The Nature and Mechanisms of Dental Fluorosis in Man

O. FEJERSKOV, F. MANJP, and V. BAELUW

Department of Oral Anatomy, Dental Pathology and Operative Dentistry, 2Department of Periodontology and Oral Gerontology, and the WHO
Collaborating Centre for Oral Health Planning and Research in Third World Countries, The Royal Dental College, Vennelyst Boulevard, DK-
8000 Aarhus C, Denmark, and 1Kenya Medical Research Institute, Medical Research Centre, PO Box 20752, Nairobi, Kenya

Any use offluorides, whether systemic or topical, in caries prevention little was actually known about pathogenic mechanisms in den-
and treatment in children results in ingestion and absorption offluo- tal fluorosis (Fejerskov et al., 1977). At the same time, Myers
ride into the blood circulation. The mineralization of teeth under for- (1978), in a review on dose-response relationships in dental
mation may be affected so that dental fluorosis may occur. Dental fluorosis, stated that "while levels of 2 ppm of fluoride (in
fluorosis reflects an increasing porosity of the surface and subsurface
enamel, causing the enamel to appear opaque. The clinical features water) are associated only with the mildest form of dental
represent a continuum of changes ranging from fine white opaque fluorosis, a margin of safety of 2 can be regarded as impres-
lines running across the tooth on all parts of the enamel to entirely sively small". The statement reveals that the underlying as-
chalky white teeth. In the latter cases, the enamel may be so porous sumptions about dose-response relationship were strongly related
(or hypomineralized) that the outer enamel breaks apart posteruptively to the concept that an optimal dose of fluoride was well-de-
and the exposed porous subsurface enamel becomes discolored. These fined. A proper understanding of the nature and mechanisms
changes can be classified clinically by the TF index to reflect, in an of dental fluorosis necessitates a thorough understanding of
ordinal scale, the histopathological changes associated with dental how fluoride affects amelogenesis.
fluorosis. Compared with Dean's and the TSIF index, we consider
the TF index to be more precise. Recent studies on human enamel The aim of this paper is, therefore,
representing the entire spectrum ofdental fluorosis have demonstrated - to present what is known about the clinical features of
a clear association between increasing TF score and increasing fluo- dental fluorosis in humans and to discuss appropriate classi-
ride content of the enamel-. So far, no useful data on dose (expressed fication systems;
in mg fluoride/kg b. w.) -response (dental fluorosis) relationships are - to discuss the histopathology, biochemistry, and chem-
available. In this paper, we have, therefore, re-evaluated the original istry of human fluorotic enamel from the point of view of
data by Dean et aL (1941, 1942), Richards et aI. (1967), and Butler identification of possible pathogenic mechanisms; and
et aL (1985) from the USA, by applying the equation of Galagan and
Vermillion (1957) which permits the calculation of water intake as a - to reconsider established epidemiological data in order
function of temperature. By so doing, it can be demonstrated that to elucidate the nature of the dose-response relationship, and
there is a linear association between fluoride dose and dental fluorosis thereby derive the means for predicting the toxicological ef-
(r = 0.87). Even with very low fluoride intake from water, a certain fects to be expected when additional fluoride is provided to a
level of dental fluorosis will be found in a population. When the linear population.
dose-response curve is applied to previous data from the use offluo-
ride supplements, these data are in full accordance. This indicates
that we already have useful data available which to some extent allows Clinical features of dental fluorosis.
us to predict prevalence and severity offluorosis in a child population
which is exposed to a known amount offluoride. Because dental fluo- Much of the apparent confusion about the clinical appear-
rosis may occur in some individuals and populations to a higher prev- ance of fluoride-induced enamel defects stems from the use of
alence and degree than expected, and there exist rare cases who the term "mottled" enamel. Beyond doubt, fluoride is the one
exhibit clinical changes similar to those of fluorosis-but with no single factor most commonly responsible for causing enamel
known excessive fluoride background-it is concluded that it is im- "mottling". However, other factors, although relatively rare,
portant to intensify studies on factors which alone or in combination
can make individuals more or less susceptible to the effect offluoride. may cause opacities in the enamel (Small and Murray, 1978).
The term "mottled" enamel covers, therefore, a broader clin-
J Dent Res 69(Spec Iss):692-700, February, 1990 ical spectrum of lesions than does "dental fluorosis".
Dental fluorosis in humans has a very characteristic appear-
ance, in terms of both the single tooth surface and the distri-
Introduction. bution within the mouth of an individual (Dean, 1936, 1942;
Thylstrup and Fejerskov, 1978; Fejerskov et al., 1988a,b).
Up until the last major international meeting on cariostatic Therefore, dental fluorosis may usually be distinguished from
mechanisms of fluoride, held in Naples in 1976 (Brown and any other enamel disturbance. Apart from one study on stron-
Konig, 1977), there was almost consensus about the nature tium in water supplies (Curzon and Spector, 1977), no other
and mechanisms of dental fluorosis. The prevailing concept causative factor has so far been identified capable of inducing
was that maximum caries reduction would only be obtained the spectrum of enamel changes which is characteristic of the
through systemic use of fluorides. Dental fluorosis was there- changes induced by fluoride. Furthermore, there exists today
fore considered an unavoidable "side-effect" which was com- an extremely large amount of epidemiological data demonstrat-
monly thought to result from the toxic effect of fluoride on the ing that the occurrence of such lesions is associated with ex-
secretory phase of enamel formation. cessive fluoride intake (Dean, 1934, 1942; M011er, 1965;
Studies on human dental fluorosis during the early 1970's Richards et al., 1967; Myers, 1978; Thylstrup and Fejerskov,
(Fejerskov et al., 1974, 1975) led us to conclude that a re- 1978; Manji et al., 1986a, b; Larsen et al., 1986; Szpunar
evaluation of human dental fluorosis was necessary and that and Burt, 1988).
Fluorosis at the time of ernption.-Fluoride-induced enamel
Presented at a Joint IADR/ORCA International Symposium on Fluo- changes as they appear at the time of eruption range from thin
rides: Mechanisms of Action and Recommendations for Use, held white opaque lines corresponding to the perikymata running
March 21-24, 1989, Callaway Gardens Conference Center, Pine across the tooth surface, to an entirely chalky white enamel
Mountain, Georgia (Thylstrup and Fejerskov, 1978; Fejerskov et al., 1988b). Such
692

