppm 4 23.0 (15.3 to 28.9) <0.00001
2,400/2,500/2,800 ppm 4 36.6 (17.5 to 55.6) <0.00001

250 ppm vs.

1,000/1,055/1,100/1,250 ppm 2 16.8 (8.5 to 25.1) 0.000076

440/500/550 ppm vs.

1,000/1,055/1,100/1,250 ppm 1 0.5 (–15.0 to 16.0) n.c.
2,400/2,500/2,800 ppm 1 12.7 (–1.7 to 27.0) n.c.
1,450/1,500 ppm 6 9.6 (2.5 to 16.6) 0.0078

1,000/1,055/1,100/1,250 ppm vs.

1,700/2,000/2,200 ppm 2 9.4 (2.1 to 16.8) 0.011
2,400/2,500/2,800 ppm 6 12.2 (6.0 to 18.4) 0.00012

Source: Walsh et al. [6] (modified). n.c. = p value not calculated, as there was only 1 study for each comparison.

A recent randomized clinical trial demonstrat- toothpastes for preventing caries was reported
ed that the clinical performance of low-fluoride due to the lack of trials [6].
toothpastes is dependent on caries activity. It was Considering the evidence above, along with
shown that the clinical efficacy of a 500-ppm fluo- the conclusion of the systematic review by Walsh
ride dentifrice was similar to that of a 1,100-ppm et al. [6] on the uncertainty surrounding the es-
fluoride dentifrice when used by caries-inactive timates of dentifrices containing 450–550 ppm
children, but the low-fluoride dentifrice was less fluoride, it becomes clear that caution must be
effective than the conventional formulation in taken when recommending low-fluoride tooth-
controlling the progression of lesions in caries- pastes for the prevention of caries in the decid-
active children [32]. More recently, it has also uous dentition. While no conclusive evidence is
been demonstrated that plaque fluoride concen- available, the decision of recommending those
trations (an indirect indicator of clinical efficacy formulations will depend on caries activity, be-
of topical fluoride products [33]), observed 1 h af- sides professional and patient choices regarding
ter brushing with conventional and low-fluoride risks/benefits. It seems reasonable to recommend
toothpastes, were not significantly different be- low-fluoride (500 ppm) toothpastes for young
tween children residing in communities with children who are at risk of developing fluorosis
fluoridated and non-fluoridated drinking water in the permanent maxillary central incisors (less
[34]. Finally, for the deciduous dentition, uncer- than 3 years of age) but at low caries risk, espe-
tainty regarding the effectiveness of low-fluoride cially if they live in a fluoridated area. In all other

Topical Fluorides for Caries Control 125

3,500 3,500
a a
3,000 3,000

⌬Z (vol% min × µm)

⌬Z (vol% min × µm)

b b
2,500 2,500
d c
b 2,000
2,000 b
c c d d
1,500 c 1,500
d e e
1,000 c f,g 1,000
e e
500 f,g 500 e
0 0
0 275 550 825 1,100 1,375 0 275 550 825 1,100
a Dentifrice (µg F/g) b Dentifrice (µg F/g)

Fig. 1. Mean mineral loss (ΔZ) for the different fluoride concentrations in dentifrices at pH 5.5 (a) [35] and pH 4.5
(b) [36], compared to neutral formulations. Blue squares = Neutral toothpastes; Green squares = Acidic toothpastes;
Triangle = Commercial children’s toothpaste; Circle = Commercial 1,100 μg/g toothpaste. Means followed by distinct
letters are significantly different (Kruskal-Wallis, p<0.05). Bars indicate standard deviations.

cases, toothpastes containing at least 1,000 ppm when compared with neutral formulations [38].
fluoride should be used. Formulations with a pH of 4.5 were also shown
to have similar abrasiveness in comparison with
pH of the Dentifrice neutral dentifrices with the same fluoride concen-
The controversial evidence on the use of low- trations [39].
fluoride dentifrices for caries control has led to
increasing interest in strategies to improve the ef- Fluoride Compound Used in the Formulation
fectiveness of such formulations. Due to the in- Another factor that has been traditionally as-
verse relationship between CaF2 formation and sociated with the clinical efficacy of dentifrices
pH, dentifrices with acidic pH have been tested is the active fluoride agent used in the formu-
in clinical and laboratory studies. Results from lation (table 1). Most marketed dentifrices con-
in vitro experiments showed that the effective- tain sodium fluoride (NaF), or sodium mono-
ness of dentifrices containing 550 ppm fluoride fluorophosphate (SMFP), although formulations
(pH 5.5) [35] or 412 ppm fluoride (pH 4.5) [36] with SnF2 and amine fluoride are also available
were similar to that of a 1,100-ppm dentifrice at in some countries [2, 40]. There used to be a
neutral pH (fig. 1). These results were confirmed controversy regarding the clinical efficacy ob-
in a recent clinical trial evaluating caries progres- tained by NaF and SMFP toothpastes (the most
sion in the deciduous dentition of high caries-risk widely used formulations). Based on the prem-
children living in a fluoridated area. Caries pro- ise that fluoride only exerts its effects on de- and
gression rates observed 20 months after the use remineralization as a free ion, several reports
of a 550-ppm dentifrice in pH 4.5 were similar to have claimed the superiority of NaF formula-
those seen for the conventional 1,100-ppm neu- tions (which releases free F–), in comparison
tral toothpaste [37]. The superior clinical perfor- with SMFP, where fluoride is covalently bound
mance of the acidic toothpastes can be partially to phosphate, and requires enzymatic hydrolysis
explained by an increased plaque fluoride uptake to release free F–.

126 Pessan · Toumba · Buzalaf

Table 7. Potential factors affecting clinical effectiveness of toothpastes

Interpretation or recommendation Evidence

Fluoride concentration increase in prevented fraction (permanent dentition) for 1,000 ppm fluoride 1

Rinsing behaviour discourage rinsing with large volumes of water; 4, 5

encourage young children to spit out excess toothpaste

Frequency of brushing increase in prevented fraction moving from once to twice a day 1

Supervision lower prevented fraction with unsupervised toothbrushing 1

When to brush brush last thing at night and on one other occasion 4, 5

Type of fluoride toothpastes containing sodium fluoride, sodium monofluorophosphate or 1, 5

stannous fluoride are clinically effective

Age to commence advise parents/carers to begin brushing once the primary teeth have 4, 5
brushing commenced eruption

Background fluorides higher prevented fraction in presence of any background fluoride 1

(e.g. in water)

Mean initial caries increase in prevented fraction per unit increase in mean initial level of caries 1

Evidence levels: 1 = systematic review of at least one randomized controlled trial; 2 = at least 1 randomized
controlled study; 3 = non-randomized intervention studies; 4 = observational studies; 5 = traditional reviews,
expert opinion. Sources: Marinho [5] and Davies et al. [58].

The clinical efficacy of different formulations Other Factors Affecting Clinical Efficacy
of toothpastes has been addressed by several clin- As for all other vehicles of fluoride delivery, the
ical studies and literature reviews over the last primary goal is to enrich the intraoral fluoride
decades with differing conclusions [4, 40–45]. reservoirs at levels that can interfere with the dy-
Reviews conducted in the 1990s claim a difference namics of dental caries. Therefore, it is reason-
in effectiveness of 6–7% favoring NaF formula- able to assume that toothbrushing habits able to
tions [42, 43]. A recent Cochrane review com- increase or sustain intraoral fluoride levels will
paring formulations containing SMFP (22 trials), have a positive effect on the clinical efficacy of
SnF2 (19 trials), NaF (10 trials) and amine fluo- toothpastes. Other aspects, including mean ini-
ride (5 trials) did not find an association between tial caries and background exposure to fluorides,
the main types of fluoride compounds present in can also affect the clinical outcome when using
toothpaste formulations and the magnitude of the a fluoridated dentifrice. However, the degree of
treatment effect [4]. Nevertheless, the authors of scientific evidence surrounding these assump-
that review considered their result to be less reli- tions varies considerably among the factors that
able than evidence from head-to-head compari- may influence the clinical efficacy of toothpastes.
sons. The long-term significance of this difference Some factors that potentially influence the effec-
has been the subject of debate, so the question re- tiveness of fluoride dentifrices are summarized in
mains to be further investigated. table 7.

Topical Fluorides for Caries Control 127

Slow-Release Fluoride Devices 0.15 mg/day. This device was shown to promote a
significant increase in salivary and urinary fluo-
The rationale for the use of slow-release fluoride ride concentrations for at least 1 month, but to
devices is that salivary fluoride concentrations date there has been only one trial evaluating this
are significantly increased during the entire day kind of device [52].
without relying on patient compliance [46]. Two Studies in humans and animals attest to the
main types of devices have been currently used – safety of these devices regarding toxicity if a de-
the copolymer membrane (developed in the USA) vice is swallowed [46, 53]. The copolymer device
and the glass beads (developed in the UK) – al- was shown to be clinically effective in reducing
though a third type (which contains a mixture of caries incidence in rats by 63% [49], and in re-
NaF and hydroxyapatite) has recently been intro- ducing dentine sensitivity in humans 4 weeks af-
duced [47]. ter its use in patients under periodontal therapy
The copolymer membrane device was designed [54]. In children, the only randomized controlled
as a membrane-controlled reservoir-type, having trial showed significantly lower caries increments
an inner core of a 50:50 mixture of hydroxyeth- in subjects using fluoride glass beads when com-
yl methacrylate (HEMA)/methyl methacrylate pared to placebo glass beads, both for DMFS (0.84
(MMA) copolymer (which contains NaF), sur- and 2.34, p < 0.05) and dmfs (2.26 and 8.41, p <
rounded by a 30:70 HEMA/MMA copolymer 0.001) [51]. Retention was the main problem as-
membrane (which controls the rate of fluoride re- sociated with the use of the devices.
lease from the device) [48]. It is usually attached The fluoride slow-release glass devices have
to the buccal surface of the first permanent molar a number of important potential applications
by means of stainless steel retainers or bonded to in addition to their use for caries prevention in
the tooth surfaces using adhesive resins. Salivary high-caries-risk groups and a number of current
fluoride levels were shown to remain significantly research studies are investigating their efficacy.
elevated throughout a 100-day test period when A randomized double-blind study investigating
using this type of device [49, 50]. the effect of the slow-release glass devices to pre-
The glass device dissolves slowly when moist vent demineralization around orthodontic brack-
in saliva, releasing fluoride without significantly ets is currently being conducted in Leeds, UK.
affecting the device’s integrity. The original device Preliminary in situ studies have also shown that
was dome shaped [46, 51], and usually attached these devices are beneficial for the prevention of
to the buccal surface of the first permanent molar dentinal root caries, which is an increasing prob-
using adhesive resin. Due to low retention rates, lem for the aged population. The results of these
it was later substantially changed to a kidney- clinical studies are eagerly awaited.
shaped device, and more recently to the form of
a disk that is placed within a plastic bracket, the
latter substantially improving device handling, at- Combinations of Topical Fluoride Modalities
tachment and replacement (without the need for
de-bonding). In contrast to the copolymer mem- Given the nature of the various modalities of fluo-
brane device, the glass type has been shown to ride delivery (community, self-applied and pro-
have a better longevity, releasing fluoride contin- fessional methods), it is not uncommon to ob-
uously for up to 2 years [51]. serve that individuals are frequently exposed to
The hydroxyapatite-Eudragit RS100 diffusion- two or more methods. Based on the mechanism of
controlled fluoride system is the newest type of action of fluoride, it is reasonable to assume that
slow-release device and is intended to release the use of different fluoride vehicles would lead

128 Pessan · Toumba · Buzalaf

Table 8. DMFS (pooled) estimates of treatment effects (as prevented fraction) for direct comparisons between fluo-
ride gels, varnishes, rinses, and toothpastes

TFT types Number of Prevented 95% CI, %

studies fraction, %

Varnish vs. gel 1 14 –12 to 40

Varnish vs. mouth rinse 4 10 –12 to 32
Gel vs. mouth rinse 1 –14 –40 to 12
Toothpaste vs. gel 3 0 –21 to 21
Toothpaste vs. mouth rinse 6 0 –18 to 19
Toothpaste vs. any TFT1 1 –13 to 14
Toothpaste + varnish vs. toothpaste alone 1 48 12 to 84
Toothpaste + gel vs. toothpaste alone 3 14 –9 to 38
Toothpaste + mouth rinse vs. toothpaste alone 5 7 0 to 13
Toothpaste + any TFT vs. toothpaste alone 9 10 2 to 17

TFT = Topical fluoride treatment. Source: Marinho [5].

Gel trials (n = 3) and mouth rinse trials (n = 6), but no varnish trial.

to enhanced preventive effects when compared to use of toothpaste alone or combined with differ-
a single vehicle alone. However, the magnitude of ent fluoride modalities are listed in table 8.
the benefits of topical fluoride treatments used to- Even considering that topical fluorides used in
gether, as well as the possible side effects, has been addition to fluoride toothpaste achieve a modest
recently questioned. reduction in caries compared to toothpaste used
This particular issue was addressed in a system- alone [55], it is still recommended that patients at
atic review assessing the effectiveness of a combi- high caries risk receive additional fluoride therapy.
nation of topical fluoride methods versus a single For patients at low caries risk, however, additional
topical fluoride method. The DMFS pooled pre- fluoride therapy may be not only ineffective, but
vented fraction was only 10% in favor of the com- may also increase the risk of side effects. Therefore,
bined regimens (mouth rinses, gels or varnishes the decision of using additional fluoride therapy
used in combination with toothpaste) when com- must be based in both cost/benefit and risk/benefit
pared to dentifrice alone, but most of the individ- considerations. The recommendations for topical
ual comparisons were not statistically significant fluoride therapy from the American Association
[55]. Those results, however, must be interpreted Council on Scientific Affairs [56], according to the
with caution, as it may be prematurely concluded caries risk, are listed in table 3.
that the association of two modalities of fluoride
delivery is not beneficial. The meta-analysis in-
volved a low number of trials (a small number of Concluding Remarks
trials was included in each relevant comparison)
and, according to the authors, the review has not The caries-protective benefits of the various mo-
tested all combinations of possible practical value. dalities of fluoride therapy have been confirmed
The main results from the comparisons involving by over 60 years of clinical and laboratory studies.

Topical Fluorides for Caries Control 129

There is strong evidence that the use of fluoridat- the most appropriate form of use for fluoride.
ed toothpastes, gels, varnishes and mouth rinses Presently, considering that no modality of topi-
is effective in controlling the progression of cari- cal fluoride administration (mouth rinses, gels,
ous lesions. Also, there is evidence that addition- varnishes and dentifrices) has been proven to
al reductions in dental caries can be achieved by be more substantially effective than any other
combining the use of a fluoride toothpaste with [54], the advantages of dentifrices – regarding
another form of topical fluoride, although the size cost, availability, cosmetic benefits, and impact
of the caries-preventive effect may not be substan- on both caries and periodontal health – indicate
tial [57]. In this case, the combination of different that this should be the fluoride vehicle of choice.
fluoride vehicles may not be justifiable for young Bearing in mind the different fluoride agents and
children, due to the possibility of increasing fluo- products with varying concentrations of fluoride
ride intake, which ultimately might increase the that have been scientifically proven to be effective
risk of dental fluorosis. On the other hand, any against dental caries, it can sometimes be confus-
extra benefit could have a great impact on the oral ing to the patient and challenging to the dental
health of high-caries-risk individuals. clinician to make the most appropriate choice.
The publication of systematic reviews made Professional dental academies and national so-
a considerable contribution to the understand- cieties – e.g. The American Dental Association
ing of the real benefits of fluoride therapy in ( as well as both the American (www.
caries control. Despite the substantial evidence and European Academies of Paediatric
that currently exists, there is still a need for fu- Dentistry ( – produce guidelines
ture randomized controlled trials with adequate to help dental practitioners and the general pub-
protocols to establish the actual benefits of each lic understand the appropriate use of fluoride
mode of fluoride application, so both clinicians products.
and patients can make a better decision regarding

1 Featherstone JDB: Prevention and rever- 5 Marinho VCC: Cochrane reviews of ran- 10 Horowitz HS, Ismail AI: Topical fluo-
sal of dental caries: role of low level fluo- domized trials of fluoride therapies for rides in caries prevention; in Fejerskov
ride. Community Dent Oral Epidemiol preventing dental caries. Eur Arch Pae- O, Ekstrand J, Burt BA (eds): Fluoride in
1999;27:31–40. diatr Dent 2009;10:183–191. Dentistry. Copenhagen, Munksgaard
2 Ellwood RP, Fejerskove O, Cury JA, Cal- 6 Walsh T, Worthington HV, Glenny AM, Textbook, 1996, p 311–327.
rkson B: Fluoride in caries control; in Appelbe P, Marinho VC, Shi X: Fluoride 11 Hawkins R, Locker D, Noble J, Kay EJ:
Fejerskov O, Kidd EAM (eds): Dental toothpastes of different concentrations Prevention. Part 7: professionally applied
Caries: The Disease and Its Clinical for preventing dental caries in children topical fluorides for caries prevention.
Management, ed 2. Oxford, Blackwell & and adolescents. Cochrane Database Syst Br Dent J 2003;195:313–317.
Munksgaard, 2008, p 287–323. Rev 2010;1:CD007868. 12 Ripa LW: Review of the anticaries effec-
3 Marinho VC, Higgins JP, Logan S, 7 Axelsson P: Use of fluorides; in Axelsson tiveness of professionally applied and
Sheiham A: Topical fluoride (tooth- P (ed): Preventive Materials, Methods self-applied topical fluoride gels. J Public
pastes, mouthrinses, gels or varnishes) and Programs. Slovakia, Quintessence Health Dent 1989;49:297–309.
for preventing dental caries in children Books, 2004, p 263–268. 13 van Rijkom HM, Truin GJ, van’t Hof MA:
and adolescents. Cochrane Database Syst 8 Newbrun E: Evolution of professionally A meta-analysis of clinical studies on the
Rev 2003;4:CD002782. applied topical fluoride therapies. Com- caries-inhibiting effect of fluoride gel
4 Marinho VC, Higgins JP, Sheiham A, pend Contin Educ Dent 1999;20(suppl treatment. Caries Res 1998;32:83–92.
Logan S: Fluoride toothpastes for pre- 1):5–9. 14 Marinho VC, Higgins JP, Logan S,
venting dental caries in children and 9 Ripa LW: Professionally (operator) Sheiham A: Fluoride gels for preventing
adolescents. Cochrane Database Syst Rev applied topical fluoride therapy: a cri- dental caries in children and adoles-
2003;1:CD002278. tique. Int Dent J 1981;31:105–120. cents. Cochrane Database Syst Rev 2002;

