Archs oral Eiol. Vol. 13, pp. 877-885, 1968. Pergamon Press. hinted in Gt. Britain.

AN ENAMEL BIOPSY METHOD FOR DETERMINATION OF

FLUORIDE IN HUMAN TEETH

F. BRUDEVOLD,
H. G. MCCANN and P. GRIN
Forsyth Dental Center, Boston, Mass. 02115, U.S.A.

Sunnnary-An enamel biopsy method has been developed which involves polishing off
about 0.2 mg enamel from accessible surfaces of single teeth. This is accomplished with a
rotating midget felt cone impregnated with silicon carbide. The cone is coated with
glycerine which traps and forms a slurry with the ground enamel particles. The slurry
and the felt cone are transferred to a plastic test tube and subjected to F and Ca analysis,
employing an F electrode and atomic absorption spectrophotometry. Findings showed
that the levels of F in intact anterior teeth from the same individual usually were similar,
and that mean concentrations varied from 400 to 2500 ppm in the anterior teeth of the
different persons studied. A second biopsy taken from the same tooth surface showed
lower concentrations of F than the first biopsy, confirming previous findings in extracted
teeth of F gradient in enamel. It was calculated that a layer of enamel between 1 and 2~
in thickness usually was removed by the biopsy procedure. Patients have no objection
because the treatment is similar to a pumice prophylaxis and produces a highly polished
surface. The biopsy procedure is harmless to the patient, takes little chair time and is
simple to perform.

INTRODUCTION

PRESENT knowledge of fluorine in human enamel has been derived from analyses
of extracted teeth. New and valuable information could be obtained if it were
possible to determine F in enamel of teeth in situ. It would then be feasible to study
the relations between the level of F in enamel and susceptibility to caries and to investi-
gate the deposition of F from ingested and topically applied F on a larger scale than
the limited availability of extracted teeth permits. It would be more important to
determine the level of F in the surface enamel than in the bulk of the enamel because
resistance to caries is related to the relatively high concentrations of F in the surface,
and F reactions with fully formed enamel are restricted to the outermost portion.
No biopsy method for determining F in surface enamel has been reported in the
literature. CANDELI et al. (1967) measured F in pieces of enamel which had been
taken from deciduous molars with burs and a scalpel. Samples of 3-12 mg of enamel
were employed, and the tooth material removed was replaced with amalgam. This
method can be used only in exceptional cases, and surface-limited F reactions may
not necessarily be revealed by the analyses of pieces of enamel.
The purpose of this paper is to describe a biopsy procedure for determining F
in the outer 1-2~ layer of enamel and to present some findings obtained in individuals
with different levels of F in the enamel. The F in the minute sample was determined
by F electrode. Biopsies were taken from a group of military personnel in order to
demonstrate the usefulness of the method in a random population.
877
878 F. BRUDEVOLD,H. G. MCCANN AND P. GRIN