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

Vol. 69 Special Issue DENTAL FLUOROSIS IN MAN 693

clinical features reflect that fluoride given in low concentra- ronment may influence the clinical appearance of the enamel
tions over the long period of tooth development results in var- once the tooth has erupted into the mouth. However, the degree
ious degrees of enamel porosity (or hypomineralization). In its of post-eruptive modification of the enamel will to some extent
mildest forms, the porosity is to be found in the outermost also be determined by the degree of hypomineralization at the
enamel only, but the entire tooth surface is involved. With time of eruption.
increasing severity, both the depth of enamel involvement and Post-eruptive changes in fluorosis. - In the milder forms of
degree of porosity of the enamel increase. Assuming a rela- dental fluorosis, mechanical attrition, e.g., resulting from the
tively constant exposure level (most commonly water-borne use of abrasive toothpaste, will over time cause an apparent
fluoride), al1 surfaces of a given tooth will be equal1y affected "remission" of the f1uorotic lesions most likely due to surface
(Thylstrup and Fejerskov, 1978). enamel removal. The upper incisorsare particularlysusceptible
Within-mouth distribution of fluorosis. - The changes are to such environmental modifications. Often the incisal I?art of
symmetrical1y distributed within the oral cavity, but the se- the central incisors may be exposed to air when there is in-
verity varies among the different types of teeth (Dean, 1934, sufficient lip closure. Consequently, the incisal part will be-
1942; Meller, 1965; Thylstrup and Fejerskov, 1978; Larsen come dried out for long periods, and any porosities will therefore
et al., 1985, 1986; Manji et al., 1986a). The degree of severity be discerned. In addition, the incisal edges/cuspal tips are not
appears in principle to reflect the stage in life at which the overlying dentin, so any change in pore volume in these areas
various tooth types are formed and mineralized (Larsen et al ., will reflect itself as clinically different from that of the re-
1987), regardless of whether the child is born and reared in a maining parts of the teeth. This may give the impression that
low « 0.2) or a higher (1-2 ppm) water fluoride area (Fig. the incisal area is more affected than the remainder of the
1). The teeth which form and mineralize early in life are those surface, which in reality is equally porous. This will not be
that are least affected, whereas the later in life the teeth min- apparent clinically unless the surface has been dried properly.
eralize, the more severely they will be affected. In accordance The incisal edges may also be subjected to attrition, and, de-
with this concept, exposure to very low fluoride levels will pending on the degree of enamel porosity, this leads to ex-
result in only a few teeth exhibiting subsurface porosities to a posure of the underlying more porous enamel. There wil1 then
degree that wil1 be clinically manifest, although the underlying be a strong tendency for stains to be taken up by the thus-
pattern will nevertheless remain the same. In areas of high exposed enamel corresponding to the position of the upper lip
fluoride exposure, the primary dentition can also be involved, across the surface (Fejerskov et af., 1988b). General1y, the
and in principle the same within-mouth distribution is found uptake of stain will depend on the degree of porosity, but will
(Thylstrup, 1978; Larsen et aI., 1988). also be influenced by the nature of the individuals' dietary
Given that dental fluorosis is characterized by increased po- habits. Thus, discoloration in itself is not an appropriate mea-
rosity of the enamel, it will be appreciated that the oral envi- sure of severity. However, since discoloration occurs most

Per cent of teeth Figs. la,b- Percentage of teeth

(a) with dental fluorosis recorded ac-
100 1 r--r--r--.--,---,....--~-~
,
cording to Thylstrup and Fejerskov
(l978) in an area of Denmark with
about 0.1 ppm F in drinking water
(a) and in an area with 1.5·2.0 ppm
F (b). From Larsen et al. (1985),
Larsen et al. (1986). The numbers
50 along the x-axes refer to tooth num-
bers with + indicatingmaxillaryand
- = mandibular teeth.

-- 1

1-1 2-2 6-6 6+6 1+1 2+2 3-3 3+3 4-4 4+4 5-5 5+5 7-7 7+7
CJ TFscoreO rJ TFscore1 _ TF score ~ 2

Per cent of teeth (b)

100
1 _.r-- -~

1-1 1+1 2-2 2+2 3-3 3+3 4-4 4+4 5-5 5+5 6-6 6+6 7-7 7+7
L.J TF score 0 TF score 1 c=J
_ TF score z 2

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

694 FEJERSKOVet af. J Dent Res February 1990

often in the upper incisors, and since these are the teeth most presumed etiology (Jackson, 1961; Al-Alousi et al., 1975;
easily seen by lay persons, such features have been mistakenly FDl, 1982). According to this concept, all defects in enamel
considered as constituting intrinsic and principle characteristics are recorded based solely on descriptive criteria. If it is con-
of dental fluorosis. sidered important for the entire spectrum of enamel defects to
In more severe forms of dental fluorosis, the tooth erupts be recorded regardless of causative factors, such an approach
into the oral cavity entirely chalky white. The degree of po- may be useful. However, when such systems are used, the
rosity (hypomineralization) of such teeth results in a dimin- authors usually spend most of their discussions trying to decide
ished physical strength of the enamel, and parts of the superficial retrospectively which types of lesions might have been induced
enamel may break away. This type of damage ranges from by fluoride. Of importance here is the fact that in such studies
small scattered round defects (pits) to more confluent areas the very lesion types ("diffuse white opacities") which would
with loss of the outermost enamel. Such loss of outermost be expected to be associated with fluoride intake are indeed
enamel is particularly frequent along incisal edges and cusp so! (Cutress et al., 1985) Therefore, in contrast to others (e.g.,
tips. In the latter case, this will also involve the occlusal sur- Clarkson and O'Mullane, 1989), we consider such classifica-
faces which become rapidly worn. In the most severe forms tions to be of no value, as was also recently stressed by Ho-
of dental fluorosis, the extent and degree of porosity within rowitz (1986).
the enamel are so severe that most of the outermost enamel The continuum of clinical changes which characterize dental
will be chipped off immediately following eruption. This ex- fluorosis can best be classified by use of the TF index to re-
poses the underlying porous enamel, which will take up stain flect, in an ordinal scale, the histopathological changes asso-
rapidly. This post-eruptive damage may increase over time, ciated with fluoride-induced enamel changes. In a recent review
depending on the degree of severity at time of eruption. Since (Fejerskov et al., 1988b), it has been emphasized that the TF
post-eruptive damage may proceed over several years and thereby index represents logical improvements and extensions of Dean's
result in more severe enamel changes in older age cohorts original approach. Dean's classification has certain limitations
(Baelum et al., 1986), it is obvious that meaningful compar- when one is dealing with both the early signs as well as the
isons between populations of the severity of dental fluorosis severe forms of dental fluorosis (Thylstrup and Fejerskov, 1978).
can only be made between groups of individuals of equivalent The first difficulty lies in its weak definitions of the various
ages. categories in the milder forms. This is perhaps best illustrated
by comparing the photographic illustrations in a recent WHO
publication (WHO, 1987) with the features of dental fluorosis
Diagnostic problems in dental fluorosis. presented by Fejerskov et al. (1988b).
As indicated previously, much doubt has been raised as to Based on existing knowledge at the time, Dean included
the possibility of diagnosing the early stages of dental fluorosis staining in his diagnostic criteria. As previously indicated,
with any certainty (Jackson, 1961; Small and Murray, 1978; however, such post-eruptive modification of the enamel "is
Suckling and Pearce, 1984). Some of these difficulties result not a diagnostic sign" (Eklund et al., 1987). The second prob-
from the terminology which Dean proposed to describe his lem in Dean's index relates to his "severe" category. Consid-
category "questionable" (Dean, 1934). Even though this cat- erable confusion about the mechanisms by which fluoride affects
egory, as originally defined, may include enamel opacities of human enamel has arisen from imprecise reading of the criteria
non-fluoride origin (Zimmerman, 1954), the association be- provided by Dean. Dean et al, (1935) were clear that true
tween the category "questionable" and fluoride exposure is hypoplasias were not a feature of human dental fluorosis, de-
now well-established (Myers, 1983). Since the early signs of spite their use of the same word to describe the pits. It is
dental fluorosis reflect the degree and extent of porosity of the therefore very unfortunate that an illustration of Dean's severe
outermost enamel, it is mandatory that these porosities be made category, in the abovementioned WHO publication, includes
clinically visible by wiping dry the tooth surface. As with the a classic true hypoplasia without commenting on this important
diagnosis of early caries lesions, a proper examination can only diagnostic problem.
be made if plaque is first removed. If this is done, the signs Apart from being more precise, sensitive, and easier to use
of dental fluorosis are those of fine striae or lines across the than Dean's index, the TF index has the advantage that it also
enamel surface (Thylstrup and Fejerskov, 1978). This fact has allows for comparisons to be made between the two indices
also been emphasized by Zimmerman (1954), and by Iizuka (Wenzel and Thylstrup, 1982; Granath et al., 1985; Burger et
and Yasaki (1976). The latter set up differential diagnostic al., 1987). This is particularly important if, in the future, it is
criteria between fluoride- and non-fluoride-induced enamel felt necessary to make comparisons with the findings of earlier
opacities, and when these differential diagnostic characteristics studies in which Dean's index has been used.
are taken into account, much less confusion about early di- Recently, Horowitz et al. (1984) published what was stated
agnosis of dental fluorosis will prevail (Fejerskov et al., 1988b). to be a new classification system. However, when it is com-
pared with both Dean's and the TF index, it is apparent that
this method mixes the basic principles of the two indices. In
their method, plaque is not removed, nor are the tooth surfaces
Classification systems of dental fluorosis. dried. As apparent when the illustrations provided in their two
In order for the prevalence and severity of dental fluorosis publications are compared (Driscoll et al., 1983; Horowitz et
to be described within a population or within an individual, a al., 1984), Dean's category "questionable" has been ex-
sensitive, precise, and valid (in the sense that it measures what cluded. Otherwise their scores 1, 2, and 3 are (illustration-
it is supposed to measure) classification system is necessary. wise) identical to Dean's categories "very mild", "mild",
However, there are diverging opinions in this context. One and "moderate". As such, it is hampered by the same weak-
line of thinking emphasizes the need for a classification system ness as Dean's index in being imprecise in its definitions of
which specifically measures fluoride-induced enamel changes the various categories in the milder forms. According to the
in order to reflect increasing severity of the lesions. Typical proposal by Horowitz et al, (1984), it seems that all enamel
examples are those of Dean (1934), and Thylstrup and Fejer- defects included in Dean's category "questionable" should be
skov (1978). Opposing this concept is the view that no diag- considered "sound" or normal. However, Driscoll and Ho-
nosis of enamel defects should be made on the basis of a rowitz (personal communications, this meeting, 1989) have