130 Pessan · Toumba · Buzalaf

15 Poulsen S: Fluoride containing gels, 26 Toumba KJ: Guidelines on the use of 37 Vilhena FV, Olympio KPK, Lauris JRP,
mouthrinses and varnishes: an update of fluoride in children: an EAPD policy Delbem ACB, Buzalaf MAR: Low-
efficacy. Eur Arch Paediatr Dent 2009; document. Eur Arch Paediatr Dent 2009; fluoride acidic dentifrice: a randomized
10:157–161. 10:129–135. clinical trial in a fluoridated area. Caries
16 Jiang H, Tai B, Du M, Peng B: Effect of 27 Stephen KW, Creanor SL, Russell JI, Res 2010;44:478–484.
professional application of APF foam on Burchell CK, Huntington E, Downie CF: 38 Buzalaf MAR, Vilhena FV, Iano FG,
caries reduction in permanent first A 3-year oral health dose-response study Grizzo L, Pessan JP, Sampaio FC,
molars in 6–7-year-old children: of sodium monofluorophosphate denti- Oliveira RC: The effect of different fluo-
24-month clinical trial. J Dent 2005; frices with and without zinc citrate: anti- ride concentrations and pH of denti-
33469–33473. caries results. Community Dent Oral frices on plaque and nail fluoride levels
17 Wei SHY, Chik FF: Fluoride retention Epidemiol 1988;16:321–325. in young children. Caries Res 2009;43:
following topical fluoride foam and gel 28 Richards A, Banting DW: Fluoride tooth- 147–154.
application. Pediatr Dent 1990;12: pastes; in Fejerskov O, Ekstrand J, Burt 39 Alves KM, Pessan JP, Buzalaf MA, Del-
368–374. BA (eds): Fluoride in Dentistry. Copen- bem AC: Short communication: in vitro
18 Whitford GM, Adair SM, Hanes CM, hagen, Munksgaard Textbook, 1996, p evaluation of the abrasiveness of acidic
Perdue EC, Russel CM: Enamel uptake 328–346. dentifrices. Eur Arch Paediatr Dent
and patient exposure to fluoride: com- 29 Twetman S: Caries-prevention with fluo- 2009;10(suppl 1):43–45.
parison of APF gel and foam. Pediatr ride toothpaste in children – an update. 40 Volpe AR, Petrone ME, Davies R,
Dent 1995;17:199–203. Eur Arch Paediatr Dent 2009;10:162– Proskin HM: Clinical anticaries efficacy
19 Marinho VC, Higgins JP, Logan S, 167. of NaF and SMFP dentifrices: overview
Sheiham A: Fluoride varnishes for pre- 30 O’Mullane DM, Kavanagh D, Ellwood and resolution of the scientific contro-
venting dental caries in children and RP, Chesters RK, Schafer F, Huntington versy. J Clin Dent 1995;(6 Spec No):
adolescents. Cochrane Database Syst Rev E, Jones PR: A three-year clinical trial of 1–28.
2002;3:CD002279. a combination of trimetaphosphate and 41 Chaves SC, Vieira-da-Silva LM: Anticar-
20 Ekstrand J, Koch G, Petersson LG: sodium fluoride in silica toothpastes. ies effectiveness of fluoride toothpaste: a
Plasma fluoride concentration and uri- J Dent Res 1997;76:1776–1781. meta-analysis. Rev Saude Publica
nary fluoride excretion in children fol- 31 Nordström A, Birkhed D. Fluoride reten- 2002;36:598–606.
lowing application of the fluoride- tion in proximal plaque and saliva using 42 Stamm JW: Clinical studies of neutral
containing varnish Duraphat. Caries Res two NaF dentifrices containing 5,000 sodium fluoride and sodium monofluo-
1980;14:185–189. and 1,450 ppm F with and without water rophosphate dentifrices; in Bowen WH
21 Pessan JP, Pin MLG, Martinhon CCR, da rinsing. Caries Res 2009;43:64–69. (eds): Relative Efficacy of Sodium Fluo-
Silva SMB, Granjeiro JM, Buzalaf MAR: 32 Lima TJ, Ribeiro CC, Tenuta LMA, Cury ride and Sodium Monofluorophosphate
Analysis of fingernails and urine as bio- JA: Low-fluoride dentifrice and caries as Anti-Caries Agents in Dentifrices.
markers of fluoride exposure from den- lesion control in children with different London, Royal Society of Medicine,
tifrice and varnish in 4- to 7-year-old caries experience: a randomized clinical 1995, pp 43–58.
children. Caries Res 2005;39:363–370. trial. Caries Res 2008;42:46–50. 43 Johnson MF: Comparative efficacy of
22 Marinho VC, Higgins JP, Logan S, 33 Duckworth RM, Morgan SN, Gilbert RJ: NaF and SMFP dentifrices in caries pre-
Sheiham A: Fluoride mouthrinses for Oral fluoride measurements for estima- vention: a meta-analytic overview. Car-
preventing dental caries in children and tion of the anti-caries efficacy of fluoride ies Res 1993;27:328–336.
adolescents. Cochrane Database Syst Rev treatments. J Dent Res 1992;( 71Spec 44 Stookey GK, DePaola PF, Featherstone
2003;3:CD002284. No):836–840. JD, Fejerskov O, Möller IJ, Rotberg S,
23 Yamaguchi N, Saito T, Oho T, Sumi Y, 34 Pessan JP, Alves KMRP, Ramires I, Taga Stephen KW, Wefel JS: A critical review
Yamashita Y, Koga T: Influence of the MFL, Sampaio FC, Whitford GM, of the relative anticaries efficacy of
discontinuation of a school-based, Buzalaf MAR: Effects of regular and low- sodium fluoride and sodium monofluo-
supervised fluoride mouthrinsing pro- fluoride dentifrices on plaque fluoride. rophosphate dentifrices. Caries Res
gramme on the prevalence of dental car- J Dent Res 2010;89:1106–1110. 1993;27:337–360.
ies. Community Dent Health 1997;14: 35 Brighenti FL, Delbem ACB, Buzalaf 45 Kingman A: Methods of projecting long-
258–61. MAR, Oliveira FAL, Ribeiro DB, Sassaki term relative efficacy of products exhib-
24 World Health Organization: Oral Health. KT: In vitro evaluation of acidified iting small short-term efficacy. Caries toothpastes with low fluoride content. Res 1993;27:322–327.
index1.html (accessed 9 December Caries Res 2006;40:239–244. 46 Toumba KJ: Slow-release devices for
2010). 36 Alves KMRP, Pessan JP, Brighenti FL, fluoride delivery to high-risk individu-
25 Pessan JP, Sicca CM, de Souza TS, da Franco KS, Oliveira FAL, Buzalaf MAR, als. Caries Res 2001;35(suppl 1):10–13.
Silva SMB, Whitford GM, Buzalaf MAR: Sassaki KT, Delbem ACB: In vitro evalu- 47 Pessan JP, Al-Ibrahim NS, Buzalaf MA,
Fluoride concentrations in dental plaque ation of the effectiveness of acidic fluo- Toumba KJ: Slow-release fluoride
and saliva after the use of a fluoride den- ride dentifrices. Caries Res 2007;41: devices: a literature review. J Appl Oral
tifrice preceded by a calcium lactate 263–267. Sci 2008;16:238–246.
rinse. Eur J Oral Sci 2006;114:489–493.

Topical Fluorides for Caries Control 131

48 Mirth DB: The use of controlled and 51 Toumba KJ, Curzon MEJ: A clinical trial 56 American Dental Association Council on
sustained release agents in dentistry: a of a slow-releasing fluoride device in Scientific Affairs: Professionally applied
review of applications for the control of children. Caries Res 2005;39:195–200. topical fluoride – evidence-based clini-
dental caries. Pharmacol Ther Dent 52 Altinova YB, Alaçan A, Aydin A, Saniso- cal recommendations. J Am Dent Assoc
1980;5:59–67. glu SY: Evaluation of a new intraoral 2006;137:1151–1159.
49 Mirth DB, Adderly DD, Amsbaugh SM, controlled fluoride release device. Caries 57 Salanti G, Marinho V, Higgins JP: A case
Monell-Torrens E, Li SH, Bowen WH: Res 2005;39:191–194. study of multiple-treatments meta-
Inhibition of experimental dental caries 53 Curzon MEJ, Toumba KJ: In vitro and in analysis demonstrates that covariates
using an intraoral fluoride-releasing vivo assessment of a glass slow fluoride should be considered. J Clin Epidemiol
device. J Am Dent Assoc 1983;107: releasing device: a pilot study. Brit Dent 2009;62:857–864.
55–58. J 2004;196:543–546. 58 Davies RM, Davies GM, Ellwood RP, Kay
50 Mirth DB, Shern RJ, Emilson CG. 54 Marini I, Checchi L, Vecchiet F, Spiazzi EJ: Prevention. Part 4: what advice
Adderly DD, Li SH, Gomez IM: Clinical L: Intraoral fluoride releasing device: a should be given to patients?
evaluation of an intra-oral device for the new clinical therapy for dentine sensitiv- Br Dent J 2003;195:135–141.
controlled release of fluoride. J Am Dent ity. J Periodontol 2000;71:90–95. 59 Hawkins R, Locker D, Noble J, Kay EJ:
Assoc 1982;105:791–797. 55 Marinho VC, Higgins JP, Sheiham A, Prevention. Part 7: professionally applied
Logan S: Combinations of topical fluo- topical fluorides for caries prevention.
ride (toothpastes, mouthrinses, gels, var- Br Dent J 2003;195:313–317.
nishes) versus single topical fluoride for
preventing dental caries in children and
adolescents. Cochrane Database Syst Rev

Prof. Juliano Pelim Pessan

Department of Pediatric Dentistry and Public Health, Araçatuba Dental School, São Paulo State University
Rua José Bonifácio, 1193
16015–050 Araçatuba – SP (Brazil)
Tel. +55 18 3636 3274, E-Mail

132 Pessan · Toumba · Buzalaf

Impact of Fluoride in the Prevention of Caries and Erosion
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 133–145

Systemic Fluoride
Fábio Correia Sampaioa ⭈ Steven Marc Levyb
aDepartment of Clinical and Social Dentistry, Health Science Centre, Federal University of Paraiba, João Pessoa, Brazil; bDepartment

of Preventive and Community Dentistry and College of Public Health, Department of Epidemiology, College of Dentistry, University
of Iowa, Iowa City, Iowa, USA

Abstract frice. Safe and effective doses of fluoride can be achieved

There is substantial evidence that fluoride, through dif- when combining topical and systemic methods.
ferent applications and formulas, works to control caries Copyright © 2011 S. Karger AG, Basel
development. The first observations of fluoride’s effects
on dental caries were linked to fluoride naturally present The idea of ingesting fluoride to prevent and
in the drinking water, and then from controlled water fluo- control dental caries is not new – in fact the ben-
ridation programs. Other systemic methods to deliver flu- efits of ingesting fluoride for this purpose have
oride were later suggested, including dietary fluoride sup- been suggested since the 1870s [1]. At that time,
plements such as salt and milk. These systemic methods fluoride lozenges were recommended in England,
are now being questioned due to the fact that many stud- and later in Germany and Scandinavia. In the
ies have indicated that fluoride’s action relies mainly on its beginning of the 1900s, the product ‘fluoridens’
post-eruptive effect from topical contact with the tooth (calcium fluoride, Cross & Co.) was available in
structure. It is known that even the methods of delivering several European countries and the USA. The
fluoride known as ‘systemic’ act mainly through a topical lozenges were probably not very popular among
effect when they are in contact with the teeth. The effec- dental practitioners, and they were therefore for-
tiveness of water fluoridation in many geographic areas gotten [2, 3]. As a result, the strategy of using flu-
is lower than in previous eras due to the widespread use oride by ingestion to control carious lesions was
of other fluoride modalities. Nevertheless, this evidence also forgotten until the 1930s, when water fluo-
should not be interpreted as an indication that systemic ridation gained status as a reasonable strategy to
methods are no longer relevant ways to deliver fluoride reduce the prevalence of dental caries [4].
on an individual basis or for collective health programs. In the beginning of the 20th century, brown
Caution must be taken to avoid excess ingestion of fluo- stains (the so-called ‘mottled enamel’) in sub-
ride when prescribing dietary fluoride supplements for jects from Colorado Springs (USA) intrigued the
children in order to minimize the risk of dental fluorosis, American dentist Frederick McKay. The obser-
particularly if there are other relevant sources of fluoride vation of similar stains in individuals from other
intake – such as drinking water, salt or milk and/or denti- areas of the USA led him to conclude that some
substance in the water was responsible for the high end of the 1980s, when understanding of its mech-
prevalence of ‘mottled enamel’. After advanced anism of action shifted to the primary topical ef-
chemical analyses of many samples of drinking fect rather than the systemic one [see Buzalaf et al.,
water, the substance was identified as fluoride and this vol., pp. 97–114]. In addition to the acceptance
the term ‘mottled enamel’ became ‘dental fluoro- of a strong systemic effect, the outstanding suc-
sis’, thus relating the dental condition to its caus- cess of these first publications on water fluorida-
ative agent [5, 6]. As a result, at that time, fluoride tion created the belief that the beneficial effects of
was a chemical compound better known for caus- this method would endure for a lifetime, and this
ing stains in the teeth rather than a substance for turned out to not be the case [12].
controlling dental caries. Moreover, the first in- Fluoride remains the cornerstone of modern
vestigations exploring the fluoride/dental fluoro- non-invasive dental caries prevention and man-
sis relationship indicated that excessive ingestion agement, but its mechanism of action remains a
of fluoride from the drinking water could lead to matter of debate [13]. Bibby et al. [14] conducted a
dental fluorosis in a dose-response pattern [5, 7]. study in the 1950s comparing the efficacy of fluo-
Meanwhile, a series of epidemiological investi- ride lozenges intended to be sucked with coated
gations known as the ‘21-city study’ conducted by fluoride pills intended to be swallowed. The re-
Trendley Dean concluded that water with 1 mg/l sults of this trial demonstrated that the group us-
(which is equivalent to 1 ppm) of natural fluo- ing lozenges had fewer carious lesions compared
ride could provide protection against dental car- with the group that used the coated pills. This
ies with minimal prevalence of dental fluorosis, study provided clear evidence that the mechanism
and with almost all of it mild (only opacity, not of action of fluoride is mostly post-eruptive [11,
staining). This impressive observation eventually 13]. Several other studies confirmed the scientific
trigged the water fluoridation trials in the USA evidence favoring the primacy of fluoride’s post-
and officially introduced systemic fluoride meth- eruptive over pre-eruptive effects in cariostasis
ods into dentistry [6, 8, 9]. [12]. These observations have led to a rethink-
The first experiences with water fluoridation in ing of the rationale for using a systemic method
Grand Rapids in 1945, and later in Newburgh and of delivering fluoride on an individual or collec-
Kingston, provided relevant results that clearly in- tive basis. Even though systemic fluoride meth-
dicated a caries reduction of about 60% in chil- ods were originally designed to promote caries
dren living in the cities with fluoride in the drink- protection by ingestion, anti-caries benefits are
ing water versus those without fluoridated water delivered primarily through topical effects due to
[3, 10]. Since the ingestion of excessive fluoride in the direct contact of fluoride on the tooth surface
water could lead to dental fluorosis, then came the prior to ingestion. The beneficial effect can also be
question: what would be the effect of a lower in- explained by the fluoride ingested that returns to
gestion of fluoride? It was suggested that lowering the oral cavity when incorporated into the saliva
the amount of fluoride ingested would provide [see Buzalaf et al., this vol., pp. 97–114]. Hence, in
caries protection without producing objection- order to obtain the maximum benefits of fluoride
able dental fluorosis. The concept that fluoride’s from systemic methods, continued exposure has
preventive action comes from being ingested and to occur. Interruption of systemic fluoride meth-
incorporated into tooth mineral became the obvi- ods can jeopardize the beneficial effects of fluo-
ous mechanism of action [3, 6, 11]. ride in reducing caries incidence [15].
All these historical facts are important for un- The salts most often used for systemic fluorida-
derstanding the great popularity of systemic meth- tion are presented in table 1. The choice and for-
ods of delivering fluoride from the 1950s until the mulas depend on the dose, stability and practicality

134 Sampaio · Levy

Table 1. Fluoride compounds and concentrations usually used in different systemic methods of fluoridation

Fluoridation methods Fluoride compounds Fluoride concentrations

Water fluoridation hydrofluorosilicate (FSA), sodium fluorosilicate, 0.7–1.2 mg/l

sodium fluoride

Salt fluoridation potassium fluoride, sodium fluoride 250–300 mg/kg

Milk fluoridation sodium fluoride or disodium 5 mg/l


Dietary fluoride supplements sodium fluoride, acidulated phosphate fluoride, 0.25–1.0 mg/day
potassium fluoride, calcium fluoride

[16]. The best-known systemic fluoride method is Water fluoridation is a rather simple technique
water fluoridation. However, it is also important that consists of adding a controlled amount of flu-
to recognize that other ways of delivering fluoride oride to the water supply in concentrations rang-
– such as salt fluoridation, milk fluoridation and ing from 0.7 to 1.2 mg/l (equivalent to 0.7–1.2
fluoride-containing supplements – still have their ppm) depending on the local average tempera-
importance in dentistry. This is particularly rele- ture. In warm climate regions, the recommended
vant in developing countries where access to oral concentrations of fluoride are low (0.7 mg/l) due
health care can be very limited. to higher consumption of water, whereas in more
temperate climate regions the concentration can
be higher since water consumption is lower [16,
Water Fluoridation 23, 24]. Recently, the Department of Health and
Human Services of the United States proposed
Recent publications have suggested that over 300 a new standardized level of 0.7 ppm fluoride
million people in almost 40 countries are exposed throughout the country as an appropriate level for
to fluoride from adjusted fluoridated water sup- maximizing benefits while minimizing any risks
plies [17, 18]. In the USA alone, an estimated associated with excess ingestion [25].
195 million people (approximately 72.4% of the There is no central source that collects and
population) are currently receiving the benefits of updates information on costs of water fluorida-
optimally fluoridated water [19, 20]. Considering tion programs all over the world. Nevertheless,
these estimates, more than half of the popula- the average annual cost of this method per per-
tion of the world receiving the benefits of water son has been estimated to range from USD 0.10
fluoridation live in the USA. A national health up to USD 5.41, which makes fluoridation a very
promotion and disease prevention initiative in the cost-effective measure for reducing dental caries
USA, known as ‘Healthy People 2010’, included [16, 26]. Water fluoridation can be regarded as a
an objective to increase the proportion of the US low-cost method to deliver fluoride, particularly
population served by community water systems for those communities where oral health care and
with optimally fluoridated water to 75% [21]. For particularly fluoride dentifrices are not available
2020, it is expected that nearly 80% of Americans and/or not affordable. Variables that influence
will be served by community water fluoridation the costs per capita of a fluoridation project in-
[22]. clude: (1) the size of the community (the smaller

Systemic Fluoride 135

the community, the higher the per capita cost); due to political or legal reasons [15]. Another im-
(2) the prevalence of dental caries in the popula- portant issue related to water fluoridation is the in-
tion; (3) the number of water sources; (4) the type creased consumption of bottled water over the last
of equipment; (5) the fluoride compound; (6) the two decades, even in countries where tap water is
availability of technical support. considered safe and of excellent quality. The poten-
The most common fluoride compounds used tial causes for this behavior are dissatisfaction with
are hexafluorosilicic acid (H2SiF6), also known as tap water organoleptics (especially taste) and gen-
fluorosilicic acid, which comes in a liquid form, eral health risk concerns [35]. Nevertheless, water
and disodium hexafluorosilicate (Na2SiF6), also fluoridation is still an important way to deliver flu-
known as sodium fluorosilicate (a powder). It oride to control caries [10, 18, 33].
must be emphasized that these compounds are not The multiple sources of fluoride and expan-
from industrial waste. Fluorosilicic acid is more sion of fluoride therapies have created a complex
frequently used. When it is introduced into the scenario for evaluating total fluoride ingestion
water system, it dissociates to release fluoride ions and isolating the beneficial effects of water fluori-
into the water. This process is similar to what hap- dation. Hence, the magnitude of benefit from wa-
pens to the fluoride ion when it is naturally present ter fluoridation is no longer 50–70%, as found in
in the water supplies. Whitford et al. [27] observed early studies when caries was prevalent in many
that the major features of human fluoride metabo- parts of the world and water fluoridation was the
lism are not affected by different chemical com- main source of fluoride [28, 36, 37]. This benefi-
pounds commonly used to fluoridate water, or cial effect is certainly lower than in previous years
whether the fluoride is present naturally or added and probably it is also confounded by other meth-
artificially. Hence, there is no difference chemical- ods of fluoride delivery, especially the topical ones
ly between natural and artificial fluoridation [28]. [13]. Exposure of individuals to many sources of
The US Centers for Disease Control and fluoride has raised concern about the potential
Prevention recognized fluoridation as one of the to increase the prevalence of dental fluorosis [see
major public health measures of the 20th cen- Buzalaf and Levy, this vol., pp. 1–19]. It must be
tury [29]. Several other organizations – such pointed out that the majority of the dental fluoro-
as the World Health Organization [30] and the sis cases related to water fluoridation are at very
American Dental Association [31] – have recog- mild or mild levels, and are often not even noticed
nized the effectiveness of water fluoridation in re- by those who are affected. Thus, the prevalence of
ducing the prevalence of dental caries. In spite of aesthetically objectionable dental fluorosis due to
this, water fluoridation is frequently questioned water fluoridation is low, and does not represent
by anti-fluoridationists who cite freedom-of- a health issue [15, 38–40].
choice issues or the potential dangers to humans Several studies have compared caries preva-
from fluoride [32]. However, there is no evidence lence in towns with and without water fluori-
of harmful effects of fluoride related to optimal dation programs or after fluoridation cessation.
water fluoridation, with the exception of the po- Some studies reported that caries prevalence re-
tential to increase the prevalence of dental fluoro- mained almost the same after water fluoridation
sis [24, 29, 33, 34]. cessation [12, 13, 41]. More aggressive use of seal-
Although water fluoridation has been demon- ants and other fluorides may party explain the re-
strated to be effective and safe, anti-fluoridationist sults in some regions of the world. However, these
challenges persist in the USA and elsewhere. Some findings need to be taken with caution, since there
European countries were not successful in estab- is still a noticeable effect of water fluoridation in
lishing or maintaining water fluoridation programs reducing carious lesions incidence in other parts