MATERIALS AND METHODS

The following supplies are needed :
Midget felt cones 0-25x0.5 in. (William Dixon Inc., Newark, N.J.). The cones
may be shortened with a razor blade since smaller cones are easier to work with.
Felt cylinders 0.25 x 0.25 in. made to order by the manufacturers are recommended.
(Large batches may be obtained from Bacon Felt Company, 11 Fifth Street,
Taunton, Mass. 02781.)
Screw mandrel with shoulder for straight handpiece.
Tissue forceps No. 16 (Hu-Friedy Co., Chicago, Ill.).
Disposable 10 ml syringe (Am. Hospital Supply Corp., Evanston, Ill.) equipped
with 1 in. hypodermic needle No. 26.
Sterile disposable stoppered test tubes, 17 x 100 mm (Falcon Plastics No. 2001).
Cotton pellets No. 2.
Cotton rolls.
Kleenex tissues.
Silicon carbide, grit 600 (Carborundum Co., Niagara Falls, N.Y.).
Glycerine (reagent).
The silicon carbide, felt cones and cotton pellets must be purified prior to use.
After purification they are tested for F and Ca by the same procedures as used for
the analysis of biopsies. If any F or Ca is found in any of these materials the purifica-
tion is repeated. The silicon carbide is suspended in 0.5 M HCIO1, and the felt cones
and cotton pellets in 0.1 M HCIO, for at least 1 hr. They are then washed repeatedly
in deionized water until the washings are free of F and Ca. The cotton pellets must
be washed gently in order to avoid clumping. Following these procedures the silicon
carbide and pellets are dried and stored in closed containers. The felt cones are sus-
pended in ethanol and then shaken in a suspension of purified silicon carbide and
glycerine until they are completely covered by the abrasive. Some penetration of
the abrasive into the felt can be achieved by supersonic vibration, but vigorous
mechanical shaking is adequate. The cones thus prepared are stored wet with glycerine
in a closed container.
The biopsy is carried out as follows:
(1) The mounted mandrel and the tissue forceps are rinsed in a stream of
deionized water from the syringe and wiped with Kleenex.
(2) A felt cone is mounted on the mandrel with forceps.
(3) The tooth to be biopsied is isolated with cotton rolls, soft debris is removed,
and the surface dried with compressed air.
(4) The biopsy is performed by polishing the surface for 40-60 sec. The surface
must be polished evenly, in order to remove a uniform enamel layer, and
at the lowest speed of the engine in order to avoid splashing. The enamel
particles ground off will collect in the glycerine on the cone or in the
glycerine-abrasive slurry which is deposited on the tooth during polishing.
To accomplish this the cone must be well coated with glycerine.
(5) The sampled enamel with the abrasive and the cone are now transferred
to the plastic test tube. Using the forceps the slurry which has accumulated
 DETERMINATION OF FLUORIDE IN HUMAN TEETH 879

on the tooth is wiped off with a cotton pellet moistened with deionized
water. The pellet is placed in the bottom of the tube, and the procedure is
repeated with another pellet to assure complete collection. The felt cone
is also placed in the test tube with the forceps, and any slurry left on the
forceps is washed into the tube with a few drops of deionized water.
(6) The test tube is stoppered and stored until chemical analysis can be under-
taken.

Chemical analysis
Fluoride is determined by means of the fluoride electrode (Orion Research, Inc.)
using the procedure for analysis of enamel described in detail in a recent publication
by MCCANN(1968). One ml of O-5 M HClO, is added to the biopsy tube and allowed
to stand for at least 1 hr in order to dissolve the enamel present. Four ml of O-5 M
Na, citrate is then added, the tube stoppered and contents thoroughly mixed by
shaking. After the abrasive has settled the supernatant is decanted into a plastic
cup and the F concentration obtained with the electrode. Calcium is determined by
atomic absorption spectrophotometry on aliquots diluted 1 : 5. The amount of
enamel in the sample is calculated assuming a calcium content in enamel of 38 per cent.
Since the amount of fluoride in some biopsy samples is near the lower limit of
sensitivity for the electrode (0.05pg F/5ml) it is important that the HClO, and sodium
citrate be very low in F. Concentrated HClO, is purified by adding 100 ml of deionized
water to 1 litre of the 60 per cent reagent grade acid and evaporating to the original
volume. Mallinckrodt Analytical Reagent Grade Na,C,H,O,.2H,O has been
found by analysis with the electrode to be essentially free of F (less than O*OOOOl
per cent). However it does contain a trace of calcium. Since both F and Ca may
vary from batch to batch, new fluoride and calcium standards should be made if
a different citrate is used. Fluoride standards should be made up to be O-4 M in
citrate, O-1 M in perchlorate, and to contain 0,0~02,0~05,0~1,0~25,0~5, 1 and 1Opg
F/5*0 ml of sample. Calcium standards should be 0.08 M in citrate and 0.02 M in
perchlorate (allowing for the 1 : 5 dilution). A finding of 1 y Ca/ml at this dilution
is equivalent to 0.066 mg of enamel in the original biopsy.
The method was tested in vitro on blocks of enamel. A number of biopsies were
then taken from the labial surfaces of the upper anterior teeth of young or middle
aged adults at a military base. In some instances a second biopsy was obtained either
a few days or 4-6 weeks later in order to find out how the level of F in the enamel surface
was affected by the procedure. Only intact teeth were included, and all the biopsied
teeth had been given a prophylaxis the previous day. The biopsies were taken by 2
dental hygienists who previously had been instructed in the method.

