Fluoride in Caries Prevention: Is the Effect Pre- or Post-eruptive?

A. GROENEVELD, A.A.M.J. VAN ECK, and O. BACKER DIRKS

Department of Community Dental Health and Epidemiology, Netherlands Institute for Preventive Health Care, PO Box 124,
2300 AC Leiden, The Netherlands

A longitudinal study of children from ages 7 to 18 showed that, if Disagreements about the pre-eruptive effect of fluoride prob-
enamel lesions were included, the overall number was the same in ably result from this difference.
fluoridated and non-fluoridated areas. However, a significant redu- In the Dutch water fluoridation study, scoring of enamel
cing effect of pre-eruptive fluoride could be seen in the number of
dentinal lesions in a fluoridated area, provided that fluoride was also lesions was performed longitudinally every two years, and
consumed post-eruptively for a considerable period of time. A precise children were followed up from 7 to 18 years of age (Backer
estimation of both pre- and post-eruptive effects was obtained when Dirks, 1966). This study, carried out between 1953 and 1971,
teeth were classified according to their eruption time as related to the is the most appropriate for correlation of in vivo with in vitro
onset of water fluoridation. About 66% of the greatest reduction in results.
pit and fissure caries came from pre-eruptive fluoride, while in smooth
surfaces, this effect was reduced to 25%. In approximal surfaces, the
reduction was due halfto pre- and halfto post-eruptive fluoride. Post- The enamel lesion.
eruptive fluoride became more important with decreasing severity of
caries attack. Thus, assuming a continuing decline in dental caries, The early stages of the enamel dissolution process occur at
the majority of such populations will benefit most from the use of a submicroscopic level. Clinically, the initial enamel lesion
topical fluorides. However, in high-risk groups, supplementation of appears at a relatively late stage when the loss of mineral
pre-emptive fluoride can still be of major importance. becomes visible as a white spot.
The enamel lesion can be seen in its most pure form on the
J Dent Res 69(Spec 155):751-755, February, 1990 free smooth surfaces of teeth. In pits and fissures, these initial
lesions are soon stained, and on approximal surfaces they are
Introduction. seen on radiographs as areas of dissolution confined to the
enamel. On bite-wing radiographs, the number of approximal
Since dental fluorosis is caused by pre-eruptive consumption enamel lesions is always underestimated (Rugg-Gunn, 1972),
of fluoride, it is not surprising that the cariostatic properties but by appropriate calculation it is possible for a more realistic
of fluoride were thought to be due to the incorporation of estimate of the actual number to be obtained (Groeneveld,
fluoride into the enamel. 1985).
McKay stated, in a review on water fluoridation as late as Shown in Fig. 1 are the total number of lesions grouped for
1952, that "it is not necessary to continue the use of fluori- the fluoride (Tiel, 1 ppm F-) and non-fluoride (Culemborg,
dated water after the enamel has been calcified". However, 0.1 ppm F-) towns. The children included were permanent
some authors (Klein, 1946; Russell, 1949a, 1949b) had pre- residents of their towns and participated in every examination.
viously demonstrated an effect of the post-eruptive consump- Those living in fluoridated Tiel had received fluoride contin-
tion of fluoride. Later, Backer Dirks et al. (1961) and Marthaler uously since birth.
(1967) confirmed this observation and drew attention to the All examinations were completed in duplicate and, in cases
fact that the pre-eruptive effect would be lost unless fluoride of differing diagnoses, both investigaters re-examined that sur-
was also consumed post-eruptively. face again. The error in diagnoses was thus kept as small as
In vitro studies revealed that the reduction in enamel solu- possible (Backer Dirks, 1966). .
bility by pre-eruptive incorporation of fluoride is small
(Newesely, 1972). It is therefore unlikely that the ability of
fluoride to produce a decrease in the rate of enamel dissolution
when incorporated into the apatite lattice, mainly as fluoridated
N/CHILD
70
mAl
r7:1
i lesions F-
hydroxyapatite, plays an important role in the observed caries
reduction (Larsen, 1973; Fejerskov et al., 1981). It is now tillJ All lesions F+
60
believed that the effectiveness of fluoride is due to its presence
in the aqueous phase during enamel dissolution and that it acts .D-Lesions F-
50
in three different ways: (a) inhibition of demineralization, (b)
enhancement of remineralization, and (c) by effects on bacteria 40
~ D-Lesions F-+
(Ten Cate and Duijsters, 1983a, 1983b; Theuns, 1986).
It is tempting to compare these in vitro findings with the 30
results of clinical studies. However, most clinical studies have
used the DMF-T or DMF-S index, and therefore only present 20
data on the prevalence of cavities and dentinal lesions. Clinical
studies have revealed information about the occurrence of enamel
lesions, while in vitro studies deal with enamel lesions only.
o 7 9 11 13
Presented at a Joint IADR/ORCA International Symposium on Fluo-
rides: Mechanisms of Action and Recommendations for Use, held Fig. i-Total number of lesions and number of dentinal lesions per
March 21-24, 1989, Callaway Gardens Conference Center, Pine child in a fluoridated and a non-fluoridated area. Children were followed
Mountain, Georgia from 7 to 18 years of age.
751