136 Sampaio · Levy

of the world [15]. Recent data from the National
Brazilian Epidemiologic Survey showed a 30– 0.900
40% lower decayed (D) component of the DMFT
in 12-year-old children from fluoridated towns 0.850

where water fluoridation had been implement-

ed for more than 5 years [42]. It is interesting to

note that high-concentration fluoride dentifric-
es (1,500 ppm) have been used widely since the
1990s in both fluoridated and non-fluoridated
towns in the country. In preliminary analyses of
2010 data from several communities in Brazil, 0.650
most of them capitals of their states, a water fluo-
1 2
ridation preventive effect of about 20–30% is still Fluoridation
present when comparing fluoridated with non-
fluoridated towns (raw data from the National Fig. 1. Human development index (HDI) in Brazilian
Brazilian Epidemiologic Survey, 2010). These re- States where water fluoridation does not reach 40% of
sults are similar to those observed in the USA in the territory (1) and in those where water fluoridation is
the mid-1980s [43], and in line with data obtained available in more than 60% of the cities (2).
recently in Australia [39]. In this later study, a
3-year follow-up of caries status was carried out
in adolescents. The effectiveness of water fluori-
dation was observed even in the presence of the the access to potable water. These regions gener-
effect of fluoride from other methods. ally have certain similarities: technical capabilities
Several studies have shown that water fluori- are limited and political support for water fluori-
dation reduced the prevalence of caries and the dation is often less favorable.
number of affected teeth, as well as social in- Human development and potable water avail-
equalities, among groups with a different socio- ability has been subject of study of many inves-
economic status [44–50]. However, the features tigations [55]. Evaluation of the impact of water
of social complexity of the populations in each accessibility can provide relevant data on wa-
country complicate these comparisons. For in- ter quality and health parameters as well. The
stance, in New Zealand and Finland, water fluo- Human Development Index (HDI) has been suc-
ridation had a similar beneficial effect for all so- cessfully used as suitable ranking of development
cial classes, whereas in some other countries this of countries and standard of living in target re-
was not the case [51–53]. In a Brazilian study gions within countries. This index is a composite
carried out in the wealthiest part of the country, statistic reference number (from zero = lowest, to
communities with better social status had lower one = highest level) which is calculated based on
caries experience, probably due to the mediating life expectancy, education and standard of living
effect of receiving fluoridation earlier than in oth- [56]. The relationship of water fluoridation to so-
er areas [49]. In general, there is a consensus that cial inequalities can be evaluated by the HDI. This
water fluoridation can be most advantageous for is particularly important for developing coun-
more deprived communities where other health tries where water fluoridation is feasible. Figure
policies are less available [10, 54]. On the other 1 shows the HDI in Brazilian States where water
hand, one must bear in mind that in underprivi- fluoridation does not reach 40% of the territory
leged communities there are limitations even in and in those where water fluoridation is available

Systemic Fluoride 137

in more than 60% of the cities. There is a clear in- prevention measures, and particularly fluoride
dication that water fluoridation reaches the most toothpastes, are not available [59].
privileged groups (HDI >0.80) more efficiently The first epidemiological studies to evaluate
than it does those who are in greater need of this the effectiveness of fluoridated salt in reducing
health benefit (HDI <0.75). Since potable water caries prevalence were performed in Colombia,
is a basic unmet need in many underprivileged Hungary and Switzerland [58]. The outcomes of
populations in developing countries, one poten- these studies indicated that salt fluoridation gen-
tial strategy for expanding water fluoridation in erally showed very similar beneficial results to
such areas might be linking fluoridation projects those observed for water fluoridation [60].
with the need for potable water – though imple- Recent statistics indicate that salt fluoridation
mentation of this plan may face limitations in cer- is available in nearly all Latin American coun-
tain geographical areas. tries, except Brazil, Chile and Panama. It is still
In summary, water fluoridation remains the available in several European countries, including
most cost-effective way to deliver fluoride for France, Germany and Switzerland. There are na-
prevention and control of caries at the commu- tional regulations or authorizations for the pro-
nity level. Opponents of community water fluo- duction and marketing of fluoridated salt in eight
ridation have been overstating adverse health European countries: Austria, Czech Republic,
effects, including concerns with aesthetically ob- France, Germany, Romania, Slovakia, Spain and
jectable dental fluorosis, without scientific basis. Switzerland [58, 59, 61–63].
Community water fluoridation can be the most The costs for implementing salt fluoridation
important and sometimes the only feasible oral are similar to those for water fluoridation regard-
health program for some underprivileged groups. ing the equipment for initial operation. However,
Although the currently measured percentage level during operation, salt fluoridation has an estimat-
of effectiveness of water fluoridation in many ar- ed cost 10–100 times lower than that associated
eas is lower than in previous eras due to the more with water fluoridation programs. According to
widespread use of other fluoride modalities, its Gillespie and Marthaler [64], the costs of salt flu-
importance as a general health measure must not oridation can vary from USD 0.015 up to USD
be underestimated. 0.030 per capita/year, which is so low that many
producers do not raise the price of the product
after fluoridation is implemented.
Salt Fluoridation In contrast with water fluoridation, which is
readily available to the whole community, salt flu-
Salt fluoridation (at a concentration of about 250– oridation can provide a choice for the consum-
350 mg/kg) can be considered as an alternative to er. According to Jones et al. [65], the individual
fluoridation of drinking water. It was introduced choice is one positive aspect of a fluoridated salt
in Switzerland in the 1950s based on the success of program, since it can be sold alongside a non-
the use of iodized salt to prevent goiter [3, 57, 58]. fluoridated alternative. Individual choice makes
As mentioned in the introduction to this chapter, salt fluoridation more acceptable for some people
the objective of any fluoridation method in the from ethical and social policy perspectives. On the
1950s was to promote the ingestion of fluoride in other hand, it can weaken its caries-preventive im-
order to achieve its cariostatic effect. Hence, the pact since salt is not used similarly on an individ-
concept of using salt fluoridation in a community ual basis [16]. Another aspect to consider is that
has a different aim today, which is to reach com- many variants of the commercial distribution or
munities and regions in the world where oral care ‘channels’ to reach the consumer may exist. These

138 Sampaio · Levy

channels include: domestic salt, meals at schools, fluoridation use urine as a biomarker [see Rugg-
large kitchens, and in food items such as bread. In Gunn et al., this vol., pp. 37–51]. There are several
Switzerland and other Latin American countries, studies that show that ingestion of fluoridated salt
all commercial channels are utilized. In some oth- can increase fluoride excretion, and consequently
er European countries (France and Germany), the this can be a useful way to monitor compliance
salt fluoridation program is mainly based on do- of individuals with a salt fluoridation program, as
mestic salt [62, 65]. well as a good alternative to monitor possible ex-
One point of concern is that promoting salt cessive fluoride ingestion [58, 68].
fluoridation could be contraindicated from the One interesting experience using salt fluori-
perspective of general public health, since great- dation was noted in Jamaica, where a salt fluo-
er salt consumption is linked to hypertension. ridation program started in 1987. At that time,
However, according to Jones et al. [65] people do the mean DMFT of 12-year-old children was 6.7
not need to change their usual behavior to ben- (very severe) and recently the DMFT was about
efit, and if a secular decline in salt consumption 1.1 (low) [69]. The salt fluoridation program was
were to take place, an increase in fluoride concen- considered appropriate for the island due to geo-
tration could be considered. To support this view, graphical conditions, the low concentrations of
Bürgi and Zimmerman [57] expressed the opin- water-borne fluoride (which do not exceed 0.3
ion that preventing hypertension through restrict- mg/l), and the availability of bottled water also
ing salt intake and eliminating iodine deficiency having the same levels of fluoride. A recent study
through iodized salt are not in conflict. It is esti- observed that 96% of rural and 100% of urban
mated that among communities or groups usually Jamaican children in the sample were consuming
consuming low-salt diets (<5 g NaCl per person fluoridated salt [59].
per day), essential hypertension will be uncom- Similar to issues raised with regard to fluori-
mon. Moreover, there is no doubt that some salt dated water, there has been some concern about
is required by man, and estimates of normal daily the simultaneous combination of fluoride ingest-
requirements for adults have ranged up to 15 g ed from both dentifrice and salt. Available data
per day [66]. suggest that this combination has not resulted
The fluoride compounds used are usually so- in objectionable enamel fluorosis levels [70, 71].
dium fluoride and potassium fluoride (table 1), However, increased mild dental fluorosis was ob-
which are included in the salt during manufacture served in children who used fluoride tablets in as-
of the product. A wide range of concentrations of sociation with fluoridated salt [37].
fluoride have been tested with concentrations Although salt fluoridation has received sup-
varying from 90 mg/kg up to 350 mg/kg. Most port from official health agencies such as the
programs have used 250 or 350 mg/kg, and some World Health Organization, regular ongoing sur-
studies have suggested that the ideal concentra- veillance of fluoride concentrations in the salt is
tion of fluoride in the salt should be about 250 necessary [59]. The concentration of 250–300 mg/
mg/kg [58, 60]. This level of fluoride was support- kg of fluoride in salt is regarded as the ideal con-
ed by a study that measured salivary fluoride after centration, while the concentration of 200 mg/kg
a meal prepared with fluoridated salt at 250 mg/ of fluoride is regarded as the minimal acceptable
kg. The results of this trial indicated that, at this level of fluoride in salt to achieve a meaningful ef-
concentration, the level of fluoride in saliva was fect on caries control. Salt fluoridation should be
very similar to that found in the saliva of individ- considered when water fluoridation is technically
uals exposed to water fluoridation at 1 mg/l [67]. difficult or due to economic or sociocultural rea-
Most of the studies designed for monitoring salt sons it cannot be implemented. In summary, the

Systemic Fluoride 139

advantages of using salt as a vehicle for delivering with an impressive average of 183.9 liters, fol-
fluoride outweigh the drawbacks related to this lowed by Sweden with 145.5 liters. On the other
method, such as variation in ingestion, difficulties hand, in China only 8.8 liters per capita were con-
in maintaining the ideal concentration and con- sumed [76]. It is predicted that the worldwide av-
cerns with hypertension. erage consumption of milk could reach 90 kg per
person/year in the year of 2030 [74].
Sodium fluoride or disodium monofluoro-
Milk Fluoridation phosphate are the fluoridating compounds includ-
ed in milk (table 1). The manufacture of fluori-
The idea of introducing fluoride into milk was first dated milk involves simple production techniques
published in 1953 [72], and the first outcomes of a regardless of its various forms (pasteurized, steril-
clinical trial with milk fluoridation were available ized, UHT or powdered) [74, 77]. All the products
in 1959 [73, 74]. Milk fluoridation became more have been shown to be stable, with considerable
popular only decades later when a charitable foun- amounts of fluoride remaining throughout their
dation started to promote this method [75]. The shelf-life. The rationale for ingestion of fluoridat-
first community milk fluoridation program was ed milk is that it increases the concentration of flu-
implemented in Bulgaria in the cities of Plovdiv oride in saliva to levels similar to those observed
and Asenovgrad in 1988, reaching 15,000 chil- for optimally fluoridated water. Considering the
dren [74]. In the 1990s, milk fluoridation projects amounts of water and milk usually consumed, in
were implemented in Russia, China, Chile, Peru terms of caries prevention the fluoride concentra-
and the UK. However, Peru later ceased milk flu- tion equivalent to 1 mg/l of fluoride in water is 5
oridation due to the introduction of salt fluori- mg/l of fluoride in milk [74, 78].
dation. In 2000, a milk fluoridation program was The main constituents of whole cow’s milk,
introduced in Thailand, and recently the Republic other than carbohydrate (4.5%), considered rel-
of Macedonia started milk fluoridation [75]. evant for the de- and remineralization processes
It is estimated that about 800,000 children are are fat (up to 3.9%), protein (3%), phosphorus
receiving fluoridated milk [75]. Most data avail- (92 mg/100 g) and calcium (118 mg/100 g). The
able for this method are from studies with children, amount of carbohydrate present in milk would be
since the child population is the target age group sufficient to classify this food item as cariogenic.
with school-based programs. Milk consumption Nevertheless, lactose is regarded as the least cario-
varies considerably when comparing different re- genic of the common dietary sugars [77]. Studies
gions of the world. For instance, the worldwide focusing on the positive aspects of having milk
average consumption of milk was estimated to be as a vehicle for fluoride delivery have indicated
78 kg per person/year, but consumption was high- that milk would appear to reduce the cariogenic
est in industrialized countries (212 kg per person/ potential of dental plaque due to: (1) lactose be-
year) when compared with developing countries ing the least cariogenic of dietary sugars; (2) the
(45 kg per person/year) [74]. Latin America has protective role of casein, and possibly fats; (3) the
the highest estimates among developing coun- protective role of calcium and phosphorus in the
tries with 110 kg per person/year. There has been de- and remineralization processes. However, the
a modest increase in milk consumption in most favorable features of milk can be strongly compro-
industrialized countries over the last 30 years, but mised when sucrose is added [79]. In fact, cow’s
in Western European countries a decline was ob- milk is essentially non-cariogenic, but the addi-
served during the same period. In 2006, Finland tion of sucrose in the milk can promote early car-
had the highest consumption per capita of milk, ies in young children [80].

140 Sampaio · Levy

A systematic review on fluoridated milk con- compounds used as supplements are shown in
cluded that high-quality randomized clinical trials table 1. The most common type of fluoride used
concerning the effectiveness of fluoridated milk is sodium fluoride.
for caries prevention are lacking. This gap com- Several studies support the view that the ma-
promises the evidence for supporting fluoridated jor beneficial cariostatic effect of fluoride supple-
milk as an effective method for caries control [81]. ments is due to its post-eruptive effect [12, 81, 83].
However, from the few available studies, it can be The pre-eruptive effect of dietary fluoride sup-
concluded that children should begin to drink flu- plements has been questioned and received little
oridated milk at an early age, preferably before 4 credit. Moreover, the pre-eruptive effect and the
years, in order to reduce caries in their primary early use of supplements have been linked to den-
teeth. In addition, children should be drinking tal fluorosis in children. As a result, in some coun-
fluoridated milk when their first permanent mo- tries (e.g. Canada), dietary fluoride supplements
lars erupt in order to help protect these teeth. are not indicated for most citizens and particular-
ly for those before the third year of age, because of
the potential risk for dental fluorosis [16, 83].
Dietary Fluoride Supplements The balance between the beneficial effect on
caries prevention and the potential increased
The use of dietary fluoride supplements for con- prevalence of dental fluorosis is the key point in
trolling dental caries was introduced by the end considering the use of fluoride supplements. This
of the 1940s when it was assumed that the ‘sys- balance between benefits and risks can be influ-
temic’ effect of fluoride was its main mechanism enced by the child’s age and caries risk status. It is
of action. It is important to note that the dietary also important to point out the fluoride concen-
fluoride supplements were introduced before the tration of the child’s primary sources of drinking
widespread use of fluoride dentifrices, varnishes, water as an important variable [84].
gels and other professional methods for apply- In spite of the potential risk for dental fluoro-
ing fluoride. Thus, when dietary fluoride supple- sis, dietary fluoride supplements are regarded as
ments were introduced, the major source of fluo- effective in preventing caries and are still available
ride was the drinking water [16]. Since that time, in several countries [85]. In addition, the current
dietary fluoride supplements have been intended evidence indicates that the incidence of carious
to substitute for fluoridated water in areas where lesions can be reduced in both the primary and
water fluoridation is not available or feasible. permanent dentition by the regular use of dietary
In the beginning of the 1950s, the American fluoride supplements [86]. However, a recent sys-
Dental Association recommended supplement- tematic review indicated that the effectiveness
ing the domestic drinking water with of 0.25 mg/l of fluoride supplements is weak for the primary
fluoride for children up to 6 years of age. The rec- dentition [83]. Due to the fact that dietary fluo-
ommendation was to mix fluoride with the water ride supplements can increase children’s intake
in a bottle that would be stored in the refrigerator. of fluoride, this method is recommended only
The children were asked to drink this fluoridated for groups of or individual children at high car-
water and eat food prepared with it [82]. ies risk. These recommendations emphasize that,
The term dietary fluoride supplements can for an appropriate prescription of dietary fluo-
mean different forms of manufactured products: ride supplements, there is a need for caries risk
(1) tablets or drops (to be swallowed); (2) tablets assessment in association with some estimate of
for chewing; (3) lozenges (to be sucked or dis- the total fluoride intake. In summary, the balance
solved in the mouth) [16]. The different fluoride between the caries-preventive benefits of dietary

Systemic Fluoride 141

Table 2. Fluoride supplement dosage schedule for 2010 (mg/day)

Fluoride ion level in drinking water

<0.3 mg/l 0.3–0.6 mg/l >0.6 mg/l

Birth to 6 months none none none
6 months to 3 years 0.25 none none
3–6 years 0.50 0.25 none
6–16 years 1.0 mg 0.50 none

Approved by the American Dental Association, American Academy of Pediatrics and American
Academy of Pediatric Dentistry.

fluoride supplementation and the risk of dental not receiving other systemic or topical fluoride
fluorosis has to be evaluated for the appropriate modalities. Because dietary fluoride supplements
implementation of this method at both group and seem to be more effective in preventing caries in
individual levels. the permanent dentition, and that their use before
In addition to child’s age and caries risk status, the age of 6 years (but especially before the age of
the sources of fluoride intake such as drinking wa- 3 years) is associated with an increased prevalence
ter have to be considered when prescribing dietary of dental fluorosis, they should be prescribed only
fluoride supplements. Hence, it is recommended for high-caries-risk groups and individuals from
that the fluoride content in the water should be the age of 3 years on. In addition, they can be well
determined, whether a public water source or bot- suited to some remote populations not receiving
tled water is the primary source of drinking water. other methods of fluoride delivery and in areas
There is evidence that the use of both fluoridated where this method is accepted and compliance is
salt and fluoride tablets is more effective in reduc- achieved [88].
ing caries in children when compared with the use
of fluoridated salt alone [37]. However, similarly
to what happens in water fluoridation, the fluo- Conclusion
ride intake from salt and tablets simultaneously
can increase the occurrence of mild fluorosis in Concerning the systemic effect of fluoride, most
permanent incisors [37]. studies support the view that the caries-preventive
Since 1994, the American Dental Association effect of fluoride is mainly post-eruptive. This evi-
has recommended that fluoride supplements dence should not be interpreted as a limitation of
should not be given from birth up to 6 months systemic fluoride methods, since some topical ef-
(table 2) [87]. This is an interesting precaution fect cannot be disregarded when someone is in-
for avoiding significantly increased fluoride in- gesting fluoride in water, milk or salt or sucking a
take during the early stages of tooth formation, fluoride lozenge. However, since multiple sourc-
and thus reducing the risk of dental fluorosis. es of fluoride exposure and ingestion may exist, a
Finally, there is no doubt that dietary fluoride well coordinated approach for promoting fluoride
supplements can play a role in caries control, par- delivery is essential, particularly if the individual
ticularly for children at elevated risk of caries and can be exposed to different systemic and topical

142 Sampaio · Levy

methods simultaneously. In fact, the classification fluorosis when they are used before 3 years of
of fluoride delivery methods into ‘systemic’ and age. Moreover, evaluation of additional sourc-
‘topical’ is misleading, since it is recognized that es of fluoride is important when using fluoride
the topical effect is more important. A more ratio- supplements.
nal classification would be as follows: individual To date, there is strong evidence that water
methods (fluoride dentifrices, mouth rinses and fluoridation is still an effective method of caries
supplements), collective methods (fluoridated prevention. However, technical requirements for
water, salt and milk) and professional methods implementing water fluoridation may pose limi-
(fluoride varnishes, gels, solutions and foams). tations for some developing countries, resulting
This approach can simplify the consideration of in water fluoridation not being available where it
benefits and risks when combining different fluo- is most needed.
ride methods, particularly individual and collec- Finally, it must be pointed out that fluoride
tive ones. has been used to prevent caries for more than
There is strong evidence that fluoride under 60 years, and several clinical and laboratory
different applications and formulas works for studies have demonstrated that fluoride used
controlling caries development. Nevertheless, in both topical and systemic ways has the abil-
there is a lack of high-quality clinical trials con- ity to control caries. Thus, fluoride is still of the
cerning the caries-preventive effect of fluoridated utmost importance for preventing dental caries,
milk and salt. as well as for modern non-invasive dental caries
Regarding dietary fluoride supplements, there management.
is strong evidence for an increased risk of dental

1 Hunstadbraten K: Fluoride in caries pro- 9 Horowitz HS: The effectiveness of com- 15 Newbrun E: What we know and do not
phylaxis at the turn of the century. Bull munity water fluoridation in the United know about fluoride. J Public Health
Hist Dent 1982;30:117–120. States. J Public Health Dent Dent 2010;70:227–233.
2 Hoffmann-Axthelm W: History of Den- 1996;56:253–258. 16 Fejerskov O, Ekstrand J, Burt B: Fluoride
tistry. Chicago, Quintessence, 1981, pp 10 Centers for Disease Control and Preven- in Dentistry, ed 2. Copenhagen, Munks-
318–322. tion: Achievements in Public gaard, 1996.
3 Murray JJ, Rugg-Gunn AJ, Jenkins GN: Health,1990–1999: fluoridation of 17 Browne D, Whelton H, O’Mullane D:
Fluorides in Caries Prevention, ed 3. drinking water to prevent dental caries. Fluoride metabolism and fluorosis.
Oxford, Butterworth-Heinemann, 1991. Morb Mortal Wkly Rep 1999;48:933. J Dent 2005;33:177–186.
4 Maier FJ: Manual of Water Fluoridation 11 Fejerskov O: Changing paradigms in 18 Pizzo G, Piscopo M, Pizzo I, Giuliana G:
Practice. New York, McGraw-Hill, 1963. concepts on dental caries: consequences Community water fluoridation and car-
5 Dean HT: Endemic fluorosis and its rela- for oral health care. Caries Res ies prevention: a critical review. Clin
tion to dental caries. Public Health Rep 2004;38:182–191. Oral Invest 2007;11:189–193.
1938;53:1443–1452. 12 Limeback H: A re-examination of the 19 Bailey W, Barker L, Duchon K, Maas W:
6 McClure FJ: Water Fluoridation: The pre-eruptive and post-eruptive mecha- Populations receiving optimally fluori-
Search and the Victory. Bethesda, nism of the anti-caries effects of fluo- dated public drinking water – United
National Institutes of Health, 1970. ride: is there any anti-caries benefit from States, 1992–1996. MMWR
7 McKay FS: The relation of mottled swallowing fluoride? Community Dent 2009;57:737–741.
enamel to caries. J Am Dent Assoc Oral Epidemiol. 1999;27:62–71. 20 Center for Disease Control and Preven-
1933;20:1137–1149. 13 Hellwig E, Lennon AM: Systemic versus tion: 2008 Water Fluoridation Statistics.
8 Horowitz HS: Effectiveness of School topical fluoride. Caries Res 2004;38: Atlanta, CDC, 2010.
Water Fluoridation and Dietary Fluoride 258–262. dation/statistics/2008stats.htm.
Supplements in School-aged Children 14 Bibby BG, Wilkins E, Witol E: A prelimi- 21 Center for Disease Control and Preven-
J Publ Hlth Dent 1989;49(5 Spec No): nary study of the effects of fluoride loz- tion: About Healthy People. Atlanta,
290–296. enges and pills on dental caries. Oral Surg CDC, 2010.
Oral Med Oral Pathol. 1955;8:213–216. people/hp2010.htm.