RESULTS
Blocks of enamel with the surface layer removed so that a relatively low and uniform
fluoride content would be obtained were used to check the correctness of measure-
ments obtained by the biopsy procedure. Successive layers were shown to contain
540 ppm F by etching vs. 570 by biopsy and 230 by biopsy vs. 250 by etching. It was
880 F. BRUDEVOLD,H. G. MCCANN AND P. GRIN

also shown by polishing hard tool steel that no fluoride was released from the felt
cones or the abrasive as a result of exposure of new surfaces through the abrasive
action.
Table 1 shows some results obtained on the labial surfaces of the 6 upper anterior
teeth. Data from persons with different levels of F are included in order to demonstrate
how the biopsy method performs within the range of concentrations which may be
expected in a random population. Patient 1 and 2 have low levels of enamel F with
the same means of 430 ppm and respective ranges from 250 to 600 ppm and from 400
to 550 ppm. The amount of biopsied enamel ranges from O-07 to O-26 mg and the
amount of F in the samples from 0.04 to 0.11 pg. The concentrations of F are fairly
consistent and suggest that a relatively constant level of F may be expected in the

TABLE 1. BIOPSIESFROM LABIAL SURFACESOF UPPER ANTERIOR TEETH OF PERSONS WITH MW AND
HIOH LEVELS OF F IN THESURFACEENAMEL

Risht Left
Canine Lateral Central Central Lateral Canine Mean SD.

Patient 1 24 yr
Enamel (mg) 0.10 0.21 o* 15 0.16 0.20 0.07 o-15 o-055
Fluoride &g) o-05 0.07 0.07 O-04 0.09 O-04 0.06 o-020
Fluoride @pm) 500 350 500 250 450 600 430 117

Patient 2 27 yr
Enamel (mg) o-11 0.20 o-20 0.11 0.26 0.18 0.065
Fluoride &g) O-04 O-08 O-09 O-06 0.11 0.08 O-027
Fluoride @pm) 400 400 450 550 400 430 74

Patient 3 19 yr
Enamel (mg) 0.10 0.11 o-14 o-10 0.11 O-08 0.11 o-020
Fluoride bg) o-07 0.07 0.10 0.08 0.08 0.10 0.08 o-014
Fluoride (ppm) 700 650 700 800 750 1250 800 224

Patient 4 38 yr
Enamel (mg) o-19 0.18 0.21 0.20 O-18 0.11 O-18 0.035
Fluoride &g) O-28 0.34 0.34 o-33 0.19 0.16 0.27 O-080
Fluoride (ppm) 1450 1900 1600 1650 1050 1450 1520 277

Patient 5 20 yr
Enamel (mg) o-10 o-15 0.20 o-15 o-14 0.15 O-036
Fluoride (pg) o-22 O-29 O-42 0.28 0.29 0.30 o-073
Fluoride (ppm) 2200 2000 2150 1950 2150 2080 115

Patient 6 20 yr
Enamel (mg) O-08 O-27 o-19 0.12 0.17 0.074
Fluoride bg) 0.25 0.44 O-38 O-24 0.35 0.099
Fluoride (ppm) 3150 1650 2000 2000 2190 561

Patient 7 35 yr
Enamel (mg) 0.08 o-15 0.23 o-19 O-18 0.17 0.056
Fluoride (pg) O-25 0.33 o-45 O-52 O-48 0.41 0.112
Fluoride @pm) 3150 2200 1950 2750 2650 2540 462
 DETERMINATION
OF FLUORIDEIN HUhfANTEETH 881