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

Vol. 69 Special Issue DENTAL FLUOROSIS IN MAN 695

informed us that they include Dean's category "questionable" matrix secretion at any given point in the enamel, the under-
in their category 1. According to Dean and Elvove (1937), lying enamel will be at a stage of development where there is
"the quantitative aspects of a survey would probably be en- a gradual decrease in mineral content from the EID junction
tirely lost, or judged erroneously, if the 'questionable' cases toward the outer enamel. At this stage, the outer enamel is
were thrown into one grade or another...". Thus, in the same very hypomineralized (as compared with normal mature enamel)
population in which the same individuals have been examined except for a very thin outer layer of surface enamel which even
by use of both Dean's and Horowitz et ai.'s indices, it would at this stage contains more mineral than does the subsurface
be expected that the prevalence of dental fluorosis using Ho- enamel (Fejerskov and Thylstrup, 1987). From then on until
rowitz' method should be the same or lower. It is intriguing eruption of the tooth, the "amelogenins" and water will have
that quite the reverse was the case in the data presented (Dris- to be removed concomitant with crystal growth to allow the
coll et ai., 1983; Horowitz et al., 1984). The most important enamel to become fully mineralized (or mature). If this mat-
limitation is that Horowitz et al. (1984) proposed that staining uration process is interfered with by fluoride, the result would
should be taken as a diagnostic criterion so important that, if from a histopathological point of view be identical with what
present, it supersedes all the lower scores (i.e., 1, 2, and 3). characterizes dental fluorosis. In support of this hypothesis was
This may explain the apparent contradiction mentioned above. the finding that human fluorosed enamel, when compared with
That staining should be given such weight in the light of cur- normal mature enamel, had a similar total protein content, but
rent knowledge about the post-eruptive changes in fluorotic the fluorosed enamel retained a relatively high proportion of
enamel is surprising. The risk with this approach is that it immature matrix proteins (Eastoe and Fejerskov, 1984).
might overestimate severity in populations exhibiting staining Recently, the fluoride content has been examined throughout
but with otherwise relatively mild enamel changes, and would the enamel in teeth representing the complete range of macro-
thus be inappropriate for use under field conditions in a variety scopically defined degrees of severity of dental fluorosis class-
of different populations. ified according to the TF index (Richards et al., 1989a). These
data show that with increasing severity of fluorosis, the fluo-
ride concentration increases not only in the superficial enamel,
but throughout the whole tissue. These findings are contradic-
Histopathology and chemistry tory to previous observations by Olsen and Johansen (1978),
of human fluorotic enamel. who did not find any association between fluoride concentra-
tions in the outer 100 [Link] of enamel and the surface appearance
During the past two decades, new information has been of the teeth. The authors classified their teeth according to the
gathered about the light and electron microscopic changes in modification of Dean's index proposed by Meller (1965). As
human enamel exhibiting various degrees of dental fluorosis previously stressed, discoloration is misleadingly included in
(Kerebel et al., 1973; Fejerskov et al., 1974, 1975; Kerebel this classification, and this may have led to confusion as to
and Daculsi, 1976; Sundstrom et al., 1978; Sundstrom and the true degree of severity.
Myhrberg, 1978; Thylstrup, 1978; Thylstrup et at., 1978). The Because of the porous nature of fluorotic enamel, it would
overall findings support the concept that fluoride affects the be expected that post-eruptive uptake of fluoride might take
forming enamel to cause porosity of the enamel. The degree place to a substantial degree. This was reflected in a rather
and extent of porosity depend on the tissue fluid concentration great variation in fluoride concentration in severely affected
of fluoride during tooth development. The structural arrange- teeth (TF scores 7-9) where substantial destruction of the sur-
ment of the crystals appears normal, but the width of the in- face enamel was recorded. The observation of increased fluo-
tercrystalline spaces increases-hence the pores. Likewise, the ride concentrations at all enamel depths in teeth exhibiting a
arcade-shaped gaps which partly surround the enamel rods dur- TF score 4 as compared with lower TF categories indicates,
ing normal enamel development(Fejerskov and Thylstrup, 1987) however, that the fluoride concentrations represent fluoride
become widened (Fejerskov et al., 1974). incorporated into the enamel prior to eruption. That this as-
The pits found in more severe cases of dental fluorosis are sumption holds true has recently been demonstrated in a study
a result of post-eruptive breakdown of the surface enamel, the of unerupted human fluorotic teeth (Richards et al., 1989b).
hypomineralized lesion being located deep to a well mineral- How fluoride may interfere with the complex processes of
ized surface zone which is very fragile to mechanical stress enamel formation and maturation cannot be resolved easily
(Thylstrup and Fejerskov, 1979; Fejerskov et al., 1983). This from human studies. Human clinical studies have demonstrated
is consistent with the clinical finding that pits develop after that it is possible for children to develop dental fluorosis when
tooth eruption (Thylstrup, 1983; Baelum et al., 1986). This they have been exposed to fluoride supplements during the
observation is very important in derivation of hypotheses about period when the teeth are undergoing enamel maturation (Ishii
pathogenic mechanisms in dental fluorosis. Thus, if the old and Nakagaki, 1984; Larsen et ai., 1985). It can be speculated
concept is perpetuated that dental fluorosis reflects hypoplastic that fluoride may affect the maturation ameloblasts by influ-
enamel (Horowitz, 1986; WHO, 1987), it lends support to the encing their ability to remove protein and water from maturing
concept that such "hypoplasias" are a direct result of a toxic enamel and/or may interfere with the ameloblast's capacity to
damage to the secretory ameloblasts by fluoride. Already Dean produce proteolytic enzymes necessary to initiate amelogenin
(1934) and Dean et al, (1935) were stressing the fact that breakdown. It is surprising that a limited fluoride intake such
typical hypoplasias should not be included in his classification as that obtained from a Danish fluoride supplement program
system, and true hypoplasias have not been found to be as- (Thylstrup et ai., 1979) should be sufficient to result in tissue
sociated with previous fluoride exposure (Cutress et al., 1985). fluid concentrations of fluoride of a magnitude which may have
Based on our light and electron microscopic findings (Fe- any effect on enzymatic processes in vivo. Studies on rats
jerskov et al., 1974), we therefore proposed that the hypo- indicate, however, that plasma fluoride concentrations as low
mineralized regions of fluorosed enamel might be a result of as 1.5 umol/L may be sufficient to induce enamel changes
an arrest of enamel maturation. During normal enamel devel- (Angmar-Mansson and Whitford, 1984), so it seems highly
opment, a substantial crystal growth takes place in the inner relevant for the dose-response relationship in humans to be re-
enamel following the gradual removal of proteins, while outer evaluated before hypotheses on pathogenesis of dental fluo-
enamel is still being secreted. At the time of cessation of enamel rosis are further developed.