Downloaded from jdr.sagepub.com at UNIV OF WESTERN ONTARIO on February 10, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

752 GROENEVELD et al. J Dent Res February 1990

At the age of 18 years, both groups registered the same FREE SMOOTH SURFACES
AGE AT START
number of lesions, with the small differences in the younger
age groups apparently indicating a retardation of dental caries % REDUOCOTION
1
fWiiI 8888l
~ ~ ~
%V
initiation in Tiel. o 4 8
Also shown is the ratio of cavities to all lesions for Culem-
borg (non-F) and Tiel (F), respectively. In 18-year-olds, about
50% of lesions were confined to enamel in the non-fluoridated
area, and about 75% in the fluoridated area. The larger number
of enamel lesions in the fluoridated area suggests that pro-
gression was slower.
Pot et al, (1977) showed that a larger number of white-spot
enamel lesions remained unchanged in fluoridated Tiel than in
CuIemborg, over a period of six years.
9 11 13 15
The similar number of lesions suggests a small difference AGE IN YEARS AT EXAMINATION
in solubility between enamel mineral of both areas, despite a
higher fluoride concentration in the outer layers of fluoridated Fig. 2-DMF-S percentage reductions of free smooth surfaces at dif-
teeth. ferent examination ages in three groups of children with different ages at
onset of water fluoridation. Arrow indicates the percentage reduction in
Lemke et al. (1970) found that after discontinuation of water one separate group consuming fluoride from age 7 and examined at age
fluoridation, the caries incidence of children increased to the 12.
level of those in the non-fluoridated area in a period of 4 to 6
years. These types of data provide strong support for a post- PITS AND FISSURES
eruptive effect. Also, the observation that the pre-eruptive ef- AGE AT START
fect will soon be lost without post-eruptive continuity points % REDUCTION. IliI [iliI
to a post-eruptive fluoride effect (Backer Dirks et al., 1961; 60 0 4 8
Marthaler, 1967). Hence, it can be concluded that there is no
post- or pre-eruptive effect of fluoride on the initiation of den-
tal caries. Progression of enamel caries is mainly affected post-
eruptively by inhibition of demineralization and promotion of
early enamel lesion remineralization.

The dentinal lesion.