Systemic Fluoride 143

22 Healthy People: Oral Health. www.healt- 35 Doria MF: Bottled water versus tap 47 Jones CM, Worthington H: Water fluori- water: understanding consumers’ prefer- dation, poverty and tooth decay in
tives2020/overview.aspx?topicid=32. ences. J Water Health 2006;4:271–276. 12-year-old children. J Dent 2000;28:
23 Murray JJ: Efficacy of preventive agents 36 Levy SM: An update on fluorides and 389–393.
for dental caries. Systemic fluorides: fluorosis. J Can Dent Assoc 2003;69: 48 Burt BA: Fluoridation and social equity.
water fluoridation. Caries Res 1993; 286–291. J Public Health Dent 2002;62:195–200.
27(suppl 1):2–8. 37 Meyer-Lueckel H, Grundmann E, Stang 49 Gabardo MC, da Silva WJ, Moysés ST,
24 Whitford GM: The Metabolism and Tox- A: Effects of fluoride tablets on caries Moysés SJ: Water fluoridation as a
icology of Fluoride, ed 2. Basel, Karger, and fluorosis occurrence among 6- to marker for sociodental inequalities.
1996. 9-year olds using fluoridated salt. Com- Community Dent Oral Epidemiol 2008;
25 Department of Health and Human Ser- munity Dent Oral Epidemiol 2010;38: 36:103–107.
vices: Proposed HHS recommendation 315–323. 50 Fischer TK, Peres KG, Kupek E, Peres
for fluoride concentration in drinking 38 Griffin SO, Beltrán ED, Lockwood SA, MA: Primary dental care indicators:
water for prevention of dental caries. Barker LK: Esthetically objectionable association with socioeconomic status,
Federal Register Notices 2011;76: fluorosis attributable to water fluorida- dental care, water fluoridation and Fam-
2383–2388. tion. Community Dent Oral Epidemiol ily Health Program in Southern Brazil
26 Ciketic S, Hayatbakhsh MR, Doran CM: 2002;30:199–209. (in Portuguese). Rev Bras Epidemiol
Drinking water fluoridation in South 39 Spencer AJ, Armfield JM, Slade GD: 2010;13:126–138.
East Queensland: a cost-effectiveness Exposure to water fluoridation and car- 51 Hausen H, Milen A, Heinonen OP,
evaluation. Health Promot J Austr ies increment. Community Dent Health Paunio I: Caries in primary dentition
2010;21:51–56. 2008;25:12–22. and social class in high and low fluoride
27 Whitford GM, Sampaio FC, Pinto CS, 40 Chankanka O, Levy SM, Warren JJ, Chal- areas. Community Dent Oral Epidemiol
Maria AG, Cardoso VES, Buzalaf MAR: mers JM: A literature review of aesthetic 1982;10:33–36.
Pharmacokinetics of ingested fluoride: perceptions of dental fluorosis and rela- 52 Evans RW, Beck DJ, Brown RH, Silva PA:
lack of effect of chemical compound. tionships with psychosocial aspects/oral Relationship between fluoridation and
Arch Oral Biol 2008;53:1037–1041. health-related quality of life. Commu- socioeconomic status on dental caries
28 Lennon MA: One in a million: the first nity Dent Oral Epidemiol 2010;38: experience in 5-year-old New Zealand
community trial of water fluoridation. 97–109. children. Community Dent Oral Epide-
Bull World Health Organ 2006;84: 41 Maupomé G, Shulman JD, Clark DC, miol 1984;12:5–9.
759–760. Levy SM, Berkowitz J: Tooth-surface 53 Spencer AJ, Slade GD, Davies M: Water
29 Center for Disease Control and Preven- progression and reversal changes in flu- fluoridation in Australia. Community
tion: Recommendations for using fluo- oridated and no-longer-fluoridated com- Dent Health 1996;13(suppl 2):27–37.
ride to prevent and control dental caries munities over a 3-year period. Caries 54 Petersen PE, Bourgeois D, Ogawa H,
in the United States. MMWR Morb Mor- Res 2001;35:95–105. Estupinan-Day S, Ndiaye C: The global
tal Wkly Rep 2001;50:1–42. 42 Ministério da Saúde: Projecto SB Brasil burden of oral diseases and risks to oral
30 Fawell J, Bailey K, Chilton J, Dahi E, 2003: Condições de Saúde Bucal da Pop- health. Bull World Health Organ 2005;
Fewtrell L, Magara Y: Fluoride in ulação Brasileira 2002–2003. Resultados 83:661–669.
drinking-water. London, IWA (World principais (in Portuguese). Brasília, 55 Human Development Report Team,
Health Organization publication), 2006 Ministério da Saúde, 2004. Klugman J, et al: Human Development
31 American Dental Association: Fluoride 43 Brunelle JA, Carlos JP: Recent trends in Report 2010: 20th Anniversary Edition.
& Fluoridation. Chicago, ADA, 2010. dental caries in U.S. children and the The Real Wealth of Nations: Pathways to effect of water fluoridation. J Dent Res Human Development. New York, Pal-
32 Cross DW, Carton RJ: Fluoridation: a 1990;(69 Spec No):723–727, discussion grave Macmillan (for United Nations
violation of medical ethics and human 820–823. Development Programme), 2010.
rights. Int J Occup Environ Health 44 Ellwood RP, O’Mullane DM: The associa- 56 Webb BC, Simpson SL, Hairston KG:
2003;9:24–29. tion between area deprivation and den- From politics to parity: using a health
33 McDonagh MS, Whiting PF, Wilson PM, tal caries in groups with and without disparities index to guide legislative
Sutton AJ, Chestnutt I, Cooper J, Misso fluoride in their drinking water. Com- efforts for health equity. Am J Public
K, Bradley M, Treasure E, Kleijnen J: munity Dent Health 1995;12:18–22. Health 2011;101:554-560.
Systematic review of water fluoridation. 45 Lawrence HP, Sheiham A: Caries pro- 57 Bürgi H, Zimmermann MB: Salt as a
BMJ 2000;321:855–859. gression in 12 to 16-year-old schoolchil- carrier of iodine in iodine deficient
34 Committee on Fluoride in Drinking dren in fluoridated and fluoride- areas. Schweiz Monatsschr Zahnmed
Water, Board on Environmental Studies deficient areas in Brazil. Community 2005;115:648–650.
and Toxicology, Division on Earth and Dent Oral Epidemiol 1997;25:402–411. 58 Marthaler TM: Overview of salt fluori-
Life Studies, National Research Council 46 Riley JC, Lennon MA, Ellwood RP: The dation in Switzerland since 1955, a short
of the National Academies: Fluoride in effect of water fluoridation and social history. Schweiz Monatsschr Zahnmed
Drinking Water: A Scientific Review of inequalities on dental caries in 2005;115:651–655.
EPA’s Standards. Washington, National 5-year-old children. Int J Epidemiol
Academies Press, 2006. 1999;28:300–305

144 Sampaio · Levy

59 Baez RJ, Marthaler TM, Baez MX, 70 Solórzano I, Salas MT, Chavarría P, 82 Horowitz HS: Water fluoridation and
Warpeha RA: Urinary fluoride levels in Beltrán-Aguilar E, Horowitz H: Preva- other methods for delivering systemic
Jamaican children in 2008, after 21 years lence and severity of dental caries in fluorides; in Stallard RE (eds): A Text-
of salt fluoridation. Schweiz Monatsschr Costa Rican schoolchildren: results of book of Preventive Dentistry, ed 2. Phil-
Zahnmed 2010;120:21–28. the 1999 national survey. adelphia, W.B. Saunders Company, 1982,
60 Marthaler TM: Increasing the public Int Dent J 2005;55:24–30. pp 147–169.
health effectiveness of fluoridated salt. 71 Menghini G: Dental fluorosis in salt flu- 83 Ismail AI, Hasson H: Fluoride supple-
Schweiz Monatsschr Zahnmed 2005;115: oridation schemes. Schweiz Monatsschr ments, dental caries and fluorosis: a sys-
785–792. Zahnmed 2005;115:1026–1030. tematic review. J Am Dent Assoc 2008;
61 Tramini P: Salt fluoridation in France 72 Ziegler E: Cariesprophylaxe durch Fluo- 139:1457–1468.
since 1986.Schweiz Monatsschr Zah- rierung der Milch. Schweiz Med 84 Levy SM, Broffitt B, Marshall TA,
nmed 2005;115:656–658. Wochenschr 1953;83:723–724. Eichenberger-Gilmore JM, Warren JJ:
62 Schulte AG: Salt fluoridation in Ger- 73 Imamura Y: Treatment of school meals Associations between fluorosis of per-
many since 1991. Schweiz Monatsschr with sodium fluoride as a means of pre- manent incisors and fluoride intake
Zahnmed 2005;115:659–662. venting tooth decay. J Oral Dis Acad from infant formula, other dietary
63 Gillespie GM, Baez R: Development of 1959;26:180–199. sources and dentifrice during early
salt fluoridation in the Americas. Sch- 74 Bánóczy J, Petersen PE, Rugg-Gunn AJ: childhood. J Am Dent Assoc 2010;
weiz Monatsschr Zahnmed 2005;115: Milk Fluoridation for the Prevention of 141:1190–1201.
663–669. Dental Caries. Geneva, World Health 85 Pendrys DG, Haugejorden O, Bårdsen A,
64 Gillespie GM, Marthaler TM: Cost Organization, 2009. Wang NJ, Gustavsen F: The risk of
aspects of salt fluoridation. Schweiz 75 World Health Organization: Oral health enamel fluorosis and caries among Nor-
Monatsschr Zahnmed 2005;115: promotion: an essential element of a wegian children: implications for Nor-
778–784. health-promoting school (WHO infor- way and the United States. J Am Dent
65 Jones S, Burt BA, Petersen PE, Lennon mation series on school health, docu- Assoc 2010;141:401–414.
MA: The effective use of fluorides in ment 11). Geneva, World Health Organi- 86 Rozier RG, Adair S, Graham F, Iafolla T,
public health. Bull World Health Organ zation, 2003. Kingman A, Kohn W, Krol D, Levy S,
2005;83:670–676. 76 International Dairy Federation: The Pollick H, Whitford G, Strock S,
66 Dahl LK: Possible role of salt intake in World Dairy Situation 2007 (423/2007). Frantsve-Hawley J, Aravamudhan K,
the development of essential hyperten- Brussels, Bulletin of the International Said M, Whall C, Meyer DM: Evidence-
sion. Int J Epidemiol 2005;34:967–972. Dairy Federation, 2007. based clinical recommendations on the
67 Hedman J, Sjöman R, Sjöström I, Twet- 77 Rugg-Gunn AJ: Nutrition and Dental prescription of dietary fluoride supple-
man S: Fluoride concentration in saliva Health. Oxford, Oxford University Press, ments for caries prevention: a report of
after consumption of a dinner meal pre- 1993. the American Dental Association Coun-
pared with fluoridated salt. Caries Res 78 Twetman S, Nederfors, Petersson LG: cil on Scientific Affairs. JADA 2010;141:
2006;40:158–162. Fluoride concentration in whole saliva 1480–1489.
68 Marthaler TM, Schulte AG: Monitoring and separate gland secretions in school- 87 American Dental Association: Fluoride
salt fluoridation programs through uri- children after intake of fluoridated milk. supplement dosage schedule – 1994.
nary excretion studies. Schweiz Monatss- Caries Res 1998;32:412–416.
chr Zahnmed 2005;115: 679–684. 79 Prabhakar AR, Kurthukoti AJ, Gupta P: (accessed 26 December 2010).
69 Baez R, Horowitz H, Warpeha R, Suther- Cariogenicity and acidogenicity of 88 Gagnon F, Catellier P, Arteau-Gauthier I,
land B, Thamer M: Salt fluoridation and human milk, plain and sweetened Simard-Tremblay E, Lepage-Saucier M,
dental caries in Jamaica. Community bovine milk: an in vitro study. J Clin Paradis-Robert N, Michel J, Lavallière A:
Dent Oral Epidemiol 2001;29:247–252. Pediatr Dent 2010;34:239–247. Compliance with fluoride supplements
80 Bowen WH, Lawrence RA: Comparison provided by a dental hygienist in homes
of the cariogenicity of cola, honey, cow of low-income parents of preschool chil-
milk, human milk, and sucrose. Pediat- dren in Quebec. J Public Health Dent
rics 2005;116:921–926. 2007;67:60–63.
81 Espelid I: Caries preventive effect of flu-
oride in milk, salt and tablets: a litera-
ture review. Eur Arch Paediatr Dent

Fábio Correia Sampaio

Department of Clinical and Social Dentistry, Health Science Centre
Federal University of Paraiba
João Pessoa, 58051–900 (Brazil)
Tel. +55 83 3216 7795, E-Mail

Systemic Fluoride 145

Impact of Fluoride in the Prevention of Caries and Erosion
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 146–157

Oral Fluoride Reservoirs and the Prevention of

Dental Caries
Gerald Lee Vogel
American Dental Association Foundation, Paffenbarger Research Center, Gaithersburg, Md., USA

Abstract of decreasing the F content of topical agents while simul-

Current models for increasing the anti-caries effects of taneously increasing their cariostatic effect.
fluoride (F) agents emphasize the importance of main- Copyright © 2011 S. Karger AG, Basel
taining a cariostatic concentration of F in oral fluids. The
concentration of F in oral fluids is maintained by the Recent in vitro and in vivo studies as well as
release of this ion from bioavailable reservoirs on the clinical observations [1, 2] have emphasized the
teeth, oral mucosa and – most importantly, because of importance of maintaining a concentration of flu-
its association with the caries process – dental plaque. oride (F) in oral fluids that is cariostatically effec-
Oral F reservoirs appear to be of two types: (1) mineral tive. This focus on oral fluid F has in turn drawn
reservoirs, in particular calcium fluoride or phosphate- attention to the importance of the ‘bioavailable’
contaminated ‘calcium-fluoride-like’ deposits; (2) biologi- F reservoirs that can persistently increase these
cal reservoirs, in particular (with regard to dental plaque) concentrations in the fluids surrounding the site
F held to bacteria or bacterial fragments via calcium-fluo- of de- and remineralization on the teeth.
ride bonds. The fact that all these reservoirs are mediated
by calcium implies that their formation is limited by the
low concentration of calcium in oral fluids. By using novel Types of Oral Fluoride Reservoirs
procedures which overcome this limitation, the forma-
tion of these F reservoirs after topical F application can Oral F reservoirs can be divided into two broad types,
be greatly increased. Although these increases are asso- both of which involve calcium (Ca): (1) the mineral
ciated with substantive increases in salivary and plaque deposits of F which include calcium fluoride (CaF2)
fluid F, and hence a potential increase in cariostatic effect, and fluorapatite (FAp); (2) biologically/bacterially
it is unclear if such changes are related to the increases bound calcium-fluoride (Ca-F) deposits.
in the amount of these reservoirs, or changes in the
types of F deposits formed. New techniques have been
developed for identifying and quantifying these depos- Fluorapatite
its which should prove useful in developing agents that
enhance formation of oral F reservoirs with optimum F The formation and dissolution of FAp, Ca10
release characteristics. Such research offers the prospect (PO4)6F2, is governed by the relationship of a
constant, the solubility product (KSP) of this min- however has a very slow rate of dissolution due
eral and the ion activity product (IAP) of the ions to adsorption of oral fluid phosphate onto
of which it is composed. The KSP of FAp (KSPFAp) the surface of the mineral [7, 15], probably in
is 10–121 [3] while the IAP of FAp can be calculated the form of fluorapatite [7]. These phosphate
from IAPFAp = {Ca2+}10 × {PO43–}6 × {F–}2 where { } stabilized CaF2 deposits lose PO4–3 ions under
is the chemical activities of these ions. The activity low pH (cariogenic conditions) by protonation of
of an ion is the free unbound concentration of the these phosphate groups and thus dissolve rapidly
ion multiplied by an ionic-strength-dependent ac- [7, 9]. However CaF2 formed in the presence of
tivity coefficient [4]. The relationship between KSP phosphate has considerably different proper-
and IAP is a step function: an all-or-nothing rela- ties from pure CaF2, and some studies have sug-
tionship in which dissolution/formation of a min- gested a moderately rapid rate of dissolution for
eral occurs when the IAPFAp is above the KSPFAp the type of phosphate-containing CaF2 deposits
(supersaturation) or below the KSPFAp (undersatu- formed in the oral environment [8] while oth-
ration). Even under cariogenic conditions, IAPFAp er studies have suggested that such deposits can
values for oral fluid are usually much greater than be persistent [9, 16]. In this chapter, these phos-
KSPFAp [5, 6]; thus, FAp does not dissolve, and phate-contaminated deposits are referred to, in
hence FAp mineral is a poor source of oral fluid F. accordance with other authors [7, 9], as ‘calcium
Furthermore, it is possible, as noted below, to re- fluoride-like’ (CaF2-like).
duce the release of F to oral fluids by inducing the
formation of this mineral.
Biologically/Bacterially Bound Fluoride

Calcium Fluoride Bacterially Bound Fluoride

In an extensive study of bacterial Ca-F bind-
Oral fluids, such as saliva and the fluid phase of ing, Rose et al. [17] proposed a model (fig. 1) in
plaque (plaque fluid), are highly supersaturated which F reacts with intercellular or intracellu-
with respect to CaF2 after application of topical lar Ca ‘bridges’ to form calcium-fluoride (Ca-F)
F agents, and hence this mineral has long been bonds at fixed anionic bacterial sites. An impor-
regarded as the primary source of bioavailable F tant feature of this model is that the application of
in the oral environment [7–9]. CaF2 dissolution F breaks bidentate Ca bonds, leading to more Ca
and formation is governed by the same type of and F binding (fig. 1A, B). These authors demon-
relationship noted for fluorapatite, with KSPCaF2 strated that that bacterial Ca-F binding was quite
= 3 × 10–10.4 [3] and IAPCaF2 = {Ca2+} × {F–}2. unlike the binding of F by the mineral deposits
Since resting oral fluids such as plaque fluid or described above: it is a continuous function of the
saliva have a {Ca} of about 1 mmol/l [5, 10, 11], {Ca2+} and {F–} which is also dependent on the
the critical value of free F that will induce a dis- number of binding sites on the bacteria (i.e. bind-
solution of pure CaF2 can be calculated to be ing capacity) and the pH [17]. The pH depen-
about 450 μmol/l [11]. This fluoride concentra- dence (fig. 1D) is a consequence of the competi-
tion is reached within about 10 min after a use tion of Ca2+ and H+ for the same anionic sites on
of a conventional strength topical agent [12–14], the bacterial surface.
and thus in theory CaF2 deposits should rapidly
dissolve in oral fluids within a short period after Non-Bacterial Biological F Binding
use of a topical F agent. Pure CaF2 exposed to Other ‘biological’ F binding sites exist in the oral
phosphate-containing solutions, such as saliva, environment besides bacteria – such as proteins,