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

696 FETERSKOVet at. J Dent Res February 1990

TABLE
Dose-response relationship in humans. FORMULA USED FOR COMPUTING THE DAILY DOSE OF
There has been some concern expressed in recent years that FLUORIDE FROM DRINKING WATER
in North America and Europe the prevalence of dental fluorosis
DOSE WATER F CONC. x WATER INTAKE
may have increased slightly (Szpunar and Burt, 1987; Lever- (mg F/kg body weight) (mg FIL) (L/kg body weight)
ett, 1986). Such concern has arisen principally in light of the
increasing use of a variety of fluoride regimes in these popu- Galagan and Vermillion provided a formula from which the water intake
lations. In North America, in particular, where rather large could be computed based on knowledge of the mean maximum air tem-
populations are provided with artificially fluoridated waters, perature:
the addition of a variety of topical (and systemic) agents might
be expected to influence both the prevalence and severity of WATER INTAKE = -0.038 + 0.0062 x MEAN MAXIMUM
dental fluorosis (Whitford et al., 1987). (fl. oz./lb body weight) TEMPERATURE
(OFahrenheit)
At the same time, an increasing number of reports have
appeared from countries in which no organized fluoride pro- Using the following relationships:
phylactic programs have been implemented, indicating higher 1 [Link] (USA) = 0.0295735 liters
prevalences and severity of dental fluorosis in low-fluoride 1 lb. = 0.4535924 kg
areas than would be expected (Meller et al., 1970; Olsson,
1978; Glass, 1984; Subbareddy and Tewari, 1985; Manji et the above formula may be transformed:
al., 1986a,b,c; Brouwer et al., 1988; Evans, 1988; Grobler
et al., 1988). WATER INTAKE = -0.0024775 + 0.00040423 x MEAN
(L/kg body weight) MAXIMUM TEMPERATURE
The principal difficulty in interpretation of recent and past ("Fahrenheit)
data on dental fluorosis is that there have been misapprehen-
sions about the nature of the dose-response relationship. The Therefore, the equation
result of a long-lasting low-fluoride exposure in humans is first
recordable in children after the age of 6, and precise estimates DOSE WATER F CONC. x WATER INTAKE
of the most severe effects of fluoride can only be obtained (mg F/kg body weight) (mg FIL) (L/kg body weight)
when the premolars and the second molars have erupted. This
means that the dose responsible for such changes has to be is equivalent to
estimated from a past cumulative history of about 10-12 years,
DOSE = WATER F CONC. x (- 0.0024775 + 0.00040423 x
during which period rapidly growing children will have been (mgIL) (mgIL) MEAN MAX TEMPERATURE)
exposed to considerable changes in dietary patterns and prac- ("Fahrenheit)
tices (from weaning, bottlefeeding, etc.). The relative contri-
bution of food-borne fluoride for the development of dental
fluorosis is still debatable. As seen from the literature (Taves, al, (1985). Two striking features about the data are apparent
1983; Smith and Ekstrand, 1988), it is difficult to arrive at a in this Fig. First, regardless of the source of the data, it is
firm agreement as to how much fluoride is actually ingested- clear that the slope of the regression line for the association
and the bioavailability of fluoride in food is as yet uncertain. between fluoride dose and dental fluorosis is such that even
However, from epidemiological studies throughout the world with very low fluoride intake from water a certain level of
there is a positive relationship between water-borne fluorides dental fluorosis will be found (r2 = 0.87). Second, the dose-
and the occurrence of dental fluorosis (Dean, 1942; Meller, response relationship is linear. These data indicate that for
1965; Richards et al., 1967; Myers, 1978; Thylstrup and Fe- every increase of dose of 0.01 mg Flkg b.w., an increase in
jerskov, 1978; Manji et al., 1986a,c; Larsen et al., 1987). Fci of 0.2 will be expected.
In order to understand the relationship between fluoride ex- From the studies by Aasenden and Peebles (1974), it may
posure (dose) and dental fluorosis (response), it is obvious that be calculated that the median daily dose of fluoride from tablets
we should consider all signs of dental fluorosis, including its would range between 0.042 and 0.070 mg Flkg b.w., with an
earliest manifestations. average dose over this period of 0.056 (Baelum et al., 1987).
It is a prevailing concept that temperature affects the prev- This fluoride dosage resulted in an Fci value of 1.23 (based
alence and severity of dental fluorosis simply as a result of on the Fci weights recommended by Dean), a value which is
variations in daily consumption of water (Galagan et al., 1957). consistent with the expected value for the Fci based on the
Galagan and Vermillion (1957) formulated an equation which epidemiological evidence in Fig. 4. Similarly, in a Swedish
permits the calculation of water-intake as a function of tem- study on fluoride tablet supplementation (Granath et al., 1985),
perature (Table). When this formula is applied to data, for the median dosage of fluoride ranged from 0.024 to 0.042 mg
example, from Dean's studies in Illinois and Ohio (Dean et Flkg b.w., resulting in an Fci value of about 0.4. Thus, it
al., 1941, 1942), it is possible for the dosage of fluoride from would appear that regardless of whether fluoride is obtained
water (mg Flkg body weight) associated with dental fluorosis from drinking water or from supplements, the expected out-
in these communities at that time to be estimated. From Fig. come in terms of levels of dental fluorosis is approximately
2 it is apparent that with an intake from water equivalent to the same.
0.02 mg Flkg b.w. a prevalence of about 40-50% was observed Currently, the recommendation in the USA for fluoride sup-
(about 15-25% excluding questionables). The community in- plements in, for example, a < 0.3 ppm fluoride area (ADA,
dex values, Fci, at this intake would be between 0.3 and 0.4 1982) is that 0.25 mg of fluoride should be provided as sup-
(Fig. 3). Even if fluoride from food might double this dose in plements from birth to 2 years of age, and thereafter 0.5 mg
small children (Ophaug et al., 1985), these data from Dean should be provided until the age of 3, after which 1 mglday
suggest estimates that are considerably below that of Forsman should be provided. In those receiving such supplements, we
(1977), who stated that dental fluorosis is unlikely to occur at would expect, based on Fig. 4, that the Fci values would be
a dosage below 0.1 mg Flkg body weight. approximately 1.0, with a prevalence of about 70-80%.
In Fig. 4 we have also included data from both Richards et However, such values will of course only be found provided
at. (1967), adjusted for temperature variations, and Butler et that the dose regime is strictly followed and that there is 100%