9 11 13 15
As Murray (1976) has pointed out, results from water fluor- AGE IN YEARS AT EXAMINATION
idation studies in different countries showed remarkably sim-
ilar dentinal caries reductions ranging from 50 to 60% in areas Fig. 3- DMF-S percentage reductions of pits and fissures at different
examination ages in three groups of children of different ages at the onset
with controlled water fluoridation. of water fluoridation.
Also from these studies it can be concluded that the effect
increases with decreasing age at start of water fluoridation, % REDUCTION APPROXIMAl
with the maximum effect achieved when fluoride was con-
sumed continuously from birth.
Marthaler (1960) discussed these data and came to the widely
accepted conclusion that the post-eruptive effect was increased
for each year of pre-eruptive fluoride consumption. However,
the pre-eruptive contribution of fluoride was not similar for
different sites of the tooth (Groeneveld, 1986).
A reduction of 67% of cavities in free smooth surfaces oc-
curred at age 15 when consumption started as late as age 8 9 8 7 5 4 o
(Fig. 2), while children consuming fluoridated water from 7 AGE AT START
years old showed 75% fewer cavities at age 12. From these FigA- DMF-S percentage reductions of approximal surfaces at age 15
data it follows that smooth surfaces benefited most by post- in six groups of children with different ages at the onset of water fluori-
eruptive fluoride. However, for pits and fissures, the reduction dation.
at age 15 (when fluoride started at age 8) was only 12%, i.e.,
less than one-third the maximum effect (Fig. 3). Hence, pre-
eruptive fluoride was more important for pits and fissures than types, fluoride was available pre-eruptively even if consump-
for smooth surfaces. Here the maximum effect was attained tion was delayed until age 8, e.g. , premolars and second mo-
when fluoride was taken from birth, but three-quarters of this lars. Although a fair indication about the magnitude of pre-
effect was achieved at ages 13 and 15 years when fluoride and post-eruptive fluoride effects can be obtained from the
consumption started at 4 years of age. above results, a more precise estimation is derived when teeth
The relation between percentage caries reduction and age at are classified according to their time of eruption in relation to
start is shown in Fig. 4 for approximal surfaces in children start of fluoride consumption.
examined at age 15. By starting at age 8, the reduction was Van Eck (1987) analyzed the Dutch data on water fluori-
approximately 50%, while starting at age 4 produced near- dation, and in Fig. 5 the data on approximal surfaces from 15-
maximum results, the difference being only 5%. For approx- year-old children, given as percentage reductions, were re-
imal surfaces, the contribution of pre-eruptive fluoride seemed arranged as a function of the annual time intervals between
to have a value somewhere between that for fissures and that start of water fluoridation and eruption. Here, when water
for free smooth surfaces. However, for a number of tooth fluoridation coincided with eruption (year 0), the percentage

Downloaded from jdr.sagepub.com at UNIV OF WESTERN ONTARIO on February 10, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

Vol. 69 Special Issue PRE- AND POST-ERUPTIVE EFFECTS OF FLUORIDE 753

APPROXIMAL % REDUCTION APPROXIMAL

80 POST POST- +PRE-ERUPTIVE
% REDUCTION
POST POST- +PRE-ERUPTIVE ~

60 60

40 40

20 20

-1 0 1 3 4 5 6 7 4 3 2 1 0
YEARS BETWEEN START F/H 20 AND ERUPTION
AGE AT START
Fig. 5-DMF-S percentage reductions of approximal surfaces. Data Fig. 7-0bserved percentage reductions (bars) of caries in first molar
from 15-year-old children were re-arranged in relation to the time interval, approximal surfaces at age 15 in children with different ages at the onset
in years, between onset of water fluoridation and time of eruption. Col- of water fluoridation. Lines A and B indicate, theoretically, curves of pure
umns are not related to a group of children, but to collections of teeth post- and pre-eruptive effects, respectively. Line C indicates similar pre-
with the same time intervals. and post-eruptive effects.