Oral F Reservoirs and Caries 147

A F–
Ca uptake
Ca Adding F– breaks
bidentate Ca binding;
exposes new sites
F–/Ca2+ Ca
F — Ca –

Loss of F– to saliva, –
system returns to Ca — F
initial state –

Ca — F
F — Ca
Ca — F Ca — F – B
Ca — F
Additional sites
C Ca — F
Ca — F
Ca — F
Ca — F Uptake on
Ca — F
pH falls, H+ newly exposed sites
displaces Ca2+
and F–
F — Ca H

F–/Ca2+ H
released H Ca — F

Ca — F

Ca — F H

Fig. 1. The fluoride/calcium binding model of Rose et al. [17]. Before bacteria is exposed to fluo-
ride (A), calcium (Ca) is intercellularly and intracellularly attached to bacteria in a bidentate fashion.
Adding fluoride (F–) breaks Ca bonds and exposes new binding sites (B), which leads to more F– and
Ca2+ uptake (C). With time, F is lost to saliva and the bacteria returns to the initial state (A). However,
when the bacteria is exposed to low pH, the increase in H+ displaces Ca2+ and F– (D).

mucosal tissue and (most importantly with re- Location of Oral Bioavailable Fluoride
gard to plaque) bacterial fragments. It appears Reservoirs
however that these moieties have binding prop-
erties that are similar to those described for bac- Bioavailable Fluoride Reservoirs on or in the Teeth
teria in that they all appear to involve calcium-
to-fluoride binding [11]. In this article these Fluorapatite
deposits are referred to generically as biological/ Due to the relative insolubility, and thus resis-
bacterial Ca-F. tance to the acidic attack characteristic of caries,

148 Vogel
the formation of fluorapatite (FAp) in/on the a constant composition procedure has been de-
tooth had been considered as the primary goal scribed, which not only quantifies these deposits,
of F therapies for many years [18]. Although a but also measures their rate of release into a ‘saliva-
significant amount of FAp is found on the tooth like’ solution of a chosen F concentration [21].
surface (especially in caries-prone areas), this in-
soluble mineral is a poor source of oral fluid F.
Thus, FAp provides little protection to adjacent Bioavailable Fluoride Reservoirs in the Mucosa
F-poor tooth minerals, i.e. during caries progres- and Salivary Fluoride
sion, demineralization simply bypasses the FAp
F-rich minerals in the outer layers of the lesion Salivary Fluoride
body and dissolves F-poor minerals at the ad- During and shortly after administration of topi-
vancing front. cal F agents, high levels of F are delivered to teeth
and plaque via the saliva. However, salivary F con-
CaF2-Like Deposits centrations rapidly fall below plaque fluid F levels
Many studies have identified CaF2-like depos- [10, 12, 22], which tends to discount salivary F as a
its as the most important labile source of F on/in persistent source of plaque F. Salivary F, however,
the tooth surface [7, 9, 15, 16]. CaF2-like depos- doubtlessly plays an important role in the remin-
its can be formed by the reaction of tooth-bound eralization of the plaque-free area of the teeth.
Ca with the applied F. However, at ‘resting’ pH,
the low solubility of tooth mineral limits the rate Location of Salivary F reservoirs
of release of this ion. Thus, it appears doubtful Zero et al. [23] found that edentulous subjects had
that clinically significant amounts of this mineral higher salivary F levels than a dentate panel, sug-
could form on teeth [1, 6, 19] unless the pH of gesting that oral soft tissue is the major source of
the topical agent is low (as in the case of APF [9, salivary F. Surprisingly, it has been shown that flu-
20]), the application time is quite long (as with id recovered by scraping the mucosal surface af-
fluoride varnishes [19]), or a high concentration ter a F rinse is not only higher than saliva samples
of the fluoride topical agents are employed [9, 16, [24], but higher in F than plaque fluid samples re-
19]. Once formed, however, tooth deposits of this covered at the same time (unpublished data pre-
mineral have desirable properties as a F source, as sented at the 1997 IADR meeting, abstract 174 ).
evidenced by the cariostatic effect of infrequent- Little is known of the nature of mucosal F reser-
ly applied APF rinses and varnishes [9]. The rea- voirs, other than they are easily depleted by water
sons for this beneficial effect include the location exposure, since post-water rinsing dramatically
of these deposits at the site of the caries activity decreases post-topical application salivary F lev-
and their ability to release additional F during a els [25, 26]. However, in view of the fact that: (1)
cariogenic challenge [1, 7]. increases in salivary-free F appear to predict in-
creases in plaque fluid F [5, 10, 12, 22, 27], and (2)
Quantification of Fluoride Reservoirs on or in the salivary-free F and plaque fluid F appear to be cor-
Teeth related, at least in samples collected at times ≥2 h
Because CaF2 is soluble in a basic low-phosphate post application [12, 27], these deposits are likely
solution, while FAp is nearly insoluble in such so- to be similar in nature to those found in plaque.
lutions, a sequential extraction of the tooth sur-
face by base and then acid is often used to quan- Measurement of Whole and Free Salivary F
tify CaF2-like and FAp deposits on human teeth Although free saliva F, which is measured on cen-
[16, 19]. For measuring small amounts of labile F, trifuged or filtered saliva, is more relevant to the

Oral F Reservoirs and Caries 149

cariostatic effect of F, salivary F is often reported
on the whole saliva, which also includes F held in
salivary particulates. Whole saliva F can be con- 120

Plaque fluid F (μmol/l)

siderably larger than free saliva F, especially after 100
use of Ca and phosphate-containing experimental
agents [27].

Bioavailable Fluoride Reservoirs in Plaque
and Plaque Fluid F
0 2 4 6 8 10
Plaque Fluid F Total plaque F (μmol/g)
Changes in the concentration of ions in plaque
fluid is the major factor governing the de/remin- Fig. 2. Relationship between total plaque and plaque
eralization characteristic of caries [4, 5, 28, 29], fluid F of samples recovered 1 h after a NaF (filled solid
and it is in this milieu that F exerts its major anti- square), MFP (filled solid diamond) or 2 experimental rins-
caries effects [11, 29]. Thus ‘bioavailable’ plaque es (filled circles) containing high levels of calcium and/or
phosphate. The average values for these rinses are shown
F reservoirs that substantively increase the plaque by the corresponding large open symbols. The individual
fluid F concentrations are of critical importance MFP and NaF data points are unpublished data from refer-
in the cariostatic effect of this ion. ence 22. The red circle experimental rinse data is unpub-
lished data from an abstract presented at the 2001 IADR
meeting (abstract 1294) and refers to a controlled release
Plaque F Reservoirs rinse with a high Ca content (see text). The black circle
Plaque fluid is greatly supersaturated with respect data is unpublished data obtained by the author follow-
to FAp. Because FAp is sensitive to the high con- ing the procedure of reference 22. This rinse contained
centration of mineralization inhibitors found in high levels of both Ca and PO4.
the oral environment [30, 31], no significant stores
of this mineral have been detected in plaque. More
importantly, the insolubility of this mineral ap-
pears to negate its value as a plaque source of oral rinse. Although, there is no correlation between
fluid F, and thus CaF2-like deposits and biologi- plaque fluid F and total plaque F, the average total
cal/bacterial Ca-F appear to be the major plaque plaque F after these 2 rinses reflects the average
F reservoirs that increase plaque fluid F [7, 9, 17, change in plaque fluid F. This is not always the
32, 33]. As described further below, recent stud- case: the 2 experimental rinses shown in this fig-
ies [11] have suggested that unless additional Ca is ure deposited large amounts of plaque F, yet pro-
supplied with conventional ‘over the counter’ topi- duced no increase in plaque fluid F. These rinses,
cal agents, these reservoirs appear to be primarily which had high concentrations of Ca, and/or PO4,
in the form of biological/bacterial Ca-F deposits. appear to have formed primarily insoluble FAp.

Relationship of Plaque and Plaque Fluid F Location of Plaque Fluoride Reservoirs, and the
The relationship between plaque fluid and total Effect of Water Rinsing on Plaque Fluoride
plaque F in resting plaque is illustrated by the data Studies performed on plaque samples formed in
in figure 2; samples were collected 1 h after a 228- a 1-mm height ring and recovered after a topical
μg/g F rinse given as NaMFP or NaF [22]. Here F application have found a step gradient in total
μg/g (ppm) refers to the mass fraction of F in the plaque F from the saliva to the enamel interface

150 Vogel
[34]. In view of the tenuous relationship between the most insoluble F phases will completely dis-
total plaque and plaque fluid F noted above, the solve (even at neutral pH). Such techniques are
effect of this distribution on the subsequent dis- also sensitive to sample handling procedures,
tribution of plaque fluid F is unclear. However, such as loss of fluid, which may induce conver-
given the shallow deposition of F found in these sion of these reservoirs to FAp.
studies, the fact that a 30-second saliva-like wash
administered after the F application induced no Quantification of CaF2-Like Deposits in Plaque
significant F loss suggests that the F deposits must Recently, an extraction procedure was described
not only form quickly, but release F slowly [34]. that permits the quantification of CaF2-like de-
Noteworthy in this regard is a study that found posits in plaque recovered shortly after use of a
only a small change in total plaque F in subjects F topical agent [11]: one of a pair of matched ho-
who rinsed with water 1 h after use of a F rinse mogenized aliquots is repeatedly extracted with
[10]. a very low phosphate-containing solution having
the same {Ca2+}, {F–} and {H+} as the plaque fluid
Release of Plaque F under Cariogenic Conditions recovered from these same samples. Since, short-
Studies on plaque fluid F, such as those described ly after a F rinse, plaque fluid IAPCaF2 ({Ca2+} ×
above, were performed on resting plaque, and thus {F–}2) is well below KSPCaF2, this extraction dis-
are more relevant to the remineralization (rather solves all the CaF2-like deposits. However, plaque
than the demineralization) phase of the caries pro- reservoirs that are in insoluble (FAp), or in equi-
cess. Unfortunately, studies of post-sucrose plaque librium with, the ‘plaque fluid-like’ solution (bio-
can be complicated by several factors: (1) in high F logical/bacterial Ca-F) would not be extracted by
plaque samples, F inhibits acid production [5, 35], this procedure. Hence, by comparing the total F
raises the pH and reduces F release; (2) salivary content of this aliquot with the total F content of
clearance patterns maintain a high concentration the unextracted aliquot, the amount of CaF2-like
of sucrose and F at the same sites, which increases deposits can be determined.
F release from high F sites. As a result of these fac-
tors, studies in which plaque was recovered after a No CaF2-Like Deposits Found in Plaque after Use
sucrose rinse often found no change or a decrease of Over-the-Counter Strength Topical Agents
in plaque fluid F that was unrelated to the F con- When the CaF2 extraction procedure described
tent of the samples [5, 6, 28]. A better procedure above was applied to plaque samples recovered 30
for examining the F release from plaque F reser- min after a 228-μg/g F (NaF) rinse, the F content of
voirs at cariogenic pH appears to be the use of an the 2 aliquots was nearly identical [11]. This con-
in vitro acidification or titration [36]. centration of F delivered as a rinse appears to release
more F to saliva and plaque than over-the-counter
Extraction Techniques for Examining the dentifrices [13]. Furthermore, the dentifrice ingre-
Properties of Plaque Fluoride Reservoirs dient sodium lauryl sulfate greatly reduces the for-
Water or buffer extraction has often been used mation of CaF2-like deposits [7, 38]. Thus, these
to examine plaque reservoirs’ ability to release F. results suggest that biological/bacterial bound Ca-
Because of the difficulty in examining challenged F, rather than CaF2 or ‘CaF2-like’ deposits, is the
plaque, such procedures are especially valuable major reservoir of plaque F that releases this ion
when buffers are used whose pH is similar to car- to plaque fluid in the case of over-the-counter F
iogenic plaque [37]. Unfortunately, such methods dentifrices and rinses. However, as described be-
are sensitive to the extraction conditions: given a low, CaF2-like deposits can be formed if additional
large extraction volume and sufficient time even Ca is supplied before a topical F agent.

Oral F Reservoirs and Caries 151

Increasing the Deposition of Fluoride into Oral Two-Component Controlled Release Agents
Chow and colleagues [10, 12, 27, 36] have described
Ca-F binding appears to play a central role in a two-part ‘controlled release’ rinse in which part A
the formation of all the bioavailable oral F res- contained Na2SiF6, while part B contained CaCl2
ervoirs; in fact nearly every study examining and sodium acetate. This rinse initially also con-
the relationship of Ca and F in plaque has found tained a low concentration of PO4–3; however, this
a moderate to strong correlation [33, 37–39]. component was eliminated in later studies. These
Unfortunately, salivary and plaque fluid free-Ca rinses were called ‘controlled release’ (CR) agents
is typically 2–10% of the amount of F supplied because F was slowly released by the hydrolysis of
by topical agents. Thus, the amount of oral Ca-F the SiF62– ion after parts A and B were mixed. The
that can be formed after use of a topical F agent is slow release of F in the presence of Ca could have
limited not only by the concentration of applied two effects: (1) it permits an in-depth penetra-
F, but also by the rate at which additional Ca tion by the SiF62– and Ca ions into oral tissue and
can be scavenged from Ca reservoirs in enam- plaque before F release by hydrolysis of SiF62– and
el, plaque or saliva during the short period of F subsequent formation of CaF2 or biological/bacte-
application [11, 37]. The small amounts of oral rial Ca-F deposits, or (2) alternatively, slowly grow-
Ca-F reservoirs produced by this Ca scavenging ing nano-size crystals of CaF2 could penetrate the
explain why conventional topical agents induce above substrates before aggregation and growth.
only a transient increase in plaque fluid and sali-
vary F before these reservoirs are exhausted [12, Plaque and Salivary F after Use of the CR Rinse
13, 22, 33]. There are a number of procedures for Several studies have examined plaque F reservoirs
increasing the formation of these Ca-F oral res- 60 min after administration of a 228-μg/g F CR
ervoirs, and consequently increasing the persis- rinse. Compared to a NaF rinse with a similar F
tence of high levels of oral fluid F. Because such content, the CR rinse increased total plaque F by 4×
procedures may increase the cariostatic effect of [10], the amount of water extractable F by 11× [10],
a given F dose, they offer the possibility of de- and low pH releasable F (average pH 5.2) by 9× [36].
creasing the F content of topical agents without Most importantly, the F reservoirs produced by the
compromising the clinical effect. CR rinse appear to be bioavailable since, compared
to the NaF rinse, they induce approximately a 2×
increase in both plaque fluid and centrifuged saliva
Amorphous Calcium Phosphate Agents F [10, 12, 36]. Finally, the overnight F data of figure
3 show the persistence of these increases [27]. As
Amorphous calcium phosphate products contain- noted above, the whole salivary F greatly exceeded
ing F appear to increase plaque F. A 450-μg/g F the free (centrifuged) salivary F (fig. 3), especially
mouth rinse with casein phosphopeptide stabi- in the Ca-containing CR rinses.
lized amorphous calcium phosphate was found to
double the F content of plaque over a similar rinse Enhanced Remineralization from CR Rinses
without it [40]. However, it is difficult to separate In lieu of a clinical trial, the best predictor of the
the cariostatic effect of the F reservoirs produced cariostatic effectiveness of a topical F agent is a
by such products from the effects of the enhanced well-designed ‘in situ de/remineralization’ test
levels of Ca and phosphate. Furthermore, the in- protocol in which adequate positive and nega-
activation of some of the applied F by formation of tive controls, reflecting the range of responses of
insoluble FAp is a concern with such products. known agents, is included [41]. The 228-μg/g F

152 Vogel
30 1.4
No F rinse
NaF rinse
25 CR rinse 1.2

Saliva/plaque fluid F (μmol/l)

Whole plaque F (μmol/g)





0 0
Saliva fluid Whole saliva Plaque fluid Whole plaque
(centrifuged saliva)

Fig. 3. Fluid fluoride and total fluoride in plaque and saliva measured overnight after a CR
(controlled release, see text) or NaF rinse, both with 228 μg/g fluoride [27]. The error bars refer
to standard error (n = 13). ‘Whole’ refers to the total amount of fluoride obtained by strong acid
extraction of the sample, while ‘fluid’ refers to the fluoride in the supernatant of centrifuged sam-
ples. In all types of samples, CR rinse is significantly greater than the NaF rinse and no F rinse
samples (p < 0.05).

CR rinse described above, when tested in such a and colleagues however have described other
procedure, was found to produce a remineraliza- NaF-based two-part CR systems that avoid some
tion effect that was not statistically different from a of these problems by using inhibitors (US patent
NaF rinse with 4× the F content [42]. Such results No. 5891448) or Ca chelating agents (US patent
not only indicate the potential clinical effective- No. 5476647) to control the rate of reaction of Ca
ness of the CR rinse, but also demonstrates clearly with F.
that the cariostatic effectiveness of any F product
depends not just on the amount of applied F, but
on the ability of the treatment to form bioavailable Calcium Pre-Rinse Systems
F deposits that substantively increase oral fluid F.
A pre-application of a concentrated Ca agent short-
Problems with CR Agents ly before a topical F agent is another procedure to
There are however problems with producing a ameliorate the restriction placed on the formation
commercially viable CR rinse or dentifrice: (1) of oral Ca-F reservoirs by low oral fluid Ca.
some common dentifrice/rinse ingredients can
reduce the deposition of F from CR agents; (2) al- Plaque and Salivary F after Use of a Calcium Pre-
though Na2SiF6 is approved for water fluoridation Rinse/NaF Rinse
by the American Food and Drug Administration, Compared to the NaF rinse alone, Vogel et al. [37]
it is not approved for use in topical agents. Chow found that total plaque F was elevated 12×, when

Oral F Reservoirs and Caries 153



Saliva fluid F (μmol/l)

Fig. 4. Concentration of fluoride in
centrifuged saliva samples obtained
overnight after use of a 150-mmo/l B
calcium lactate pre-rinse immedi-
ately followed by a 228 μg/g F rinse;
228 μg/g F or 912 μg/g F (with no
pre-rinse); or a distilled water rinse
(no F rinse) [44]. All fluoride rinses
were sodium fluoride. The error bars 0
refer to the standard error (n = 12). Ca pre‐rinse/ 912 μg/g 228 μg/g No F rinse
Statistical differences are indicated 228 μg/g F rinse F rinse F rinse
by the letters.

a 150-mmol/l Ca pre rinse was used immediate- one third of the deposited plaque F appeared to
ly before a 228-μg/g F (NaF) rinse. More impor- be in the form of CaF2-like deposits. Given that
tantly, the plaque fluid or centrifuged salivary F the IAPCaF2 of saliva and plaque fluid immedi-
following the pre-rinse was about 5× greater than ately after use of topical F agents exceeds KSpCaF2
in the absence of the pre-rinse [37, 43]. These in- (supersaturation), the finding that a Ca pre-rinse
creases are persistent: in centrifuged saliva sam- is required to form CaF2-like deposits may seem
ples collected overnight after the above Ca pre- surprising. However, the IAPCaF2 ({Ca2+} × {F–}2)
rinse/NaF rinse, F was 5.5× higher than after a required for the nucleation of this mineral is many
NaF rinse alone, and 2.5× higher than after a rinse orders of magnitude higher than KSPCaF2 [45], es-
with 4 times more NaF (fig. 4) [44]. Given the re- pecially in a milieu such as plaque fluid that is rich
lationship of salivary F and plaque fluid F noted in mineralization inhibitors, such as phosphate,
above, it is noteworthy that the overnight increas- which are known to reduce the rate of CaF2 pre-
es in salivary F with the use of the Ca pre-rinse cipitation [46]. Furthermore, in the absence of
were much higher than found for the CR rinses the Ca pre-rinse, the low free Ca in plaque fluid
(fig. 3), for which the more relevant total plaque F or saliva implies that the formation of significant
and plaque fluid F data are available. amounts of plaque CaF2-like precipitates would
require scavenging of additional Ca from the
Calcium Pre-Rinse Required to Produce Plaque plaque (the small amount of CaF2 formed in sa-
CaF2-Like Deposits from a Fluoride Rinse liva would primarily migrate to the mucosa rather
An examination of plaque recovered 1 h after than into the plaque). Perhaps most importantly,
the use a Ca pre-rinse/NaF rinse using a varia- this scavenging of plaque Ca by the applied F must
tion of the CaF2 extraction technique described compete with the formation of bacterial/biologi-
above (unpublished data presented at the 2010 cal Ca-F which occurs without such Ca extrac-
ORCA congress, abstract 75), found that, unlike tion [17]. Hence it is unlikely, in the absence of
the results obtained with the NaF rise alone, about the additional Ca supplied by a pre-rinse, that a

154 Vogel
conventional topical F agent can form significant ‘bioavailable’ F reservoirs, dental plaque and the
amounts of CaF2-like deposits before {F} falls be- oral mucosa, which persistently increase the con-
low the level required for deposition of this min- centration of this ion in the fluid phases that are
eral. It is unclear however, if the large persistent associated with the caries process, i.e. saliva and
increase in oral fluid F observed after use of the especially plaque fluid. Bioavailable F in these
Ca pre-rinse/F rinse is due specifically to an in- reservoirs appears to be of two types: ‘CaF2-like’
creased formation of CaF2-like deposits or to an reservoirs and fluoride bound to bacterial/bio-
increase in the total amount of F in oral reservoirs. logical substrates by Ca. Because fluoride binding
The use of this extraction technique with samples in both these moieties appears to be mediated by
recovered overnight should be able to address this Ca, their formation is limited by the low concen-
question. tration of this ion in oral fluids. Several agents or
procedures have been described that, by supply-
Problems with the Use of a Ca Pre-Application ing additional Ca, greatly increase the deposition
There are a number of potential problems with of Ca-F oral reservoirs and consequently increase
using a Ca pre-application in conjunction with a the persistence of high levels of oral fluid fluoride.
F topical agent to increase oral F reservoirs: (1) It is unclear, however, if such changes are related
a Ca-binding detergent found in many commer- to the increases in the total amount of F in these
cial F dentifrices, sodium lauryl sulfate, appears reservoirs, or changes in the fraction of these de-
to reduce the ability of the Ca pre-application to posits in the form of CaF2-like or bacterial/bio-
increase salivary and plaque F [38, 43]; (2) patient logical reservoirs. New techniques have been
compliance with a two-part regimen is problem- developed for identifying these deposits which
atic; however, it has been shown that Ca deliv- should prove useful in engineering topical agents
ered as a dentifrice before a short F rinse (10 s) that maximize the formation of oral fluoride res-
also produced large increases in salivary F [43]. ervoirs with optimum fluoride release character-
Because such procedures are similar to current istics. Such agents, by increasing the cariostatic
oral hygiene practices (substituting the F rinse for effect of a given fluoride dose, offer the possibility
the usual post-dentifrice water rinse), they can of substantially decreasing the fluoride content of
potentially achieve a high level of compliance. over-the-counter topical agents without compro-
mising the clinical effect. Conversely, it appears
possible to formulate conventional strength rins-
Conclusion es and dentifrices that produce a considerably
higher and more sustained F release into oral
Current models for increasing the anti-caries ef- fluids than high-strength prescription products.
fects of fluoride agents emphasize the importance Several studies have concluded that such agents
of maintaining a cariostatic concentration of F do confer significant additional cariostatic effects
in oral fluids. This has focused attention on the [47–49].