anterior teeth of an adult. The lowest and highest concentrations found in patient
1,250 ppm in the left central and 600 ppm in the left canine, were obtained in samples
containing only 0.04 pg of F, and may reflect analytical inaccuracy.
Patient 3 has higher levels of F in the biopsies than patients 1 and 2; the mean is
800 ppm and the range is from 650 to 1250 ppm. The concentration is about the same
in all the teeth except the left canine. The high value of 1250 ppm in this tooth could
be due to sampling of a particularly thin layer of enamel in view of the small weight
of the biopsy.
The biopsies of patient 4 range from 1050 to 1900 ppm and the mean is 1520 ppm
of F. The 1900 ppm sample was obtained on the right lateral which was heavily
stained. In this case the two canines show the same level of F (1450 ppm) in spite of
a marked difference in weight of the biopsies (0.11 and 0.19 mg respectively). The
two biopsies may have been taken to the same depth, but have extended over different
sized areas.
In patient 5, F concentrations are very close to the mean of 2080 ppm in all
teeth biopsied. Again consistent results were obtained although the weight of the
biopsies from homologous teeth differed significantly.
Patients 6 and 7 also show high F concentrations with means of 2190 and 2540
ppm, but in both some of the teeth differ considerably in concentrations as shown
by the wide range from 1650 to 3 150 ppm in patient 6 and from 1950 to 3150 ppm
in patient 7.
Table 2 shows results of two biopsies taken from the same labial area of upper
anterior teeth. In patient 8 the biopsies were taken on 2 successive days and in patients
9 and 10 five weeks apart. With one exception (patient 9, left canine) the first biopsy
shows higher concentrations of F than the second biopsy, reflecting the well-known
gradient of F in enamel from the surface inward.
The sample size of the first biopsies ranged from 0.12 to 0.24 mg and the amounts
of F in the samples from 0.05 to 0.27 pg. All these biopsies contained sufficient

TABLE 2. Two BIOPSIESFROM THE SAMELABIAL AREA OF UPPER ANTERIORTEETHOF YOUNG ADULTS

Patient No. 8 20 yr Patient No. 9 24 yr Patient No. 10 32 yr

Tooth First Second First Second First Second

Right canine 1800 550 450 200 550 150

lateral 1150 800 350 350
central 1150 500 700 350 650 150
Left central 1500 600 500 450 300 200
lateral 350 300 550 200
canine 450 650 300 200
Mean 1390 600 470 390 460 170
&SD. 320 136 132 161 165 39
Enamel (mg) o-17 0.16 0.16 o-10 o-17 o-13
+ S.D. o-029 o-042 0*04O o-071 0.037 0.049
F (& O-23 0.09 0.08 0.04 0.07 0.02
f S.D. 0.042 0.010 O-027 o-035 0.013 0.008
882 H. G. MCCANNANDP. GRIN
F. BRUDEVOLD,

amounts of F for accurate analysis. This was not always true in regard to the second
biopsies. For example in patients 9 and 10 the F ranged from O-01 to O-11 pg with
means of 0 -04 and 0 -02 pg respectively. Although F determinations may be grossly
inaccurate for such minute amounts, the concentrations of F were within the expected
range, suggesting that analytical errors were small.
DISCUSSION
One of the prerequisites of an enamel biopsy procedure is that it be harmless
to the patient. It must also provide accurate information. Ideally it should be applicable
to all tooth surfaces in a dentition, should be easy to perform, and require little chair
time. The merits of the present biopsy method in regard to these various requirements
will now be considered.
Subjects readily accept the biopsy procedure, probably because the treatment is
similar to the prophylaxis given in a dental office. Biopsies have been taken from over
200 subjects without complaint or apparent discomfort. The biopsied surface is
actually more highly polished than a pumiced surface because of the minute particle
size of the silicon carbide abrasive employed. Although there is a gain in appearance
from the biopsy, the loss of enamel involved might conceivably be a disadvantage.
However, the amount of enamel in a biopsy was usually less than O-2 mg and the
maximal amount was only O-26 mg. Assuming a biopsied surface area of 40 mma,
corresponding to the size of the labial surface of an average lateral incisor, a 0.2 mg
sample would represent a layer with a thickness of l-7 p and a 0.26 mg sample,
a layer with a thickness of 2.2 p. In comparison, an enamel layer of at least 4 p
may be removed by a 30 second pumicing (VRBIC, BRUDEVOLDand MCCANN,
1967). Evidently no more enamel is lost from a biopsy than from a thorough pumice
prophylaxis.
The minimum amount of F needed in a biopsy for accurate analysis is about
0-05-O-10 pg (MCCANN, 1968). However, smaller amounts may be determined with
decreasing accuracy down to about O-01 pg, if before each reading the electrode is
immersed in the standard containing zero F until the mV reading indicates less than
O-01 pg F. Table 3 shows the calculated weights of biopsies which would be required
TABLE3. CALCULATED
WEIGHT,
SURFACE AREA ANDTHICKNKSS
OFBIOPSIESTAKEN
FROM
ENAMEL
WITHDIEEEREN-r
LEVELS OF F ANDCONTAINING
u-5 pg OFF