DMF-S/CHILD % REDUCTION APPROXIMAL

25 80 POST-+PRE-ERUPTIVE PRE

fl:!lCulemborg

20 DTlel 60

15

10 ~
: IJ
"'~::::";:
~~ ~~
.:.
:i :.
:::
::
o ii::::::::
I ~1
15 14 13 12 11 10 9
5 II
:
::
~:~
) \\\\
j: !I!!I Fig. 8-0bserved percentage reductions (bars) of caries in approximal
0 surfaces of the first molar at age 15 in children with different ages at
2.5 3.5 45 5.5 6.5 7.5 85 9.5 10 11 12 13 14 15 termination of water fluoridation. Lines A and B indicate, theoretically,
AGE UNTIL F/H20 curves of pure post- and pre-eruptive effects, respectively. Line C indicates
Fig. 6-DMF-S per child from nine different groups of children at age similar pre- and post-eruptive effects.
15 in a fluoridated (Tiel) and a non-fluoridated area (Culemborg). Data
on the horizontal axis indicate the ages at termination of water fluoridation.
factors were involved in the observed decline of the caries
prevalence. On the contrary, the decrease in caries experience
caries reduction was about half the greatest reduction. With in non-fluoridated Culemborg had already started in 1969
respect to free smooth and occlusal surfaces, the topical effect (Groeneveld, 1988), and at that time only 10 to 15% of tooth-
was calculated to be 76% and 27% of the highest reduction, paste was available with fluoride (Kalsbeek and Verrips, 1989;
respectively (Van Eck, 1987). this symposium).
The percentage caries reductions for approximal surfaces of
the first molar, as a function of the age at the start and at the
Discontinuation of water fluoridation. end of fluoridated water consumption, are shown in Figs. 7
Clinical studies at the start of water fluoridation can provide and 8. In both Figs., hypothetical curves are drawn indicating
data on the pure post-eruptive effect of fluoride. They can give the level of reduction when the effect was due only to (B) post-
an indication of the pre-eruptive effect only by subtraction of or pre-eruptive (A) fluoride. The midline (C) indicates the
the post-eruptive from the pre- and post-eruptive effects. Stud- situation in which the pre- is as large as the post-eruptive
ies at discontinuation of water fluoridation, however, can pro- effect. It can be seen that the observed reduction levels are in
vide data on the pure pre-eruptive effect, when the eruption of the neighborhood of the diagonally drawn curve. Data for fis-
a particular tooth coincided with the discontinuation of water sures and pits revealed that 60% of the maximum reduction
fluoridation. The DMF-S values in the test and control groups was obtained when water fluoridation was available until erup-
after the 1973 cessation of water fluoridation in Tiel are shown tion. For free smooth surfaces this amounted to 75% of the
in Fig. 6. There was little reduction in DMF-S at age 15, when maximal reduction, but again it must be borne in mind that
fluoride was consumed until the age of about 3 1/2 years. For the topical use of fluoride increased in the study period.
a significant reduction to occur, fluoride had to be consumed
at least until age 6 1/2. It is, however, difficult to assess the
value of the pre-eruptive effect, because not only did the use Deciduous dentition.
of topical fluoride increase dramatically in that period, but Very few studies have been carried out on the effect of
there was also a decline in caries prevalence of the non-fluor- fluoride on the deciduous dentition. Reduction percentages
idated area. In both groups fluoride was used, post-eruptively, similar to those of the permanent dentition were found in the
by toothpaste, topical application, or similar fluoride supple- United States (Tank and Storvick, 1964) and England (Murray,
mentations. Nevertheless, that does not mean that no other 1969). However, data from these studies are not suitable for

Downloaded from jdr.sagepub.com at UNIV OF WESTERN ONTARIO on February 10, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