1 ten Cate JM: Review on fluoride, with 2 Featherstone JD: The science and prac- 3 McCann HG: The solubility of fluorapa-
special emphasis on calcium fluoride tice of caries prevention. J Am Dent tite and its relationship to that of cal-
mechanisms in caries prevention. Assoc 2000;131:887–899. cium fluoride. Arch Oral Biol 1968;13:
Eur J Oral Sci 1997;105:461–465. 987–1001.

Oral F Reservoirs and Caries 155

4 Vogel GL, Carey CM, Chow LC, Tat- 15 Lagerlöf F, Saxegaard E, Barkvoll P, Rølla 27 Vogel GL, Mao Y, Carey CM, Chow LC:
evossian A: Micro-analysis of plaque G: Effects of inorganic orthophosphate Increased overnight fluoride concentra-
fluid from single-site fasted plaque. and pyrophosphate on dissolution of tions in saliva, plaque, and plaque fluid
J Dent Res 1990;69:1316–1323. calcium fluoride in water. J Dent Res after a two-solution rinse. J Dent Res
5 Vogel GL, Zhang Z, Chow LC, Schu- 1988;67:447–449. 1997;76:761–767.
macher GE: Changes in lactate and other 16 Øgaard B, Rølla G, Helgeland K: Uptake 28 Tanaka M, Margolis HC: Release of min-
ions in plaque and saliva after a fluoride and retention of alkali-soluble and eral ions in dental plaque following acid
rinse and subsequent sucrose adminis- alkali-insoluble fluoride in sound production: Arch Oral Biol 1999:44:
tration. Caries Res 2002;36:44–52. enamel in vivo after mouthrinses with 253–258.
6 Pearce ElF, Margolis H, Kent RL Jr: 0.05% or 0.2% NaF. Caries Res 1983;17: 29 Margolis HC, Moreno EC: Physicochem-
Effect of in situ plaque mineral supple- 520–524. ical perspectives on the cariostatic
mentation on the state of saturation of 17 Rose RK, Shellis RP, Lee AR: The role of mechanisms of systemic and topical flu-
plaque fluid during sugar induced acido- cation bridging in microbial fluoride orides. J Dent Res 1990;69(Spec No):
genesis. Eur J Oral Sci 1999;107:251– binding: Caries Res 1996;30:458–464. 606–613.
259. 18 Chow LC: Tooth-bound fluoride and 30 Hay DI, Moreno EC: Macromolecular
7 Rølla G, Saxegaard E: Critical evaluation dental caries. J Dent Res 1990;69: inhibitors of calcium phosphate precipi-
of the composition and use of topical 595–600. tation in human saliva: their roles in
fluorides with emphasis on the role of 19 Cruz R, Øgaard B, Rølla G: Uptake of providing a protective environment for
calcium fluoride in caries inhibition. KOH-soluble and KOH-insoluble fluo- the teeth; in Kleinberg I, Ellison SA,
J Dent Res 1990;69(Spec No):780–785. ride in sound human enamel after topi- Mandel ID (eds): Saliva and Dental Car-
8 Larsen MJ, Ravnholt G: Dissolution of cal application of a fluoride varnish ies. Microbiology Abstracts 1979;(sup
various calcium fluoride preparations in (Duraphat) or a neutral 2% NaF solution suppl):45–58.
inorganic solutions and in stimulated in vitro. Scand J Dent Res 1992;100: 31 Tomazic B, Tomson M, Nancollas GH:
human saliva. Caries Res 1994;28: 154–158. Growth of calcium phosphates on
447–454. 20 Tsuda H, Jongebloed WL, Stokroos I, hydroxyapatite crystals: the effect of
9 Øgaard B: CaF2 formation: cariostatic Arends J: Combined Raman and SEM magnesium. Arch Oral Biol 1975;20:
properties and factors of enhancing the study on CaF2 formed on/in enamel by 803–808.
effect. Caries Res 2001;35(suppl 1): APF treatments. Caries Res 1993;27: 32 Larsen MJ, Richards A: The influence of
40–44. 445–454. saliva on the formation of calcium fluo-
10 Vogel GL, Zhang Z, Chow LC, Schu- 21 Sieck B, Takagi S, Chow LC: Assessment ride-like material on human dental
macher GE: Effect of a water rinse on of loosely-bound and firmly-bound fluo- enamel. Caries Res 2001;35:57–60.
‘labile’ fluoride and other ions in plaque ride uptake by tooth enamel from topi- 33 Whitford GM, Wasdin JL, Schafer TE,
and saliva before and after conventional cally applied fluoride treatments. J Dent Adair SM: Plaque fluoride concentra-
and experimental fluoride rinses. Caries Res 1990;69:1261–1265. tions are dependent on plaque calcium
Res 2001;35:116–124. 22 Vogel GL, Mao Y, Chow LC, Proskin HM: concentrations: Caries Res 2002;36:
11 Vogel GL, Tenuta LM, Schumacher GE, Fluoride in plaque fluid, plaque, and 256–265.
Chow LC: No calcium-fluoride-like saliva measured for two hours after a 34 Watson PS, Pontefract HA, Devine DA,
deposits detected in plaque shortly after NaF or NaMFP rinse. Caries Res 2000; Shore RC, Nattress BR, Kirkham J, Rob-
a sodium fluoride mouthrinse. Caries 34:404–411. inson C: Penetration of fluoride into
Res 2010;44:108–115. 23 Zero DT, Raubertas RF, Pedersen AM, natural plaque biofilms. J Dent Res
12 Vogel GL, Mao Y, Carey CM, Chow LC, Fu J, Hayes AL, Featherstone JDB: Stud- 2005;84:451–455.
Takagi S: In vivo fluoride concentrations ies of fluoride retention by oral soft tis- 35 Hamilton IR: Biochemical effects of fluo-
measured for two hours after a NaF or a sues after the application of home-use ride on oral bacteria. J Dent Res 1990;
novel two-solution rinse. J Dent Res topical fluorides. J Dent Res 1992;71: 69(Special No):660–667.
1992;71:448–452. 1546–1552. 36 Vogel GL, Zhang Z, Chow LC, Schu-
13 Zero DT, Raubertas RF, Fu J, Pedersen 24 Jacobson AP, Stephen KW, Strang R: Flu- macher GE, Banting D: Effect of in vitro
AM, Hayes AL, Featherstone JDB: Fluo- oride uptake and clearance from the acidification on plaque fluid composi-
ride concentrations in plaque, whole buccal mucosa following mouthrinsing. tion with and without a NaF or a con-
saliva, and ductal saliva after application Caries Res 1992;26:56–58. trolled release fluoride rinse. J Dent Res
of home-use topical fluorides: J Dent Res 25 Duckworth RM, Knoop DT, Stephen 2000;79:983–990.
1992;71:1768–1775. KW: Effect of mouthrinsing after tooth- 37 Vogel GL, Schumacher GE, Chow LC,
14 Bruun C, Givskov H, Thylstrup A: Whole brushing with a fluoride dentifrice on Takagi S, Carey CM: Ca pre-rinse greatly
saliva fluoride after toothbrushing NaF human salivary fluoride levels. Caries increases plaque and plaque fluid F.
and MFP dentifrices with different F Res 1991;25:287–291. J Dent Res 2008;87:466–469.
concentrations: Caries Res 1984;18: 26 Sjögren K, Birkhed D: Effect of various 38 Pessan JP, Sicca CM, de Souza TS, da
282–288. post-brushing activities on salivary fluo- Silva SMB, Whitford GM, Buzalaf MAR:
ride concentration after toothbrushing Fluoride concentrations in dental plaque
with a sodium fluoride dentifrice: Caries and saliva after the use of a fluoride den-
Res 1994;28:127–131. tifrice preceded by a calcium lactate
rinse. Eur J Oral Sci 2006;114:489–493.

156 Vogel
39 Whitford GM, Buzalaf MAR, Bijella 42 Chow LC, Takagi S, Carey CM, Sieck BS: 47 Bartizek RD, Gerlach RW, Faller RV,
MFB, Waller JL: Plaque fluoride concen- Remineralization effect of a two-solution Jacobs SA, Bollmer BW, Biesbrock AR:
trations in a community without water fluoride mouth rinse – an in situ study. Reduction in dental caries with four
fluoridation: effects of calcium and use J Dent Res 2000;79:991–995. concentrations of sodium fluoride in a
of a fluoride or placebo dentifrice. Caries 43 Vogel GL, Shim D, Schumacher GE, dentifrice: a meta-analysis evaluation.
Res 2005;39:100–107. Carey CM, Chow LC, Takagi S: Salivary J Clin Dent 2001;12:57–62.
40 Cochrane NJ, Cai F, Huq NL, Burrow MF, fluoride from fluoride dentifrices or 48 Stookey GK, Mau MS, Isaacs RL, Gonza-
Reynolds EC: New approaches to rinses after use of a calcium pre-rinse or lez-Gierbolini C, Bartizek RD, Biesbrock
enhanced remineralization of tooth calcium dentifrice. Caries Res 2006;40: AR: The relative anticaries effectiveness
enamel. J Dent Res 2010;89:1187–1197. 449–454. of three fluoride-containing dentifrices
41 Proskin HM: Statistical considerations 44 Vogel GL, Chow LC, Carey CM: Calcium in Puerto Rico. Caries Res 2004;38:
related to the use of caries model sys- pre-rinse greatly increases overnight 542–550.
tems for the determination of clinical salivary fluoride after a 228 ppm fluo- 49 Tavss EA, Mellberg JR, Joziak M, Gam-
effectiveness of therapeutic agents. Adv ride rinse. Caries Res 2008;42:401–404. bogi RJ, Fisher SW: Relationship
Dent Res 1995;9:270–279. 45 Larsen MJ, Jensen SJ: Experiments on between dentifrice fluoride concentra-
the initiation of calcium fluoride forma- tion and clinical caries reduction.
tion with reference to the solubility of Am J Dent 2003;16:369–674.
dental enamel and brushite. Arch Oral
Biol 1994;39:23–27.
46 Christoffersen J, Christoffersen MR,
Kibalczyc W, Perdok WG: Kinetics of
dissolution and growth of calcium fluo-
ride and effects of phosphate. Acta
Odontol Scand 1988;46:325–336.

Gerald Lee Vogel

American Dental Association Foundation, Paffenbarger Research Center
100 Bureau Drive Stop 8546, National Institute of Standards and Technology
Gaithersburg MD 20899–8546 (USA)
Tel. +1 301 975 6821, E-Mail

Oral F Reservoirs and Caries 157

Impact of Fluoride in the Prevention of Caries and Erosion
Buzalaf MAR (ed): Fluoride and the Oral Environment.
Monogr Oral Sci. Basel, Karger, 2011, vol 22, pp 158–170

Fluoride in Dental Erosion

Ana Carolina Magalhãesa ⭈ Annette Wiegandb ⭈ Daniela Riosa ⭈
Marília Afonso Rabelo Buzalafa ⭈ Adrian Lussic
aBauru Dental School, University of São Paulo, Bauru, Brazil; bUniversity of Zürich, Zürich, and cUniversity of Bern, Bern, Switzerland

Abstract potentially effective approaches in preventing dental ero-

Dental erosion develops through chronic exposure to sion. The use of tin-containing fluoride products might
extrinsic/intrinsic acids with a low pH. Enamel erosion is provide the best approach for effective prevention of den-
characterized by a centripetal dissolution leaving a small tal erosion. Further properly designed in situ or clinical
demineralized zone behind. In contrast, erosive deminer- studies are recommended in order to better understand
alization in dentin is more complex as the acid-induced the relative differences in performance of the various fluo-
mineral dissolution leads to the exposure of collagenous ride agents and formulations.
organic matrix, which hampers ion diffusion and, thus, Copyright © 2011 S. Karger AG, Basel
reduces further progression of the lesion. Topical fluorida-
tion inducing the formation of a protective layer on dental Dental erosion is defined as substance loss by ex-
hard tissue, which is composed of CaF2 (in case of conven- ogenous or endogenous acids without bacteri-
tional fluorides like amine fluoride or sodium fluoride) or al involvement. The most important sources are
of metal-rich surface precipitates (in case of titanium tet- dietary acids [1] and those originated from the
rafluoride or tin-containing fluoride products), appears stomach, like gastric acid from regurgitation and
to be most effective on enamel. In dentin, the preventive reflux disorders [2].
effect of fluorides is highly dependent on the presence In contrast to initial caries, enamel erosion
of the organic matrix. In situ studies have shown a higher is predominantly a surface phenomenon with a
protective potential of fluoride in enamel compared to centripetal bulk substance loss combined with
dentin, probably as the organic matrix is affected by enzy- a small partly demineralized surface layer with
matical and chemical degradation as well as by abrasive decreased microhardness (fig. 1). In dentin, the
influences in the clinical situation. There is convincing erosive demineralization is mostly diffusion con-
evidence that fluoride, in general, can strengthen teeth trolled, as the increasing exposure of organic ma-
against erosive acid damage, and high-concentration flu- trix hampers ion diffusion, and thus reduces fur-
oride agents and/or frequent applications are considered ther progression of dentin erosion (fig. 2) [3, 4].

A.C.M. and A.W. contributed equally.

a b

Fig. 1. Scanning electron microscopy (a) and clinical picture (b) of enamel erosion. Images are not from the same

a b

Fig. 2. a Scanning electron microscopy of dentin erosion showing opened dentinal tubules; however, the tubules also
can be partially or totally closed in the clinical situation. Reprinted from Kato et al. [59], with permission. b Clinical pic-
ture of dentin erosion. Images are not from the same tooth.

Fluorides and Erosion 159

Fig. 3. Scanning electron microsco-
py of enamel treated with conven-
tional fluoride (AmF, 0.5 M fluoride,
pH 4.5, applied for 60 s).

There is evidence that the prevalence of ero- aprismatic enamel, the demineralization is irreg-
sion is steadily increasing [5]. Preventive strate- ular, without a clear structural pattern. If the ero-
gies in the management of dental erosion consid- sive challenge is ongoing, the dissolution process
er dietary counseling, stimulation of salivary flow, results in surface loss accompanied by a progres-
modification of erosive beverages, adequate oral sive softening of the surface. As the demineral-
hygiene measures and fluoride treatment as the ized layer of eroded enamel is considerably small
most relevant [6]. when compared to the enamel loss, fluoride ap-
This chapter will give an overview of the cur- plication predominately aims to prevent erosive
rent knowledge on the use of fluorides, includ- tissue loss rather than to remineralize softened
ing conventional and metal fluorides, for the pre- enamel.
vention of erosive and combined erosive-abrasive Conventional fluorides, whose beneficial
dental loss. Due to the fact that the histology of effect against caries is well known [7], have
enamel and dentin erosion is considerably differ- been tested for prevention or control of dental
ent, this chapter will be divided into two parts: erosion [8]. The potential of conventional fluo-
(1) fluorides and enamel erosion; (2) fluorides rides, such as sodium fluoride (NaF) and amine
and dentin erosion. fluoride (AmF), to prevent erosive demineral-
ization is mainly related to the formation of a
calcium fluoride (CaF2) layer [9, 10] (fig. 3).
Fluorides and Enamel Erosion This layer is assumed to behave as a physical
barrier that hampers the contact of the acid with
Extrinsic and/or intrinsic acids with low pH (pH the underlying enamel or to act as a mineral
1.0–3.5) initially cause either the dissolution of reservoir which is attacked by the erosive chal-
the prism cores or interprismatic areas, showing lenge. Thereafter, released calcium and fluoride
a honeycomb structure in prismatic enamel. In might increase the saturation level with respect

160 Magalhães · Wiegand · Rios · Buzalaf · Lussi

Fig. 4. Illustration of enamel treated with conventional fluoride. a Enamel surface. b Deposition
of a CaF2 layer. c CaF2 layer acting as a physical barrier for the erosive challenge. d Progressive CaF2
layer dissolution.

to dental hard tissue in the liquid adjacent to increasing concentration and frequency of appli-
the surface, thus promoting remineralization cation and decreasing pH of the agent. Fluoride
(fig. 4, 5). agents with a pH below 5 seem to induce a higher
The formation of the CaF2-like layer and its CaF2 deposition on dental surface than neutral
protective effect against demineralization is high- ones [9].
ly dependent on the pH, the concentration of flu- Ganss et al. [10] evaluated the retention
oride and the frequency of application. The de- of CaF2 on human enamel under neutral and
position of CaF2 on the surface increases with acidic conditions in vitro and in situ. Fluoride

Fluorides and Erosion 161

Fig. 5. Illustration of enamel treated with conventional fluoride. a CaF2 layer final dissolution. b
Simultaneous calcium and fluoride saturation provoking remineralization. c Subsequent erosive
challenge. d Bulk substance loss combined with a small partly demineralized surface layer.