Area (mm*)
Fill wt. of layer thickness Layer thickness
enamel biopsy 1P 2P area of 40 mm1
@pm) (md (PI

300 O-167 56 28 l-4

500 0.100 33 17 O-8
1000 0.050 17 8 0.4
1500 0.033 11 6 0.3
2000 0.025 8 4 o-2
3000 o-017 6 3 0.1
 DETERMINATION OF FLUORIDE IN HUMAN TEETH 883

to give O-05 pg of F from enamel with different F levels. The areas of enamel which
must be biopsied to provide minimal layer samples with a thickness of 1 or 2 CL,and
the minimal thickness of samples which would be obtained from an area of 40 mm*
are also listed. At the low level of 300 ppm of F the minimal weight of the biopsy is
O-167 mg. At a F level of 1500 ppm the minimal weight of the enamel biopsy is only
O-033 mg and at 3000 ppm 0.017 mg. The areas which must be biopsied or the thick-
ness of enamel which must be removed from a 40 mm* area are correspondingly
reduced. It is evident that there is no problem of obtaining adequate sample size
except when the level of F in the enamel is unusually low. In a random adult popula-
tion, biopsies of nearly O-2 mg will provide ample material for analysis and can be
taken without harming the patient.
The concentrations of F found in the biopsied teeth are in the same range as
those reported in the external layer of extracted teeth from different F areas (BRUDE-
VOLD, STEADMAN and SMITH, 1960). Our finding of lower levels of F in the second
than in the first biopsy taken from the same surface also parallels the observation
of decreasing concentrations of F from the surface inward in enamel of extracted
teeth. This agreement between the results obtained and data in the literature testifies
to the validity of the biopsy findings
No information on previous exposure to F was obtained at the time the biopsies
were taken. After the chemical analysis had been completed only three persons could
be contacted. Patient 2, who had a low F concentration (mean 430 ppm) and patient 6,
with a high F concentration (mean 2190 ppm), were born and lived their first years
in low F areas. They had received no topical F, but used a F dentifrice. Patient 7,
who had a high F concentration (mean 2540 ppm), was born and reared in a commun-
ity with 1 ppm of F in the water supply. He had received 3 topical treatments and also
used an F toothpaste. In biopsy studies aimed at correlating enamel F with past
history of F exposure it would be desirable to utilize groups of similar age and
discontinue the use of F dentifrice several days prior to taking the biopsy.
Ideally the biopsied enamel should be a layer sample of standard thickness in
order to compensate for the marked decrease in concentration of F from the surface
inward. Due to this gradient of F, a heavy layer invariably will be lower in F than a
thin layer of the same enamel. Unfortunately no method is available for removing
enamel layers of uniform thickness, but with some clinical skill consistent results
can be obtained using this simple polishing procedure for sampling. This is apparent
from the frequent finding of similar levels of F in biopsies of the six anterior teeth
from the same person. The marked deviation from the mean sometimes observed in
one or two teeth from the same dentition could be caused by the presence of slightly
demineralized enamel, which tends to be high in F, or to abraded enamel which would
be low in F. Only intact teeth were included in this study, but minute enamel defects,
or uneven functional wear or wear arising from toothbrushing or pumice prophylaxis
could account for the marked difference in F observed in some of the biopsies.
The present biopsies were done by two dental hygienists who had been given a
demonstration and had been made aware of the necessity of avoiding contamination.
The method is very simple to perform on the labial surfaces of upper anterior teeth.
884 F. BRUDEVOLD,H. G. MCCANN AND P. GRIN

The labial surfaces of lower anteriors can also be biopsied without difficulty, but due
to their small size it may be advisable to pool samples from two or more teeth.
Premolars and first molars have been biopsied, but accessibility to the posterior
teeth is limited because of the bulk of the felt cones used. Smaller and more versatile
abrasive points can undoubtedly be designed, but the felt cones have the advantage
of being commercially available, and the combined use of the felt-abrasive-glycerine
to obtain samples is superior to other methods tested. These included the use of
diamond stones and suction devices to collect the ground enamel, and the use of
acid to etch off surface enamel. The former approach produced a rough surface and
there was considerable loss of sample. Etchings gave too low F values because the
dissolved F reacted with the underlying enamel. Often only little or no F was obtained
from enamel known to contain over 1000 ppm. The present method has been em-
ployed with apparent success on the anterior teeth of children and adults. The chair
time needed for a single biopsy is approximately 2 min. Because of the simplicity
of the method it is believed suitable for use in topical F studies and clinical surveys.