754 GROENEVELD et al. J Dent Res February 1990

differentiation between pre- and post-eruptive effects of fluo- ternative, particularly in countries exhibiting a large decline in
ride, mainly because of the relatively short formative period, caries prevalence. It is also conceivable that efforts focused
which requires a high frequency of examination. on the individual, especially in relation to dietary habits and
Some studies have suggested a pre-natal effect of fluoride. sweets consumption, may result in a decrease in the severity
However, Carlos et al. (1962), Horowitz (1967), and Backer of caries attack. However, in high-risk groups it would be
Dirks (1967) could not find an indication of an extra effect of unwise not to take advantage, on an individual basis, of the
pre-natally consumed fluoride. While some fluoride tablet trials fact that pre-eruptive fluoride is capable of increasing the top-
have claimed a pre-natal effect, this may be ascribed to the ical F- effect significantly. In such cases, it is quite possible
greater dental awareness of parents, which can be an important that the added benefit of caries reduction by pre-eruptive fluo-
disturbing variable in such studies (Tijmstra, 1985), rather than ride exceeds the disadvantage of the chance of slight fluorosis.
to a true fluoride effect. Nonetheless, because the post-eruptive effect attains increasing
importance with decreasing severity of the caries attack, in the
future (assuming the current caries decline continues) the ma-
Discussion. jority of such populations will benefit from the use of topical
It is evident that fluoride has an important pre-eruptive effect fluorides. However, for non-motivated persons, community-
on the caries experience in all permanent dentition predilection based fluoridation must still play a large part in their dental
sites. The maximum DMF-S reduction in a fluoridated area at well-being.
age 15 was due about half to the pre-eruptive and about half
REFERENCES
to the post-eruptive effect of fluoride.
The initiation of the caries lesion is hardly affected by fluo- BACKER DIRKS, 0.; HOUWINK, B.; and KWANT, G.W. (1961):
ride from drinking water. The cariostatic effect occurs when Some Special Features of the Caries Preventive Effect of Water
the lesion has progressed to the dentin following cavitation. Fluoridation, Arch Oral Biol 4 (Sp.SuppI.):187-192.
BACKER DIRKS, O. (1966): Posteruptive Changes in Dental Enamel,
The best effect is achieved if fluoride is available from birth, J Dent Res 45:503-511.
but about 85% of the greatest reduction is obtained when fluo- BACKER DIRKS, O. (1967): The Relation Between the Fluoridation
ride consumption starts between ages 3 and 4. of Water and Dental Caries Experience, Int Dent J 17:582-605.
In surfaces with high. caries susceptibility, e.g., pits and BACKER DIRKS, O. and KALSBEEK, H. (1987): Vermindering
fissures, the greater part of the reduction is derived from pre- van Tandcaries in Geindustrialiseerde Landen, Ned Tijdschr Ge-
eruptive fluoride, while in surfaces with a low susceptibility, neeskd 131:1902-1907.
i.e., smooth surfaces, post-eruptive fluoride is more signifi- CARLOS, J.P.; GITTELSOHN, A.M.; and HADDON, W., Jr. (1962):
cant. However important the contribution of a pre-eruptive Caries in Deciduous Teeth in Relation to Maternal Ingestion of
Fluoride, Publ Health Rep 77:658-660.
effect is to the total effect of fluoride, it does not mean that a
FEJERSKOV, 0.; THYLSTRUP, A.; and LARSEN, M.J. (1981):
similar reduction cannot be achieved by the use of topical Rational Use of Fluoride in Caries Prevention, Acta Odontol Scand
fluoride alone. Several studies have shown caries reductions 39:241-247.
comparable in magnitude with that of the highest water fluor- FEJERSKOV, 0.; MANJI, F.; BAELUM, V.; and M0LLER, I.J.
idation effect when different forms of topical fluoride supple- (1988): Dental Fluorosis-A Handbook for Health Workers,
ments were used. Since the topical effect of fluoride is inversely Copenhagen: Munksgaard.
related to caries susceptibility, as shown above, it seems logi- GROENEVELD, A. (1985): Longitudinal Study of Prevalence of
cal that the effectiveness of topical fluorides will increase more Enamel Lesions in a Fluoridated and Non-Fluoridated Area, Com-
during a period in which caries prevalence declines, provided munity Dent Oral EpidemioI13:159-163.
that the caries incidence decrease has not been entirely due to GROENEVELD, A. (1986): Clinical Observationson Caries and Water
Fluoridation. In: Tooth Development and Caries. Vol. II, F.C.M.
an increase in the use of topical fluorides. There are, however, Driessens and J.H.M. Woltgens, Eds., Boca Raton, FL: CRC
indications that better dietary habits and altered sugar intake Press, pp.147-184.
(Backer Dirks and Kalsbeek, 1987) are important contributors GROENEVELD, A. (1988): De Gebitstoestand van 15-Jarigen, Ned
to the caries decrease, at least in The Netherlands. Tijdschr Tandheelkd 95:307-311.
HOROWITZ, H.S. (1967): Effects of Prenatal Exposure to Fluori-
dation on Dental Caries, Publ Health Rep 82:297-304.
Implications for the rational use of fluoride. KLEIN, H. (1946): Dental Caries (DMF) Experience in Relocated
Children Exposed to Water Containing Fluoride, II, J Am Dent
From the above it can be stated that topical fluoride has a Assoc 33:1136-1141.
significant caries-reducing effect which can be increased by LARSEN, M.J. (1973): Dissolution of Enamel, Scand J Dent Res
supplementation of pre-eruptive fluoride. Even when topical 81:518-522.
fluoride regimes attain effects as large as those from water LEMKE, Ch. M.; DOHERTY, J.M.; and ARRA, M.C. (1970): Con-
fluoridation, the possibility exists that this effect would have trolled Fluoridation: the Dental Effects of Discontinuation in An-
been greater if fluoride had also been given during tooth for- tigo, Wisconsin, JAm Dent Assoc 80:782-786.
MARTHALER, T.M. (1960): Kariesstatistische ResuItate der Trink-
mation. wasserfluoridering im Bleibenden Gebiss und Was Sie von der
However, it is not clear in which way pre-eruptive fluoride Salzfluoridierung Erwarten Lassen, Schweiz Mschr Zahnheilk
should be administered, since water fluoridation has not been 70:315-333.
implemented widely in most countries, and addition of fluo- MARTHALER, T.M. (1967): The Value in Caries Prevention of Other
ride, even in very low concentrations, to food, beverages, and Methods of Increasing Fluoride Ingestion, Apart from Fluoridated
sweets would have the disadvantage of uncontrolled fluoride Water, Int Dent J 17:606-618.
uptake, which may be harmful. Fluoride drops and tablet McKAY, F.S. (1952): The Study of Mottled Enamel (Dental Fluo-
administration can be used, but there is a chance of low-grade rosis), JAm Dent Assoc 44:133-137.
MURRAY, J.J. (1969): Caries Experience of Five-Year Old Children
fluorosis (Fejerskov et al., 1988). Furthermore, the daily in-
from Fluoride and Non-Fluoride Communities, Br Dent J 127:387-
take of drops and tablets requires a high degree of subject 388.
cooperation. MURRAY, J.1. (1976): Fluorides in Caries Prevention, Bristol:
In Europe, where water fluoridation has not been imple- John Wright and Sons Ltd.
mented in most countries, individual prescriptions are the al- NEWESELY, H. (1972): Mechanism and Action of Trace Elements