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

Vol. 69 Special Issue DENTAL FLUOROSIS IN MAN 697

DOSE-PREVALENCE RELATIONSHIP DOSE-RESPONSE RELATIONSHIP

FROM DEAN ET AL. 1941, 1942 FROM DEAN ET AL. 1941. 1942

100 1.8

1.6

80 1.4

% ** 1.2
P
R
E
60 t*
F
V C
A
L * * I 0,8
E
N
40 **
C 0,6
E
04
20
.* 02 *
*
0
* 0
*
0 0,01 0,02 003 0.04 005 0,06 0 0,01 0,02 0,03 0,04 0,05 0.06

DAILY DOSE (mg F/kg) DAILY DOSE (mg F/kg)

Fig. 2- The relationship between estimated daily dose of fluoride from Fig. 3- The relationship between estimated daily dose of fluoride from
drinking water and resulting prevalence of dental fluorosis in the popu- drinking waters and the community index (Fci) of dental fluorosis as
lations examined by Dean et al. (1941; 1942). The category "question- obtained from Dean et al.'s original data (1941, 1942).
able" is included.

be considerable variations with time in the levels of fluoride

compliance. The data provided by Holm and Andersson (1982) in any given drinking water supply, even in artificially fluor-
clearly indicate that the outcome in terms of prevalence of idated water supplies.
dental fluorosis is highly dependent on both time of initiation There is no doubt that there may be a number of factors
of fluoride supplementation as well as on the degree of com- which influence the susceptibility of individuals and popula-
pliance. tions to dental fluorosis, and which may account for higher
In the above calculations, we have purposely used data only than expected levels of dental fluorosis being found. Recently,
from studies in which Dean's original classification system has Manji et al, (1986c) have presented epidemiological data in-
been used. As we have indicated in Figs. 3 and 4, the dose- dicating a positive association among fluoride, altitude, and
response relationship in the three data sets suggests a linear dental fluorosis in Kenya. The biological explanation for this
relationship and implies therefore that there exists no "criti- is intriguing. While the association may reflect that individuals
cal" value below which the effect of fluoride on dental enamel living at different altitudes simply have different water con-
will not be manifest. sumption and dietary habits, it could also reflect that altitude
In presentation of Dean et al.'s data in order to estimate may make individuals more susceptible to low fluoride expo-
dose-response relationships, Hodge (1950) used a logarithmic sure. Several other factors may be important to consider when
scale for the x-axis (dose). The effect of such transformation unexpectedly high prevalence and severity of dental fluorosis
of the data in Fig. 4 is presented in Fig. 5. The conclusions are reported. There is no doubt that the bioavailability of in-
one would draw from this logarithmic transformation would gested fluoride plays a considerable role. Thus, frequency and
be that the relationship is curvilinear, such that below a certain composition of meals will highly influence bioavailability of
level of fluoride intake the toxic effects of fluoride would be concomitantly ingested fluoride. A recent example is the dem-
minimal and in any case below about an Fci value of 0.4- onstration that fluoride ingested from toothpaste is bioavailable
0.6-ironically the level which Dean suggested to be of "bor- to almost the full extent if ingested on an empty stomach,
derline public health significance". The question arises, how- whereas if taken in relation to a meal the bioavailability is
ever, whether it is appropriate to use a log-transformation where reduced to about 50% (Spak and Ekstrand, 1988). When the
the observed relationship is quite clearly already linear. frequencies and quantity of food intake, during the day, in
When the above considerations are taken into account, it is children in industrialized countries are compared with those of
to be expected that, in populations in which additional fluoride children in developing countries, it is apparent that the latter
has been prescribed, an increase in prevalence and severity of spend long periods of the day with relatively insubstantial
dental fluorosis will occur (Szpunar and Burt, 1987; Leverett amounts of food in the stomach (Manji, 1988). This alone
et al., 1988). Furthermore, water fluoride levels are seldom seems to us to be of importance when trying to explain ap-
stable. Larsen et al, (1988) have recently shown that there may parent variations in susceptibility to fluorides. It is possible

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

698 FEJERSKOVet al. J Dent Res February 1990

DOSE-RESPONSE RELATIONSHIP DOSE-RESPONSE RELATIONSHIP

2.2 2.2

2 0 2 0
0 0
1.8 1.8

1.6 + 1.6 +
1.4 + +13 1.4 + .,g I!ll
F 1.2 0 F 1.2 0
C + 0 C + 0
I I
+ +
0,8 0,8
+",
0.6
+ +
++ ** 0.6
+
++0

"
+
0.4 0.4 +0 "
+ +
+0 * *0 + 0 * *0
0,2 0.2 **
0+ +
O....--=c_-L._.L.--...L_-'-_L.----L_...L-----'_---' 0
l • * ,0.
o 0.01 0,02 0.03 0.04 0,05 0.06 0,07 0,08 0.09 0.1 0.001 0.01 0.1
DAILY DOSE (mg F/kg body weight) DAILY DOSE (mg F /kg body weight)
+ RICHAR,DS ET AL. 1967 * DEAN ET AL. 1941/2 + RICHARDS ET AL. 1967. DEAN ET AL. 1941/42
o BUTLER ET AL, 1985 - LINEAR TREND o BUTLER ET AL. 1985