(10,000 ppm, AmF) was applied once for 5 min, Although toothbrushing might affect the pro-
and the enamel specimens were exposed to ero- gression of eroded dental hard tissues adversely
sive demineralization (3 × 30 s/day, 4 days in vit- by removing the softened layer of enamel [11, 12],
ro; 3 × 2 min/day, 7 days in situ) or neutral con- it was shown that the use of fluoridated (NaF)
ditions (artificial saliva in vitro; human saliva in toothpastes might diminish the abrasive effect to
situ). It was shown that more CaF2 was lost under some extent [13–15]. However, as the overall pro-
erosive compared to neutral conditions in vitro, tective effect of toothpastes with 1,100–5,000 μg/g
while the intraoral environment was consider- fluoride is limited [14, 15], the use of highly con-
ably protective for CaF2-like precipitates, espe- centrated fluoride varnishes (22,600 μg/g) was an-
cially on enamel. ticipated to be more effective due to their capacity

162 Magalhães · Wiegand · Rios · Buzalaf · Lussi

Fig. 6. Scanning electron micros-
copy of enamel treated with 4%
titanium tetrafluoride varnish (6 h).
Reprinted from Magalhães et al. [19]
with permission.

to adhere to the tooth surface and create a CaF2 the titanium incorporation in the hydroxyapatite
reservoir [16, 17]. Indeed, the application of NaF lattice. The glaze-like surface layer observed af-
varnish (22,600 μg/g) was effective in reducing ter the application of TiF4 is assumed to be due to
enamel erosion for 30 min of acid exposure, but the formation of a new compound (hydrated hy-
the protective effect declined thereafter [18, 19]. drogen titanium phosphate) that might primar-
However, as placebo varnishes also showed some ily act as a diffusion barrier [23, 29–32] (fig. 6, 7).
protection against enamel erosion and combined The increased fluoride uptake found after appli-
erosion/abrasion, it is believed that the protective cation of TiF4 can be explained by the ability of
effect of fluoride varnishes is mainly related to the the polyvalent metal ion to form strong fluoride
mechanical rather than to the chemical protection complexes firmly bound to the apatite crystals
[20, 21]. [30, 32].
As the anti-erosive effect of conventional fluo- Information regarding the efficacy of TiF4 un-
rides requires a very intensive fluoridation regime der clinical conditions is scarce and contradicto-
[22], recent studies have focused on fluoride com- ry, as only two in situ studies showed 1.6% TiF4
pounds which might deliver a higher level of ef- (0.5 m fluoride) to be as effective as SnF2 or AmF
ficacy. In this context, compounds containing in the prevention of erosion or combined erosion/
polyvalent metal ions such as stannous fluoride abrasion [33, 34], while other did not show any
or titanium tetrafluoride were tested. protective effect of 4% TiF4 [20, 21, 35]. The effi-
Several in vitro studies have shown an inhib- cacy of TiF4 is highly dependent on the pH of the
itory effect of 0.4–10% TiF4 solution on dental agent, since it was shown that enamel erosion can
erosion [23–27], which is attributed not only to be significantly reduced by TiF4 (0.5 m fluoride)
the effect of fluoride, but mainly to the action at native pH (pH 1.2) but not at a pH buffered to
of titanium [23, 28]. Its protective effect is re- 3.5 [36]. One study indicated that TiF4 applied in
lated to the formation of an acid-resistant sur- the form of a varnish might be of higher effica-
face coating, the increased fluoride uptake and cy than as a solution [19]. However, it should be

Fluorides and Erosion 163

Fig. 7. Illustration of the formation of (CaF2) layer and an acid-resistant surface coating composed
of hydrated hydrogen titanium phosphate after the application of TiF4, or composed of metal-
rich precipitates [Ca(SnF3)2, SnOHPO4, Sn3F3PO4] after the application of tin-containing fluoride
mouth rinses. a CaF2 layer and the metal-rich precipitates (in orange). b Erosive challenge. c CaF2
layer dissolution. d CaF2 layer final dissolution and the preservation of the metal-rich precipitate. e
Progressive erosive challenges. f Final dissolution of the metal-rich layer and consequent enamel

164 Magalhães · Wiegand · Rios · Buzalaf · Lussi

Fig. 8. Scanning electron microsco-
py of enamel treated with SnF2 so-
lution (0.48 M, pH 2.7, 3 min) before
erosion. Reprinted from Yu et al. [60]
with permission.

Fig. 9. Scanning electron micros-

copy of enamel treated with SnF2
solution after erosion (6 × 1 min/
day, 5 days), showing no alteration.
Reprinted from Yu et al. [60] with

considered that the low pH of TiF4 products does rich surface precipitates [Ca(SnF3)2, SnOHPO4,
not allow self-application by the patient. Sn3F3PO4], which were shown to be of high acid
Tin-containing fluoride products have shown resistance [42] (fig. 7–9). Further, tin may pene-
promising results in several studies [37–41]. The trate and become incorporated into the deminer-
mode of action of tin-containing fluoride solutions alized layer when high concentrated tin contain-
is probably attributed to the formation of metal- ing fluoride mouth rinses are used [38, 43].

Fluorides and Erosion 165

Ganss et al. [44] evaluated the relevance of cat- (0.025% fluoride, AmF/NaF) and gel (1.25% fluo-
ions in different fluoride compounds for their ef- ride, AmF/NaF) was most effective in the preven-
fectiveness as anti-erosive agents and showed that tion of dentin erosion [22, 49]. However, after en-
SnCl2 (800 ppm tin), NaF (250 ppm fluoride), zymatic removal of the organic matrix, fluoride
AmF/SnF2 (250 ppm fluoride/390 ppm tin) and was ineffective [3, 50]. It was assumed that the
SnF2 (250 ppm fluoride/809 ppm tin) solutions demineralized organic dentin matrix has a buff-
could reduce enamel erosion. Treatment with so- ering capacity sufficient to prevent further den-
lutions containing SnF2 was most effective. The tin demineralization, especially in the presence of
combination of AmF/NaF/SnCl2 with high (2,800 high amounts of fluoride [3]. Moreover, the ex-
ppm tin/1,500 ppm fluoride) and low (700 ppm posed organic matrix of etched dentin involves
Sn/1,500 ppm fluoride) tin concentrations re- an increased surface area and increased diffusion
duced erosion by 90 and 70%, respectively [38, pathways – enhancing the amount of structural-
39]. ly bound and KOH-soluble fluoride compared to
Some possible side effects of high-concentra- sound dentin [51]. However, it remains unclear to
tion tin-containing mouth rinses may include a what extent the organic material is retained under
dull feeling on the tooth surface, astringent sen- clinical conditions, when the collagen layer might
sation and tooth discoloration (1,900 ppm tin) be affected by enzymatical and chemical degra-
[45]. Therefore, tin-containing solutions of low- dation [50, 52] as well as by abrasive influences.
er concentration (800 ppm tin/ 500 ppm fluo- From the clinical appearance of dentin-erosive
ride) were tested in vitro and in situ [46, 47]. lesions, it seems likely that the collagenous layer
Under severe erosive conditions, the SnCl2/NaF/ is at least partly removed. This hypothesis might
AmF exhibited a high potential to reduce enamel also explain why fluorides such as NaF were less
erosion (67% reduction), and showed no adverse effective in dentin than in enamel under in situ
side effects [47]. Besides mouth rinses, tin-con- conditions [10, 22, 38] but not in laboratory ex-
taining fluoride toothpastes were tested using periments [27, 53].
in vitro protocols, and shown to perform sig- The application of slightly acidic fluoride for-
nificantly better under erosive challenges when mulations such as NaF or AmF results in the for-
compared with NaF- and MFP-containing tooth- mation of CaF2 precipitates on both enamel and
pastes [41]. Further research should test spe- dentin (fig. 10), but the precipitates are less stable
cially formulated tin-containing fluoride prod- on dentin than on enamel under erosive condi-
ucts to minimize aesthetic negatives seen with tions [10]. Although the preventive potential of
high-concentration tin-containing products, NaF and AmF solution and dentifrice on den-
which may provide a highly effective means to tin erosion and combined erosion/abrasion was
help prevent dental erosion using a consumer- shown in different in situ studies [22, 34, 54], in-
friendly approach. formation about the ideal fluoride concentration
and frequency of application is scarce. Also, the
resistance of dentinal CaF2 precipitates against
Fluorides and Dentin Erosion abrasion has not so far been assessed directly;
only an in situ study indicated that the protective
The preventive effect of fluorides on dentin ero- potential of AmF against erosion is not affected by
sion is highly dependent on the presence of the additional brushing treatment [34].
organic matrix [48]. Initial studies showed that Considering the severe and chronic acid ex-
a very intensive fluoridation regimen combining posure in patients suffering from dental erosion,
toothpaste (0.15% fluoride, NaF), mouth rinse the effect of CaF2 precipitates is probably limited

166 Magalhães · Wiegand · Rios · Buzalaf · Lussi

Fig. 10. Scanning electron micros-
copy of dentin treated with conven-
tional fluoride (AmF, 0.5 M F, pH 4.5,
applied for 60 s).

Fig. 11. Scanning electron micros-

copy of dentin treated with 4% tita-
nium tetrafluoride varnish (6 h).

over time [10], and fluoride compounds with a its protective potential did not exceed the efficacy
distinct potential to resist an erosive challenge are of NaF or AmF [27, 34, 56], and the low pH re-
required. quired for the efficacy of the agents has not so far
Titanium tetrafluoride was shown to induce allowed for a clinical application [57].
some coating on dentin surfaces, which partly Tin-containing fluoride solutions have been
covered dentinal tubules [55] (fig. 11). However, demonstrated to exhibit promising anti-erosive

Fluorides and Erosion 167

effects not only on enamel but also on dentin effect of the SnF2- and AmF/SnF2-containing
[38, 44, 46]. The suggested mechanism of ac- solutions compared to all other solutions.
tion is related to the incorporation of tin in
mineralized dentin when the organic matrix
is allowed to develop and to surface precipita- Conclusion
tion when the organic matrix is enzymatically
removed [58]. In cases where the organic ma- Conventional fluorides with a known anti-cario-
trix is preserved, phosphorus, phosphorylated genic potential offer some, but limited, protection
phosphoprotein or phosphophoryn might at- against erosion as the CaF2 precipitates formed
tract the tin ion, which is then retained in the on the surface are readily soluble in acids. Metal-
organic matrix to some extent, but also accu- containing fluoride compounds showed promis-
mulates in the underlying mineralized tissue. ing results in prevention of erosion, but might in-
In cases where the organic matrix is removed, volve some adverse side effects due to the very low
tin reacts with the mineral by forming differ- pH (in case of titanium tetrafluoride) and the po-
ent salts, e.g. Sn(OH)2, Sn2(PO4)OH, Ca(SnF3), tential to cause slight discoloration, a dull feeling
Sn3F3PO4, Sn2(OH)PO4, Sn3F3PO4 or SnHPO4 on the tooth surface and an astringent sensation
[58]. Recent in situ studies demonstrated that (in case of highly concentrated tin-containing flu-
mouth rinses containing AmF/NaF/SnCl2 (500 oride solutions).
ppm F, 800 ppm Sn) reduced dentin erosion by There is convincing evidence that fluoride, in
50% and were significantly more effective than general, can strengthen enamel against erosive
an NaF-containing mouth rinse (500 ppm F) acid damage; high-concentration fluoride agents
[38, 47]. and/or frequent applications are considered po-
Comparing the protective effect of different tentially effective approaches to prevent dental
fluoride compounds on dentin erosion, Ganss et erosion. However, fluorides might be more effec-
al. [48] showed that solutions containing AmF tive in enamel than in dentin, as the organic ma-
and/or SnF2 performed only slightly better than trix influencing the efficacy of fluorides might to
solutions containing NaF and/or AmF in the pres- some extent be affected by enzymatical and chem-
ence of the organic matrix. However, continuous ical degradation as well as by mechanical abrasion.
removal of the organic matrix influenced the ef- The use of tin-containing fluoride products might
ficacy of the fluoride compounds distinctly and provide the best approach for effective prevention
demonstrated a significantly better preventive of dental erosion.

1 Lussi A, Jaeggi T, Zero D: The role of diet 4 Hara AT, Ando M, Cury JA, Serra MC, 6 Magalhães AC, Wiegand A, Rios D, Hon-
in the aetiology of dental erosion. Caries Gonzalez-Cabezas C, Zero DT: Influence orio HM, Buzalaf MA: Insights into pre-
Res 2004;38(suppl 1):34–44. of the organic matrix on root dentine ventive measures for dental erosion.
2 Bartlett D: Intrinsic causes of erosion. erosion by citric acid. Caries Res 2005; J Appl Oral Sci 2009;17:75–86.
Monogr Oral Sci 2006;20:119–139. 39:134–138. 7 ten Cate JM: Review on fluoride, with
3 Ganss C, Klimek J, Starck C: Quantitative 5 Lussi A: Erosive tooth wear – a multifac- special emphasis on calcium fluoride
analysis of the impact of the organic torial condition of growing concern and mechanisms in caries prevention.
matrix on the fluoride effect on erosion increasing knowledge. Monogr Oral Sci Eur J Oral Sci 1997;105:461–465.
progression in human dentine using 2006;20:1–8. 8 Wiegand A, Attin T: Influence of fluo-
longitudinal microradiography. Arch ride on the prevention of erosive lesions
Oral Biol 2004;49:931–935. – a review. Oral Health Prev Dent 2003;

168 Magalhães · Wiegand · Rios · Buzalaf · Lussi

9 Saxegaard E, Rolla G: Fluoride acquisi- 20 Vieira A, Ruben JL, Huysmans MC: 32 Wiegand A, Magalhães AC, Attin T: Is
tion on and in human enamel during Effect of titanium tetrafluoride, amine titanium tetrafluoride (TiF4) effective to
topical application in vitro. Scand J Dent fluoride and fluoride varnish on enamel prevent carious and erosive lesions? A
Res 1988;96:523–535. erosion in vitro. Caries Res 2005;39:371– review of the literature. Oral Health Prev
10 Ganss C, Schlueter N, Klimek J: Reten- 379. Dent 2010;8:159–164.
tion of KOH-soluble fluoride on enamel 21 Vieira A, Lugtenborg M, Ruben JL, 33 Hove LH, Holme B, Young A, Tveit AB:
and dentine under erosive conditions – a Huysmans MC: Brushing abrasion of The protective effect of TiF4, SnF2 and
comparison of in vitro and in situ eroded bovine enamel pretreated with NaF against erosion-like lesions in situ.
results. Arch Oral Biol 2007;52:9–14. topical fluorides. Caries Res 2006;40: Caries Res 2008;42:68–72.
11 Attin T, Knofel S, Buchalla W, Tutuncu 224–230. 34 Wiegand A, Hiestand B, Sener B, Magal-
R: In situ evaluation of different remin- 22 Ganss C, Klimek J, Brune V, Schurmann hães AC, Roos M, Attin T: Effect of
eralization periods to decrease brushing A: Effects of two fluoridation measures TiF(4), ZrF(4), HfF(4) and AmF on ero-
abrasion of demineralized enamel. Car- on erosion progression in human sion and erosion/abrasion of enamel and
ies Res 2001;35:216–222. enamel and dentine in situ. Caries Res dentin in situ. Arch Oral Biol 2010;55:
12 Rios D, Honorio HM, Magalhães AC, 2004;38:561–566. 223–228.
Delbem AC, Machado MA, Silva SM, 23 Buyukyilmaz T, Ogaard B, Rolla G: The 35 Magalhães AC, Rios D, Honorio HM,
Buzalaf MA: Effect of salivary stimula- resistance of titanium tetrafluoride- Jorge AM Jr, Delbem AC, Buzalaf MA:
tion on erosion of human and bovine treated human enamel to strong hydro- Effect of 4% titanium tetrafluoride solu-
enamel subjected or not to subsequent chloric acid. Eur J Oral Sci 1997;105: tion on dental erosion by a soft drink: an
abrasion: an in situ/ex vivo study. Caries 473–477. in situ/ex vivo study. Arch Oral Biol
Res 2006;40:218–223. 24 van Rijkom H, Ruben J, Vieira A, Huys- 2008;53:399–404.
13 Bartlett DW, Smith BG, Wilson RF: mans MC, Truin GJ, Mulder J: Erosion- 36 Wiegand A, Waldheim E, Sener B,
Comparison of the effect of fluoride and inhibiting effect of sodium fluoride and Magalhães AC, Attin T: Comparison of
non-fluoride toothpaste on tooth wear in titanium tetrafluoride treatment in vitro. the effects of TiF4 and NaF solutions at
vitro and the influence of enamel fluo- Eur J Oral Sci 2003;111:253–257. pH 1.2 and 3.5 on enamel erosion in
ride concentration and hardness of 25 Hove L, Holme B, Ogaard B, Willumsen vitro. Caries Res 2009;43:269–277.
enamel. Br Dent J 1994;176:346–348. T, Tveit AB: The protective effect of TiF4, 37 Schlueter N, Klimek J, Ganss C: Effect of
14 Magalhães AC, Rios D, Delbem AC, SnF2 and NaF on erosion of enamel by stannous and fluoride concentration in a
Buzalaf MA, Machado MA: Influence of hydrochloric acid in vitro measured by mouth rinse on erosive tissue loss in
fluoride dentifrice on brushing abrasion white light interferometry. Caries Res enamel in vitro. Arch Oral Biol 2009;54:
of eroded human enamel: an in situ/ex 2006;40:440–443. 432–436.
vivo study. Caries Res 2007;41:77–79. 26 Hove LH, Young A, Tveit AB: An in vitro 38 Schlueter N, Klimek J, Ganss C: Efficacy
15 Rios D, Magalhães AC, Polo RO, Wie- study on the effect of TiF(4) treatment of an experimental tin-F-containing
gand A, Attin T, Buzalaf MA: The effi- against erosion by hydrochloric acid on solution in erosive tissue loss in enamel
cacy of a highly concentrated fluoride pellicle-covered enamel. Caries Res and dentine in situ. Caries Res 2009;43:
dentifrice on bovine enamel subjected to 2007;41:80–84. 415–421.
erosion and abrasion. J Am Dent Assoc 27 Schlueter N, Ganss C, Mueller U, Klimek 39 Schlueter N, Klimek J, Ganss C: In vitro
2008;139:1652–1656. J: Effect of titanium tetrafluoride and efficacy of experimental tin- and fluo-
16 Sorvari R, Meurman JH, Alakuijala P, sodium fluoride on erosion progression ride-containing mouth rinses as anti-
Frank RM: Effect of fluoride varnish and in enamel and dentine in vitro. Caries erosive agents in enamel. J Dent 2009;37:
solution on enamel erosion in vitro. Car- Res 2007;41:141–145. 944–948.
ies Res 1994;28:227–232. 28 Tezel H, Ergucu Z, Onal B: Effects of 40 Hooper SM, Newcombe RG, Faller R,
17 Vieira A, Jager DH, Ruben JL, Huysmans topical fluoride agents on artificial Eversole S, Addy M, West NX: The pro-
MC: Inhibition of erosive wear by fluo- enamel lesion formation in vitro. Quin- tective effects of toothpaste against ero-
ride varnish. Caries Res 2007;41:61–67. tessence Int 2002;33:347–352. sion by orange juice: studies in situ and
18 Magalhães AC, Stancari FH, Rios D, 29 Mundorff SA, Little MF, Bibby BG: in vitro. J Dent 2007;35:476–481.
Buzalaf MA: Effect of an experimental Enamel dissolution. II. Action of tita- 41 Faller RV, Eversole SL, Tzeghai GE:
4% titanium tetrafluoride varnish on nium tetrafluoride. J Dent Res Enamel Protection: A Comparison of
dental erosion by a soft drink. J Dent 1972;51:1567–1571. Marketed Dentifrice Performance
2007;35:858–861. 30 Wei SH, Soboroff DM, Wefel JS: Effects Against Dental Erosion. Am J Dent 2011,
19 Magalhães AC, Kato MT, Rios D, Wie- of titanium tetrafluoride on human in press.
gand A, Attin T, Buzalaf MA: The effect enamel. J Dent Res 1976;55:426–431. 42 Babcock FD, King JC, Jordan TH: The
of an experimental 4% Tif4 varnish com- 31 Ribeiro CC, Gibson I, Barbosa MA: The reaction of stannous fluoride and
pared to NaF varnishes and 4% TiF4 uptake of titanium ions by hydroxyapa- hydroxyapatite. J Dent Res 1978;57:
solution on dental erosion in vitro. Car- tite particles-structural changes and 933–938.
ies Res 2008;42:269–274. possible mechanisms. Biomaterials

Fluorides and Erosion 169

43 Schlueter N, Hardt M, Lussi A, 48 Ganss C, Lussi A, Sommer N, Klimek J, 55 Wiegand A, Magalhães AC, Navarro RS,
Engelmann F, Klimek J, Ganss C: Tin- Schlueter N: Efficacy of fluoride com- Schmidlin PR, Rios D, Buzalaf MA, Attin
containing fluoride solutions as anti- pounds and stannous chloride as erosion T: Effect of titanium tetrafluoride and
erosive agents in enamel: an in vitro tin- inhibitors in dentine. Caries Res 2010; amine fluoride treatment combined with
uptake, tissue-loss, and scanning 44:248–252. carbon dioxide laser irradiation on
electron micrograph study. Eur J Oral Sci 49 Ganss C, Klimek J, Schaffer U, Spall T: enamel and dentin erosion. Photomed
2009;117:427–434. Effectiveness of two fluoridation mea- Laser Surg 2010;28:219–226.
44 Ganss C, Schlueter N, Hardt M, Schat- sures on erosion progression in human 56 Magalhães AC, Levy FM, Rios D, Buzalaf
tenberg P, Klimek J: Effect of fluoride enamel and dentine in vitro. Caries Res MA: Effect of a single application of
compounds on enamel erosion in vitro: 2001;35:325–330. TiF(4) and NaF varnishes and solutions
a comparison of amine, sodium and 50 Schlueter N, Ganss C, Hardt M, Schegi- on dentin erosion in vitro. J Dent 2010;
stannous fluoride. Caries Res 2008;42: etz D, Klimek J: Effect of pepsin on ero- 38:153–157.
2–7. sive tissue loss and the efficacy of fluori- 57 Wiegand A, Magalhães AC, Sener B,
45 Schlueter N, Klimek J, Ganss C: Efficacy dation measures in dentine in vitro. Acta Waldheim E, Attin T: TiF(4) and NaF at
of tin-containing solutions on erosive Odontol Scand 2007;65:298–305. pH 1.2 but not at pH 3.5 are able to
mineral loss in enamel and dentine in 51 Buchalla W, Lennon AM, Becker K, reduce dentin erosion. Arch Oral Biol
situ. Clin Oral Investig 2010, E-pub Lucke T, Attin T: Smear layer and sur- 2009;54:790–795.
ahead of print. face state affect dentin fluoride uptake. 58 Ganss C, Hardt M, Lussi A, Cocks AK,
46 Schlueter N, Neutard L, von Hinckeldey Arch Oral Biol 2007;52:932–937. Klimek J, Schlueter N: Mechanism of
J, Klimek J, Ganss C: Tin and fluoride as 52 Schlueter N, Hardt M, Klimek J, Ganss action of tin-containing fluoride solu-
anti-erosive agents in enamel and den- C: Influence of the digestive enzymes tions as anti-erosive agents in dentine –
tine in vitro. Acta Odontol Scand 2010, trypsin and pepsin in vitro on the pro- an in vitro tin-uptake, tissue loss, and
E-pub ahead of print. gression of erosion in dentine. Arch Oral scanning electron microscopy study. Eur
47 Ganss C, Neutard L, von Hinckeldey J, Biol 2010;55:294–299. J Oral Sci 2010;118:376–384.
Klimek J, Schlueter N: Efficacy of a tin/ 53 Wiegand A, Meier W, Sutter E, Magal- 59 Kato MT, Leite AL, Hannas AR, Oliveira
fluoride rinse: a randomized in situ trial hães AC, Becker K, Roos M, Attin T: RC, Pereira JC, Tjäderhane L, Buzalaf
on erosion. J Dent Res 2010;89: Protective effect of different tetrafluo- MA: Effect of iron on matrix metallopro-
1214–1218. rides on erosion of pellicle-free and pel- teinase inhibition and on the prevention
licle-covered enamel and dentine. Caries of dentine erosion. Caries Res
Res 2008;42:247–254. 2010;44:309–316.
54 Magalhães AC, Rios D, Moino AL, Wie- 60 Yu H, Attin T, Wiegand A, Buchalla W:
gand A, Attin T, Buzalaf MA: Effect of Effects of various fluoride solutions on
different concentrations of fluoride in enamel erosion in vitro. Caries
dentifrices on dentin erosion subjected Res 2010;44:390-401.
or not to abrasion in situ/ex vivo. Caries
Res 2008;42:112–116.