Acknowledgements-This investigation was supported by Research Grant DE

2183 from the N.I.D.R., N.I.H., U.S. Public Health Education and Welfare.

R&mn5-Une methode de biopsie d’email a et& mise au point. Elle consiste a prelever
environ 0,2 mg d’email a la surface dune dent a l’aide d’un petit cone rotatif en feutre,
impregn6 de carbure de silicone. Le c8ne est rev&u de glycerine qui absorbe les parti-
cules d’6rnail ainsi obtenues et forme une pate. Cette dernitre ainsi que le cane en feutre
sont places dans un tube en plastique et analyses en ce qui conceme leur contenu en F et
Ca, a l’aide dune electrode de F et la spectophotometrie d’absorption atomique. Les
concentrations de F au niveau des dents anterieures normales du meme individu sont
generalement identiques et les concentrations moyennes varient de 400 SL2500 ppm dans
les dents anterieures des differents sujets 6tudiBs. Une deuxieme biopsie, prelevee sur la
m8me surface dentaire, est moins riche en fluor, confirmant l’existence dun gradient de
F dans l’email. On estime qu’une couche d&mail de 1 a 2~ d’epaisseur, est generalement
prelev&e au tours de la biopsie. Ce pro&de n’est pas desagreable pour le patient, car il
est similaire a un brossage ii la Pierre Ponce et il polit parfaitement les surfaces dentaires.
Cette methode de biopsie est rapide et simple.

Zusammenfassung-Es wurde eine Methode zur Schmelz-Biopsie entwickelt, bei welcher

etwa 0,2 mg Schmelz von zuganglichen Oberflachen einzelner Zahne abpoliert werden.
Dies wird mit Hilfe eines kleinen rotierenden Filzkonus erreicht, der mit Silikon-
Karbid impragniert ist. Der Kegel wird mit Glyzerin bedeckt, das die abgeschlilIenen
Schmelzpartikel auff%ngt und eine Paste bildet. Paste und Filzkonus werden in einen
Plastik-Reagenzbehalter tiberftihrt und unter Verwendung einer F-Elektrode und der
Atomabsorptions-Spektrophotometrie der F- und CAnalyse unterworfen. Die Ergeb-
nisse zeigen, daD die Fluormengen bei intakten Fronttihnen desselben Individuums
gewiihnlich ahnlich waren und da13 die mittleren Konzentrationen bei den Front&en
verschiedener untersuchter Personen zwischen 400 und 2500 ppm schwankten. Eine von
derselben Zahnoberrlache entnommene zweite Biopsie ergab niedrigere Fluorkonzen-
trationen als die erste. Damit werden frtihere Befunde tlher den Fluorgradienten im
Schmelz bei extrahierten Z%hnen best&t&. Es wurde errecbnet. dal3 durch dieses Bion-
sieverfahren gewiihnlich eine Schmelzschicht zwischen 1 und 2 p Dicke entfemt wirh.
Die Patienten erheben keine Einwande, weil die Behandlung einer Zahnreinigung
lhnlich ist und eine hochgradig polierte Oberfliiche verursacht. Das Biopsieverfahren
ist ftt den Patienten harmlos, es ist schnell und einfach durchzuftihren.
 DETERMINATIONOFFLUORIDEINHUMANTEETH 885

REFERENCES
BRUDEVOLD, F., STEADMAN,L. T. and !I=, F. A. 1960. Inorganic and organic components of
tooth structure. Ann. N. Y. Acad. Sci. 85,110-132.
CANDELI, A., CAPOZZI, L., MARCI, F. and MARCHINI G. 1967. The determination of fluoride within
the teeth by means of biopsy on the enamel. Caries Res. 1, 153-161.
MCCANN, H. G. 1968. Determination of fluoride in mineralized tissues using the fluoride ion electrode.
Archs oral Biol. In press.
VRBIC,V., BRDDEVOLD, F. and MCCANN, H. G. 1967. Acquisition of fluoride by enamel from fluoride
pumice pastes. Helv. odont. Acta 11,21-26.