Fig. 4-The community index of dental fluorosis (Fci) plotted against Fig. 5-The same data presented in Fig. 4, but the x-axis has been
daily dose of fluoride from drinking water as estimated from data origi- transformed on a logarithmic scale.
nating from Dean et al, (1941, 1942), Richards et al. (1967), and Butler
et al, (1985). The dose-response relationship is linear, and data indicate
an increase in Fci of 0.2 for every increase of dose of 0.01 mg Flkg b.w.
affects mineralization of teeth and bone in humans. At what
level the changes should be considered of public health concern
that malnutrition may increase the susceptibility of populations will strongly depend on the type of society with which we are
to dental fluorosis, but this hypothesis has yet to be substan- dealing.
tiated. Finally, it is necessary to establish better studies in humans
Finally, but probably equally important, the way the body to identify other factors which alone or in combination are able
handles fluoride in hot climates, where for example urinary to influence the susceptibility of the individual to fluoride.
excretion may be much less than seen in temperate climates, Likewise, attempts should be made to identify other factors
may playa role. Furthermore, little is known about the relative which alone or in combination may be able to produce slight
role that acid-base variations in body fluids play in fluoride changes in porosity in the enamel similar (in surface appear-
susceptibility in man. ance and distribution within the mouth) to those caused by
fluoride.

Conclusion.
REFERENCES
The dose-response relationship has been estimated from data
collected using a system in which the ordinality is not consis- AASENDEN, R. and PEEBLES, T.C. (1974): Effects of Fluoride
Supplementation from Birth on Human Deciduous and Permanent
tent with the actual biological effects of fluoride on developing Teeth, Arch Oral Bioi 19:321-326.
dental enamel. In order for more precise estimates on the dose- ADA COUNCIL ON DENTAL THERAPEUTICS (1982): Fluoride
response relationships to be obtained, large-scale studies in Compounds. Accepted Dental Therapeutics, 39th ed., Chicago:
different parts of the world should be conducted using the TF American Dental Association, pp. 344-368.
index along the lines we recently proposed (Fejerskov et al., AL-ALOUSI, W.; JACKSON, D.; CROMPTON, G.; and JENKINS,
1988b). O.C. (1975): Enamel Mottling in a Fluoride and in a Non-Fluoride
Even when available data from the USA are used, however, Community. Parts I and II, Br Dent J 138:9-15, 56-60.
the dose-response curve presented shows a linear relationship ANGMAR-MAN'SSON, B. and WHITFORD, G.M. (1984): Enamel
and indicates that it may no longer be appropriate to seek for Fluorosis Related to Plasma F Levels in the Rat, Caries Res 18:25-
32.
a definite lower border below which signs of fluoride-induced BAELUM, V.; MANJI, F.; and FEJERSKOV, O. (1986): Post-
changes will not occur. However, it is important to appreciate eruptive Tooth Age and Severity of Dental Fluorosis in Kenya,
that these changes need to be identified only when the biolog- Scand J Dent Res 94:405-410.
ical effects of fluoride on mineralizing tissues are to be inter- BAELUM, V.; FEJERSKOV, 0.; MANJI, F.; and LARSEN, M.J.
preted. Unless the existence of such changes is appreciated, (1987): Daily Dose of Fluoride and Dental Fluorosis, Danish Dent
we will not be able to improve our knowledge of how fluoride J 91:452-456.