Prof. Ana Carolina Magalhães

Bauru Dental School, University of São Paulo
Al. Dr. Octávio Pinheiro Brisolla 9–75
Bauru-SP 17012–190, Brazil
Tel. +55 14 3235 8247, E-Mail

170 Magalhães · Wiegand · Rios · Buzalaf · Lussi

Author Index

Buzalaf, M.A.R. 1, 20, 37, 52, 97, Magalhães, A.C. 158 ten Cate, J.M. 97
115, 158 Toumba, K.J. 115
Pessan, J.P. 52, 97, 115
DenBesten, P. 81 Villa, A.E. 37
Rios, D. 158 Vogel, G.L. 146
Honório, H.M. 97 Robinson, C. XI
Rugg-Gunn, A.J. 37 Whitford, G.M. 20, 66
Levy, S.M. 1, 133 Wiegand, A. 158
Li, W. 81 Sampaio, F.C. 133
Lussi, A. 158

Subject Index

Abrasive systems, dentifrices 123 Bacteria

Absorption 23, 24, 38, 78 fluoride binding model 147, 148
Accidental poisonings 66, 67, 69, 70, see also Acute fluoride effects 108–110
toxicity mineral dynamics 101, 102, 107
Acid-base status, see also pH Beikost 13
metabolism effects 28, 29 Beverages, infant 13, 84
toxicity effects 77, 78 Biofilm 99, 102
Acidulated phosphate fluoride 71, 72, 75 Biological F binding sites 147, 148
Acute toxicity 66–80 Biomarkers
defined 78 classification 38, 48, 53, 63
factors influencing 75–78 contemporary
gastric effects 72–75 blood plasma 39–41, 48
probably toxic dose 67, 68, 71, 78 bone surface 27, 39–42, 48
source 68, 69, 85, 86 early investigations 38, 39
dental products 67, 69–72, 78, 79 human milk 43, 44
treatment 24, 26, 68, 78 individual versus group 46, 47
Administration methods, see Delivery methods saliva 42, 43, 48, 149, 150
Age factors, see also Children sweat 43
dental fluorosis 85–87 urine 44–46, 48, 49
recommended intake 84 defined 53
toxicity effects 76, 77 historical
Alkalosis 78 bone 61, 62
Altitude 30 teeth 62, 63, 103
Ameloblasts 89–91, 93 recent
Amelogenins 90–93 hair 60, 61
American Academy of Pediatric Dentistry 130 nails 53–60
American Dental Association Blood plasma
dietary supplement guidelines 11, fluoride distribution 25
141, 142 sample collection 39, 48
tablet guidelines 71 Blood plasma concentration
topical fluoride guidelines 130 after ingestion 22, 23
water fluoridation statement 136 biomarker 39–41, 48
Amine fluoride 117, 160, 168 factors influencing 77, 82
Ammonium fluoride 117 pharmacokinetics 25
Amorphous calcium phosphate agents 152 rhythmicity 31
Apatite 91–93, 98–101 total 25

Body burden 38, 41, 62 fluoride intake 1–19
Body fluids, specialized, fluoride distribution ‘optimal’ 2, 83, 84
26 sources 4–13
Body tissues, fluoride distribution to 25–27 renal impairment 29, 30
Bone systemic methods 84, 140–142
fluoride metabolism 23, 26, 27, 62, 77 topical methods 124, 125, 130
fluoride reservoirs 38, 41, 87 Child-resistant caps 70, 79
sample collection 41, 42, 62 Chronic toxicity, see Dental fluorosis
types 41, 42, 62 Circadian rhythm 30, 31
Bone biomarkers Cochrane Central Register of Controlled Trials 116
contemporary, bone surface 27, 39–42, 48 Collagen 99, 166
historical 61, 62 Community delivery methods 115, 116, 143
Bottled water 6, 136 milk fluoridation 135, 140, 141
Buffer extraction, plaque fluoride reservoirs 151 salt fluoridation 135, 138–140
water, see Water fluoridation
Calcified tissues, see also Mineral dynamics Community level biomarkers 46, 47
fluoride distribution 23, 26, 27, 38, 62 Compact bone 41, 42, 62
Calcium fluoride Compounds, see also specific compound
dental erosion 160–164 dental erosion 163–168
oral reservoirs 105–107, 110, 147, 149, 152 systemic fluoride 135
plaque reservoirs 110, 147, 151, 153, 154 topical fluoride 116, 117, 126, 127
Calcium fluoride-like material 104, 147, 149, 151, toxicity effects 75, 76
154, 155 Concentration, see also Biomarkers
Calcium, fluoride metabolism 24, 28, 32, 68 dentifrices 124–126
Calcium phosphate agents, amorphous 152 systemic methods 135, 139, 140
Calcium pre-rinse systems 153–155 toxicity effects by 73
Calculus formation 101 Consumer Product Safety Commission 70
Canadian Dental Association 11 Controlled release rinses 152, 153
Cancellous bone 41, 42, 62 Copolymer membrane device 128
Caries control Cortical bone 39
fluoride mechanisms of action 102–111, 134 Costs
oral fluoride reservoirs, see Oral fluoride professional topical methods 129
reservoirs salt fluoridation 138
systemic fluoride 107, 108, 136, 137 water fluoridation 135, 136
topical fluoride, see Topical fluoride Critical pH 101
Caries development Crystal surface fluoride 104, 105, 108
biochemistry 98–102
etiology 97, 98 Daily urinary fluoride excretion 45, 46, 48
plaque F release 151 DDE index 31
Cerebrospinal fluid, fluoride distribution 26 Dean, Trendley 134
Chemical compounds, see Compounds Dean fluorosis index 83, 84
Children Delivery methods 115, 116, 143
acute toxicity 67, 69–71, 76, 77, 79 community, see Community delivery methods
biomarkers professional 116, 121, 122, 143, see also Topical
hair 61 fluoride
nails 59, 60 self-applied 116, 143, see also Topical fluoride
teeth 63 systemic administration, see Systemic fluoride
urine 45–48 topical administration, see Topical fluoride
dental fluorosis, see Dental fluorosis Demineralization, see Mineral dynamics

Subject Index 173

Dental erosion 158–170 child-resistant caps 70, 79
defined 158 compounds and concentrations 135, 141
dentin 158, 159, 166–168 Disodium hexafluorosilicate 136
enamel 159–166 Disodium monofluorophosphate 24, 75, 140
prevention 160 Dissociated fluoride, see Ionic fluoride
Dental fluorosis 81–96 Distribution 23, 25–27, 38, 62
children 1–19 DMFS pooled estimates 129
biomarkers predicting 59–61, 63 Drinking water fluoridation, see Water fluoridation
epidemiology 85–87 Drops 141
nutritional status 31 Ductal saliva 42, 43
renal impairment 30
window of maximum susceptibility 2–4 Enamel
clinical manifestations 82, 83 altitude effects 30
etiology 82, 84, 85, 136 biomarkers 62, 63, 103
genetic susceptibility 32, 33, 82, 86–88 composition 98, 99
history 133, 134 dissolution 100, 101
indices 31, 83, 84 erosion 158–166
pathology 86–93 fluoride distribution to 26, 27
prevalence 5, 85, 86 fluorosis 81, 82, 86–93, see also Dental fluorosis
prevention 84 mineral dynamics 92, 93, 98–100, 111, 112
risk factors 1, 2, 14 Enamel fluid fluoride 104–106, 108
treatment 84 Environmental sources 60, 61
Dental products, see Topical fluoride, see also specific Erosion, see Dental erosion
product European Academy of Paediatric Dentistry 130
Dentifrices 122–127 Excretion
acute toxicity 69–72, 75 sweat 23
children’s fluoride intake 6–10 urinary, see Urinary excretion
clinical efficacy 119, 124–127, 129 Exercise 30
composition 123, 126, 127 Exogenous sources 60
dental erosion 162, 163, 166 Extracellular fluids 41, 78
fluoride concentration 75, 124–126 Extraction techniques, plaque fluoride reservoirs 151
low-fluoride 124, 125
pH 126 F binding sites 147, 148
plaque fluoride reservoirs 151, 152 Fecal fluoride 23, 28
Dentin Fingernails, biomarker 53–60
biomarkers 62, 63 Fluorapatite 146, 147, 149, 150
composition 98, 99 Fluorhydroxyapatite 101–103
erosion 158, 159, 166–168 Fluoride
fluoride distribution to 26, 27 administration methods, see Delivery methods
fluorosis 86, 87 cariostatic benefits, see Caries control
mineral dynamics 98, 99, 111, 112 compounds, see Compounds
Diet excretion, see Excretion
biomarker effects 48 exposure, biological markers, see Biomarkers
dental fluorosis 87 intake, see Ingestion
fluoride metabolism effects 24, 29, 31, 32 metabolism, see Metabolism
Dietary acids 158 sources, see Sources, see also specific source
Dietary reference intakes 84 toxicity
Dietary supplements 141, 142 acute, see Acute toxicity
children’s intake 10, 11, 84, 142 chronic, see Dental fluorosis

174 Subject Index

Fluoridens 133 Ingestion, see also Systemic fluoride
Fluorosilicic acid 136 acute toxicity, see Acute toxicity
Fluorosis, see Dental fluorosis plasma concentration curve 22, 23
Foams 118 Inorganic fluoride
acute toxicity 71, 122 compounds 116, 117
clinical efficacy 122 natural occurrence 20, 85
Free saliva 149, 150 In situ de/remineralization test protocol 152, 153
Institute of Medicine, US Food and Nutrition
Gastrointestinal tract Board 84
absorption 23, 24, 38, 78 Intracellular concentration 78
toxicity effects 72–76 Intraoral fluoride reservoirs, see Oral fluoride
Gels 118 reservoirs
acute toxicity 71, 72, 75, 122 Ion activity product 99, 100, 147
application guidelines 121 Ionic fluoride 21, 25, 27
clinical efficacy 119, 122, 129 blood plasma 39
Gender ground waters 85
biomarkers 58, 59, 62 human milk 44
dental fluorosis 87 renal clearance 27, 28
Genetic factors 32, 33, 82, 86–88
Geographical area, nails, biomarkers 58, 59 Kallikrein 4 90, 91
Gingival crevicular fluid fluoride levels 26 Kidneys, fluoride clearance 27–31, 44, 45, 78
Glass beads 128
Glomerular filtration rate 27, 28 Lethal dose 67, 68, 71, 78
Ground waters 85 Lozenges 133, 141
Group biomarkers 46, 47
Malnutrition 31
Hair biomarker 60, 61 Matrix proteins 90–93, 99
Healthy People 2010, 135 McKay, Frederick 133
Henderson-Hasselbalch equation 72, 73 Metabolic alkalosis 78
Hexafluorosilicic acid 136 Metabolism 20–36
Hexamethyldisiloxane-facilitated diffusion 57, absorption 23, 24
58 distribution 25–27
Hormones 30, 31 factors, modifying 28–33
Human Development Index 137, 138 fecal clearance 28
Human milk 26, 43, 44 general features 21–23
Hydrogen fluoride renal clearance 27–30, 44, 45, 78
metabolism 21, 24, 28, 29 Metal ions, polyvalent 163–165
toxicity 72–75, 78 Milk
urinary excretion 45 animal
Hydroxyapatite 91–93, 98–101, 107 calcium content 32, 68
Hydroxyapatite-Eudragit RS100 diffusion-controlled fluoridation 135, 140, 141
fluoride system 128 human 26, 43, 44
Hydroxyethyl methacrylate/methyl methacrylate Mineral dynamics
copolymer 128 biochemistry 99–101
Hypertension 139 caries formation 101–103
dental erosion 160, 161
Individual biomarkers 46, 47 dentin 98, 99, 111, 112
Infant foods, manufactured 13, 84 enamel 92, 93, 98–100, 111, 112
Infant formulas 6, 11–13, 84 enhancement 107, 152, 153

Subject Index 175

fluoride effects 92, 93, 104–107, 111, 112 fluoride reservoirs 108, 110, 111, 147, 150–155
saliva 99, 100, 107 Plasma, see Blood plasma
Mineralized tissues, fluoride distribution 23, Poisonings, accidental 66, 67, 69, 70, see also Acute
26, 27, 38, 62 toxicity
MMP-20 90, 91 Potassium fluoride 139
Monofluorophosphate-containing compounds 116, Pre-rinse systems, calcium 153–155
117 Probably toxic dose 67, 68, 71, 78
Mottled enamel, see Dental fluorosis Professional delivery methods 116, 121, 122, 143, see
Mousses 118 also Topical fluoride
Mouthwashes, see Rinses Proteinases 90, 91
Mucosa Proteolytic activity 90–92
fluoride reservoirs 110, 149, 150
toxicity effects 73–76 Reference dose 84
Remineralization, see Mineral dynamics
Nails, biomarker 53–60 Renal clearance 27–31, 44, 45, 78
National Brazilian Epidemiologic Survey 137 Rice-based diet 32
National Research Council, Committee of Rinses 122
Biomarkers 53 acute toxicity 69–72, 79
Natural sources 20, 85 calcium pre-rinse systems 153–155
Non-ionic fluoride 25, 39 clinical efficacy 120, 129
Nutritional status 31, 87, see also Diet controlled release 152, 153
oral fluoride reservoirs 152–155
‘Optimal’ fluoride intake 2, 83 plaque fluoride reservoirs 151, 152
Oral bacteria, see Bacteria tin-containing 166–168
Oral fluoride reservoirs
caries control 105–108, 110, 111 Saliva
enhancement 152–155 biomarkers 42, 43, 48, 149, 150
location 148–152 fluoride distribution 26
types 104, 105, 146–148 fluoride reservoirs 147, 149–155
Organic compounds 116, 117 mineral dynamics 99, 100, 107
Outer fluoride 104 slow-release fluoride devices 128
Overdose, see Acute toxicity Salmon thyrocalcitonin 30
Salt fluoridation 135, 138–140
Parathormone 30 Self-applied methods 116, 143, see also Topical
Perikymata 82 fluoride
pH Silane fluoride 117
critical 101 Skeleton, see Bone
dental erosion 160, 166, 167 Slow-release fluoride devices 128
dentifrice 126 Sodium fluoride
enamel formation 92 absorption 24
fluoride metabolism 21, 24–31, 45, 78 acute toxicity 66, 67, 71, 75
fluoride toxicity 73–75, 77, 78, 92 controlled release systems 153
mineral dynamics 99–103, 107, 111 dental erosion 160, 163, 166, 168
Pharmacokinetics 25, 62 dietary supplements 141
Phosphatases 75, 76 milk fluoridation 140
Physical activity 30 rinses 153, 154
Plaque salt fluoridation 139
fluid microenvironment 99, 100 topical formulations 117, 126, 127
fluoride concentration 125, 150 Sodium fluorosilicate 136

176 Subject Index

Sodium lauryl sulfate 151, 155 Toothbrushing habits 127, 162, 163
Sodium monofluorophosphate 117, 126, 127 Tooth formation
Soft tissues, fluoride distribution to 23, 25, 26 fluoride effects 2–4, 88–92
Solid phase fluoride 104, 108 mineralization, see Mineral dynamics
Solubility product 99, 147 renal impairment effects 29, 30
Solutions 72, 116–118 Toothpaste, see Dentifrices
Sorghum-based diet 32 Topical fluoride 115–132
Sources acute toxicity 67, 69–72, 78, 79
children’s intake 4–13 child-resistant caps 70, 79
dental products, see Topical fluoride, specific chronic toxicity 85, 86, see also Dental fluorosis
product combination modalities 128, 129
dietary, see Dietary supplements compounds 116, 117, 126, 127
environmental 60, 61 defined 116
infant foods/beverages 13, 84 dentifrices, see Dentifrices
infant formulas 6, 11–13, 84 gels, see Gels
milk 135, 140, 141 mechanisms of action 103–111, 134
salt 135, 138–140 mousses 118
toxicity 68, 69, 78, 79, 85, 86 reservoirs, see Oral fluoride reservoirs
water, see Water fluoridation rinses, see Rinses
Stannous fluoride 117, 163 slow-release devices 128
Steady-state distribution 25, 41 solutions 116–118
Stomach varnishes, see Varnishes
absorption 23, 24, 78 Total daily fluoride intake 45, 46, 48
toxicity effects 72–76 Total plasma fluoride 25
Stomach acids 158 Toxicity
Striae of Retzius 82 acute, see Acute toxicity
Sucrose 140 chronic, see Dental fluorosis
Sweat 23, 43 Trabecular bone 39
Systemic fluoride 133–145
acute toxicity, see Acute toxicity Underground water 85
caries control 107, 108, 136, 137 Urinary excretion
compounds 135 biomarker 44–46, 48, 49
sources, see Sources range of values 45, 47, 48
factors affecting 31, 32, 44, 45, 48
Tablets 71, 141 physiology 27, 28, 38
Tamarind 32 US Centers for Disease Control and Prevention 136
Tartaric acid 32 US Environmental Protection Agency 84
Therapeutic ratio 38, 48 US Food and Drug Administration 153
Thylstrup and Fejerskov Fluorosis Index 83, 84
Tin-containing fluoride products 165–168 Varnishes 120
Titanium tetrafluoride 117, 163–165, 167 acute toxicity 122
Toenails, biomarker 53–60 application guidelines 121
Tooth clinical efficacy 119, 122, 129
biomarkers 62, 63 for dental erosion 162, 163
caries control, see Caries control Vegetarian diet 29, 31, 32, 48
enamel, see Enamel
erosion, see Dental erosion Water
fluoride reservoirs 148, 149 bottled 6, 136
fluorosis, see Dental fluorosis underground 85

Subject Index 177

Water extraction, plaque fluoride reservoirs 151 nails, biomarker 58
Water fluoridation 135–138 opposition 136, 138
acute toxicity 68, 69 Water rinsing, plaque fluoride effects 150, 151
children’s intake 4–6 Whole saliva 42, 43, 149, 150
chronic toxicity 82, 85, 86, 136 World Health Organization
compounds and concentrations 135, 136 biomarker classification 53
costs 135, 136 biomarke