Downloaded from jdr.sagepub.com at UNIV OF WESTERN ONTARIO on February 10, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research

Vol. 69 Special Issue PRE- AND POST-ERUPTIVE EFFECTS OF FLUORIDE 755

in the Mineralization of Dental Hard Tissues. Nyon, Switzer- dren One to Six Years Old in Two Oregon Communities (Corvallis
land: Zyma S.A., p.59. and Albany). 1. Effect of Fluoride on Caries Experience and Erup-
POT, T.J.; GROENEVELD, A.; and PURDELL-LEWIS, D.J. (1977): lion of Teeth, JAm Dent Assoc 69:291-294.
The Origin and Behaviour of White Spot Enamel Lesions, Neth TEN CATE, J.M. and DUIJSTERS, P.P.E. (1983a): Influence of
Dent J 85:6-18. Fluoride in Solution on Tooth Demineralization, Caries Res 17:193-
RUGG-GUNN, A.J. (1972): Approximal Caries Lesions, A Com- 199.
parison of the Radiological and Clinical Appearances, Br Dent J TEN CATE, J.M. and DUIJSTERS, P.P.E. (1983b): Influence of
133:481-484. Fluoride in Solution on Tooth Demineralization, Caries Res 17:513-
RUSSELL, A.L. (1949a): Dental Effects of Exposure to F1uoride- 519.
bearing Dakota Sandstone Waters at Various Ages and for Various
Lengths of Time. 1. Status of the Permanent Teeth of 339 Children THEUNS, H.M. (1986): Enamel Caries and Fluoride. In: Tooth De-
Aged 11 to 15 Years Who Used Such Water for Eighteen Months velopment and Caries. Vol. II. F.C.M. Driessens and J.H.M.
Prior to Eruption of the First Permanent Molars, J Dent Res 28:298- Woltgens, Eds., Boca Raton, FL: CRC Press, pp.66-109.·
309. TIJMSTRA, T.; BRINKMAN-ENGELS, M.; and GROENEVELD,
RUSSELL, A.L. (1949b): Dental Effects of Exposure to Fluoride- A. (1978): Effect of Socio-economic Factors on the Observed Car-
bearing Dakota Sandstone Waters at Various Ages and for Various ies Reduction After Fluoride Tablet and Fluoride Toothpaste Con-
Lengths of Time. II. Patterns of Dental Caries Inhibition as Related sumption, Community Dent Oral Epidemiol 6:227-230.
to Time Since Exposure, and to Periods of Calcification and Erup- VAN ECK, A.A.M.J. (1987): Pre- and Posteruptive Effect of Fluor-
tion, J Dent Res 28:600-612. idated Drinking Water on Dental Caries Experience. Thesis, Uni-
TANK, G. and STORVICK, C.A. (1964): Caries Experience of Chil- versity of Utrecht, NIPG-TNO, No. 87021.

Downloaded from jdr.sagepub.com at UNIV OF WESTERN ONTARIO on February 10, 2015 For personal use only. No other uses without permission.

© 1990 International & American Associations for Dental Research