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

Vol. 69 Special Issue DENTAL FLUOROSIS IN MAN 699

BROUWER, I.D.; BACKER DIRKS, 0.; DE BRUIN, A.; and Humans, J.L. Shupe, H.B. Peterson, and N.C. Leone, Eds., Salt
HAUTVAST, J.G.A.J. (1988): Unsuitability of World Health Or- Lake City: Paragon Press, pp. 305-317.
ganisationGuidelinesfor Fluoride Concentrations in Drinking Water FEJERSKOV, 0.; SILVERSTONE, L.M.; MELSEN, B.; and
in Senegal, Lancet I: 223-225. M0LLER, LJ. (1975): Histological Features of Fluorosed Human
BROWN, W.E. and KONIG, K. (1977): Cariostatic Mechanisms of Dental Enamel, Caries Res 9:190-210.
Fluorides, Caries Res l1:Supplement. FEJERSKOV, o. and THYLSTRUP, A. (1987): Dental Enamel. In:
BURGER, P.; CLEATON-JONES, P.; DU PLESSIS, J.; and DE Human Oral Embryology and Histology, LA. Mjor and O. Fe-
VRIES, J. (1987): Comparison of Two Fluorosis Indices in the jerskov, Eds., Copenhagen: Munksgaard, pp. 50-89.
Primary Dentition of Tswana Children, Community Dent Oral Ep- FEJERSKOV, 0.; THYLSTRUP, A.; and LARSEN, M.J. (1977):
idemioI15:95-97. Clinical and Structural Features and Possible Pathogenic Mecha-
BUTLER, W.J.; SEGRETO, V.; and COLLINS, E. (1985): Preva- nisms of Dental Fluorosis, Scand J Dent Res 85:510-534.
lence of Dental Mottling in School-aged Lifetime Residents of 16 FORSMAN, B. (1977): Early Supply of Fluoride and Enamel fluo-
Texas Communities, Am J Publ Health 75:1408-1412. rosis, Scand J Dent Res 85:22-30.
CLARKSON, Land O'MULLANE, D. (1989): A Modified DDE GALAGAN, D.J. and VERMILLION, J.R. (1957): Determining Op-
Index for Use in Epidemiological Studies of Enamel Defects, J timum Fluoride Concentrations, Publ Health Rep 72:491-493.
Dent Res 68:445-450. GALAGAN, D.J.; VERMILLION, J.R.; NEVITT, G.A.; STADT,
CURZON, M.E.J. and SPECTOR, P.C. (1977): Enamel Mottling in Z.M.; and DART, R.E. (1957): Climate and Fluid Intake, Publ
a High Strontium Area of the U.S.A., Community Dent Oral Ep- Health Rep 72:484-490.
idemioI5:243-247. GLASS, R.L. (1984): Kuwait National Dental Health Survey, Part
CUTRESS, T.W.; SUCKLING, G.W.; PEARCE, EJ.; and BALL, I, Boston: Forsyth Dental Center.
M.E. (1985): Defects of Tooth Enamel in Children in Fluoridated GRANATH, L.; WIDENHEIM, L; and BIRKHED, D. (1985): Di-
and Non-Fluoridated Water Areas of the Auckland Region, N Z agnosis of Mild Enamel Fluorosis in Permanent Maxillary Incisors
Dent J 81:12-19. Using Two Scoring Systems, Community Dent Oral Epidemiol
DEAN, H.T. (1934): Classification of Mottled Enamel Diagnosis, J 13:273-276.
Am Dent Assoc 21:1421-1426. GROBLER, S.R.; VAN WYK, C.W.; and KOTZE, D. (1988): Re-
DEAN, H.T. (1936): Endemic Fluorosis and Its Relation to Dental lationship between Enamel Fluoride Levels, Degree of Fluorosis
Caries, Publ Health Rep 53:1443-1452. and Caries Experience in Communities with a Nearly Optimal and
DEAN, H.T. (1942): The Investigation of Physiological Effects by a High Fluoride Level in the Drinking Water, Caries Res 20:284-
the Epidemiological Method. In: Fluorine and Dental Health, 288.
F.R. Moulton, Ed., Washington, D.C.: American Association for HODGE, H.C. (1950): The Concentration of Fluorides in the Drink-
the Advancement of Science, pp. 23-31. ing Water to Give the Point of Minimum Caries with Maximum
DEAN, H.T.; ARNOLD, F.A.; and ELVOVE, E. (1942): Domestic Safety, JAm Dent Assoc 40:436-439.
Water and Dental Caries. V. Additional Studies of the Relation of HOLM, A.K. and ANDERSSON, R. (1982): Enamel Mineralization
Fluoride Domestic Waters to Dental Caries in 4425 White Chil- Disturbances in 12-Year-Old Children with Known Early Exposure
dren, Age 12-14 Years, of 13 Cities in 4 States, Publ Health Rep to Fluorides, Community Dent Oral Epidemiol 10:335-339.
57:1155-1179. HOROWITZ, H.S. (1986): Indexes for Measuring Dental Fluorosis,
DEAN, H.T.; DIXON, R.M.; andtOHEN, C. (1935): Mottled Enamel J Publ Health Dent 46:179-183.
in Texas, Publ Health Rep 50:424-442. HOROWITZ, H.S.; HEIFETZ, S.B.; DRISCOLL, W.S.; KING-
DEAN, H.T. and ELVOVE, E. (1937): Further Studies on the Min- MAN, A.; and MEYERS, R.J. (1984): A New Method for As-
imal Threshold of Chronic Endemic Dental Fluorosis, Publ Health sessing the Prevalence of Dental Fluorosis-The Tooth Surface
Rep 52:1249-1264. Index of Fluorosis, J Am Dent Assoc 109:37-41.
DEAN, J.T.; JAY, P.; ARNOLD, F.A.; and ELVOVE, E. (1941): IIZUKA, Y. and YASAKI, T. (1976): On the Differential Diagnosis
Domestic Waters and Dental Caries. II. A Study of 2832 White between Dental Fluorosis and Nonfluoride Enamel Mottlings, Bull
Children Ages 12-14 Years of Eight Suburban Chicago Commu- Kan Dent Coli 4:55-56.
nities, Including Lactobacillus acidophilus Studies of 1761 Chil- ISHII, T. and NAKAGAKI, H. (1984): Study of the Correlation be-
dren, Publ Health Rep 56:761-792. tween the Degree of Dental Fluorosis and the Duration of Fluoride
DRISCOLL, W.S.; HEIFETZ, S.B.; HOROWITZ, H.S.; KING- Present in Drinking Water. In: Tooth Enamel IV, R.W. Fearn-
MAN, A.; MEYERS, R.J.; and ZIMMERMAN, E.R. (1983): head and S. Suga, Eds., Amsterdam: Elsevier, pp. 338-341.
Prevalence of Dental Caries and Dental Fluorosis in Areas with JACKSON, D. (1961): A Clinical Study of Non-Endemic Mottling
Optimal and Above-Optimal Water Fluoride Concentrations, JAm of Enamel, Arch Oral BioI 5:212-223.
Dent Assoc 107:42-47. KEREBEL, B. and DACULSI, G. (1976): Etude Ultrastructurale et
EASTOE, J. and FEJERSKOV, O. (1984): Composition of Mature Cristallographique de [Link] Humain dans la Fluorose Ende-
Enamel Proteins from Fluorosed Teeth. In: Tooth Enamel IV, mique, J Biol Buccale 4:143-154.
R.W. Fearnhead and S. Suga, Eds., Amsterdam: Elsevier, pp. KEREBEL, B.; DUBOIS, T.; CLERGEAU-GUERITAULT, S.; and
326-330. SIAU, T. (1973): [Link] Dentaire Humain dans la Fluorose
EKLUND, S.A.; ISMAIL, A.I.; BURT, B.A.; and CALDERONE, Endemique, Schweiz Monatsschr Zahnheilkd 83:326-343.
J.J. (1987): High-Fluoride Drinking Water, Fluorosis, and Dental LARSEN, M.J.; KIRKEGAARD, E.; FEJERSKOV, 0.; and POUL-
Caries in Adults, J Am Dent Assoc 114:324-328. SEN, S. (1985): Prevalence of Dental Fluorosis after Fluoride Gel
FDI TECHNICAL REPORT (1982): An Epidemiological Index of Treatments in a Low Fluoride Area, J Dent Res 64:1076-1079.
Developmental Defects of Enamel, No. 15, Int Dent J 32:159- LARSEN, M.J.; KIRKEGAARD, E.; and POULSEN, S. (1987):
167. Patterns of Dental Fluorosis in a European Country in Relation to
FEJERSKOV, 0.; KRAGSTRUP, J.; and RICHARDS, A. (1988a): Fluoride Concentrations of Drinking Water, J Dent Res 66: 10-12.
Fluorosis of Teeth and Bone. In: Fluoride in Dentistry, J. Ek- LARSEN, M.J.; KIRKEGAARD, E.; POULSEN, S.; and FEJER-
strand, O. Fejerskov, and L.M. Silverstone, Eds., Copenhagen: SKOV, O. (1986): Enamel Fluoride, Dental Fluorosis and Dental
Munksgaard, pp. 190-228. Caries Among Immigrants to and Permanent Residents of Five
FEJERSKOV, 0.; JOHNSON, N.W.; and SILVERSTONE, L.M. Danish Fluoride Areas, Caries Res 20:349-355.
(1974): The Ultrastructure of Fluorosed Human Dental Enamel, LARSEN, M.J.; RICHARDS, A.; and FEJERSKOV, O. (1985): De-
Scand J Dent Res 82:357-372. velopment of Dental Fluorosis According to Age at Start of Fluo-
FEJERSKOV, 0.; MANJI, F.; BAELUM, V.; and M0LLER, LJ. ride Administration, Caries Res 19:519-527.
(1988b): Dental Fluorosis. A Handbook for Health Workers, LARSEN, M.L; SENDEROVITZ, F.; KIRKEGAARD, E.; POUL-
Copenhagen: Munksgaard. SEN, S.; and FEJERSKOV, O. (1988): Dental Fluorosis in the
FEJERSKOV, 0.; RICHARDS, A.; and JOSEPHSEN, K. (1983): Primary and the Permanent Dentition in Fluoride Areas with Con-
Pathogenesis and Biochemical Findings of Dental Fluorosis in Var- sumption of either Powdered Milk or Natural Cow's Milk, J Dent
ious Species. In: Fluorides. Effects on Vegetation, Animals and Res 67:822-825.

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

700 FEJERSKOVet al. J Dent Res February 1990

LEVERETT, D.B. (1986): Prevalence of Dental Fluorosis in Fluor- on the Bioavailability of Ingested Fluoride Toothpaste, J Dent Res
idated and Nonfluoridated Communities-a Preliminary Investi- 67 (Spec. Iss.): 357, Abst. No. 1953.
gation, J Publ Health Dent 46:184-187. SUBBAREDDY, V.V. and TEWARI, A. (1985): Enamel Mottling
LEVERETT, D.B.; ADAIR, S.M.; and PROSKlN, H.M. (1988): at Different Levels of Fluoride in Drinking Water in an Endemic
Dental Fluorosis Among Children in Fluoridated and Nonfluori- Area, J Indian Dent Assoc 57:205-212.
dated Communities, J Dent Res 67 (Spec. Iss): 230, Abst. No. SUCKLING, G.W. and PEARCE, E.I.F. (1984): Developmental De-
936. fects of Enamel in a Group of New Zealand Children: Their Prev-
MANJI, F. (1988): Sugar Availability, Diet, and Dental Caries, Ph.D. alence and Some Associated Etiological Factors, Community Dent
Thesis, London: University of London. Oral Epidemiol 12:177-184.
MANJI, F.; BAELUM, V.; FEJERSKOV, 0.; and GEMERT, W. SUNDSTROM, B.; JONGEBLOED, W.L.; and ARENDS, J. (1978):
(1986a): Enamel Changes in Two Low-Fluoride Areas of Kenya, Fluorosed Human Enamel, Caries Res 12:329-338.
Caries Res 20:371-380. SUNDSTROM, B. and MYHRBERG, H. (1978): Light and Scanning
MANJI, F.; BAELUM, V.; and FEJERSKOV, O. (1986b): Dental Electron Microscopy of Fluorosed Enamel from Human Permanent
Fluorosis in an Area of Kenya with 2 ppm Fluoride in the Drinking Teeth, Caries Res 12:320-328.
Water, J Dent Res 65:659-662. SZPUNAR, S.M. and BURT, B.A. (1987): Trends in the Prevalence
MANJI, F.; BAELUM, V.; and FEJERSKOV, O. (1986c): Fluoride, of Dental Fluorosis in the United States: A Review, J Publ Health
Altitude and Dental Fluorosis, Caries Res 20:473--480. Dent 47:71-79.
MYERS, H.M. (1978): Fluorides and Dental Fluorosis, Basel: Kar- SZPUNAR, S.M. and BURT, B.A. (1988): Dental Caries, Fluorosis,
ger. and Fluoride Exposure in Michigan Schoolchildren, J Dent Res
MYERS, H.M. (1983): Dose-Response Relationship between Water 67:802-806.
Fluoride Levels and the Category of Questionable Dental Fluo- TAVES, D.R. (1983): Dietary Intake of Fluoride Ashed (Total Fluo-
rosis, Community Dent Oral Epidemiol 11:109-112. ride) v. Unashed (Inorganic Fluoride) Analysis of Individual Foods,
M0LLER, I.J. (1965): Dental Fluorose og Caries. Ph.D. Thesis, Br J Nutr 49:295-301.
Copenhagen: Rhodos. THYLSTRUP, A. (1978): Distribution of Dental Fluorosis in the Pri-
M0LLER, I.J.; PINDBORG, 1.J.; GEDALIA, I.; and ROED PE- mary Dentition, Community Dent Oral Epidemiol 6:329-337.
TERSEN, B. (1970): The Prevalence of Dental Fluorosis in the
THYLSTRUP, A. (1983): Posteruptive Development of Isolated and
People of Uganda, Arch Oral Bioi 15:213-225.
Confluent Pits in Fluorosed Enamel in a 6-Year-Old Girl, Scand
OLSEN, T. and JOHANSEN, E. (1978): Fluoride Content of Human
J Dent Res 91:243-246.
Fluorosed Enamel, J Dent Res 57 (Spec. Iss.): 281.
OLSSON, B. (1978): Dental Caries and Dental Fluorosis in Arussi THYLSTRUP, A. and FEJERSKOV, O. (1978): Clinical Appearance
Province, Ethiopia, Community Dent Oral Epidemiol 6:338-343. of Dental Fluorosis in Permanent Teeth in Relation to Histological
OPHAUG, R.H.; SINGER, L.; and HARLAND, B.F. (1985): Di- Changes, Community Dent Oral Epidemiol 6:315-328.
etary Fluoride Intake of 6-Month and 2-Year-Old Children in Four THYLSTRUP, A. and FEJERSKOV, O. (1979): A Scanning Electron
Dietary Regions of the United States, Am J Clin Nutr 42:701- Microscopic and Microradiographic Study of Pits in Fluorosed
707. Human Enamel, Scand J Dent Res 87:105-114.
RICHARDS, L.F.; WESTMORELAND, W.W.; TASHIRO, M.; THYLSTRUP, A.; FEJERSKOV, 0.; BRUUN, C.; and KANN, J.
McKAY, C.H.; and MORRISON, J.T. (1967): Determining Op- (1979): Enamel Changes and Dental Caries in 7 year Old Children
timum Fluoride Levels for Community Water Supplies in Relation Given Fluoride Tablets from Shortly after Birth, Caries Res 13:265-
to Temperature, JAm Dent Assoc 74:389-397. 276.
RICHARDS, A.; FEJERSKOV, 0.; and BAELUM, V. (1989a): THYLSTRUP, A.; FEJERSKOV, 0.; and MOSHA, H.J. (1978): A
Enamel Fluoride in Relation to Severity of Human Dental Fluo- Polarized Light and Microradiographic Study of Enamel in Human
rosis, Adv Dent Res 3:147-153. Primary Teeth from a High Fluoride Area, Arch Oral Bioi 23:373-
RICHARDS, A.; FEJERSKOV, 0.; BAELUM, V.; and LIKlMANI, 380.
S. (1989b): Enamel Fluoride in Unerupted Fluorotic Human Teeth, WENZEL, A. and THYLSTRUP, A. (1982): Dental Fluorosis and
Caries Res 23:103. Localized Enamel Opacities in Fluoride and Nonfluoride Danish
SMALL, B.W. and MURRAY, 1.J. (1978): Enamel Opacities: Prev- Communities, Caries Res 16:340-348.
alence, Classifications and Aetiological Considerations, J Dent WHITFORD, G.M.; ALLMANN, D.W.; and SHAHED, A.R. (1987):
6:33--42. Topical Fluorides: Effects on Physiologic and Biochemical
SMITH, F. and EKSTRAND, J. (1988): Fluoride in the Environment Processes, J Dent Res 66:1072-1078.
and Intake in Man. In: Fluoride in Dentistry, J. Ekstrand, O. WHO (1987): Oral Health Surveys. Basic Methods, 3rd cd., Ge-
Fejerskov, and L.M. Silverstone, Eds., Copenhagen: Munks- neva: World Health Organization.
gaard, pp. 13-27. ZIMMERMAN, E.R. (1954): Fluoride and Nonfluoride Enamel
SPAK, C.-J. and EKSTRAND, J. (1988): Effect of the Food Intake Opacities, Publ Health Rep 69:1115-1120.

Downloaded from [Link] at Apollo Group - UOP on February 17, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research