Nature and Role of Loosely Bound Fluoride
In
Dental Caries
1. ARENDS and J. CHRISTOFFERSEN! Dental School, Laboratory for Materia Technica, University of Groningen, Ant. Deusinglaan 1, 9713 AV Groningen, The Netherlands; and 'Medicinsk-Kemisk Institut, Panum Instituttet, Blegdamsvej 3, DK 2200 Copenhagen N, Denmark This paper discusses loosely bound fluoride and its role in den~al caries and prevention. Loosely bound fluoride (abbr. by FJ IS fluon~e adsorbed onto enamel mineral crystallites. Several recent studies indicate that a high total level offluoride in enamel do~s not guaran~ee protection against caries. This leads to the conclusion that a major part offluoride present in the solid enamel is not active in prevention. The adsorption of F to the mineral under acidic conditions is ~e scribed. Most likely there is a dynamic equilibrium between fluoride in solution and adsorbed F at the crystal surface interface. U1zen the crystallite is completely covered by adsorbed F., there is a maximum inhibition of dissolution. The rate of dissolution of mineral depends on pH, the actual concentrations of calcium and phosphate in the liquid in contact with the crystallites, and on the fraction ofthe surface covered by adsorbed fluoride. The fluoride, F" localized in the inner part of the crystallites is relatively unimportant. "Cali-like" materi~l can be fanned on and in enamel depending on conditions. The In vivo-fanned globular "CaF;rlike" material is not pure CaF] and releases F- ions when dissolving; these ions will also be partly adsorbed as F in and on enamel. Presently, the amount and importance of F originating from in vivo-fanned "Ca F;rlike " material are not known. The level of fluoride, F necessary for strong inhibition of. enamel demineralization invitro is estimated to correspond to a fluoride concentration, FL , in the liquid phase of 1 ppm or 50 }JJ7Iol/L fluoride ions. The concentration, Fu needed in vivo may be substantial?, greater, whereas F. has a maximum value, of the order 2 }JJ7Iol/m , corresponding to the concentration of hydroxyl sites in the surface of the crystallites. The present paper indicates that frequent low-level applications offluoride are most likely more beneficial in caries prevention than few treatments with solutions having a high fluoride concentration. Q
Q
- loosely bound fluoride, - the role of loosely bound fluoride in enamel caries,. - loosely bound fluoride and the role of "CaFl-like" material, levels of loosely bound fluoride necessary for the prevention of mineral loss from enamel in vitro and in vivo. The nomenclature used in this paper (see also Fig. 1) is: Fa-fluoride adsorbed to enamel crystallites, FL -fluoride present within enamel in sol~tion, Fs-fluoride inside (solid) enamel crystallites, and Fo-fluoride in the outer solution surrounding the enamel (saliva or plaque fluid). These symbols are also used to indicate the concentrations of these fluoride species. Assuming a Langmuir adsorption, we have: K = Fj(FowFa)F L = xJ(l-x)F L
Q
Q
Q ,
in which K is the adsorption constant, FOIl is the concentration of adsorption sites, mol/rn-, and x is the fr~ction of.sites.occupied by fluoride ions. Solving the adsorption equation gives x =
F)FolI = KFJ(1 + KFd
J Dent Res 69(Spec 155):601-605, February, 1990
For low values of x, we have x = KFL • The degree of surface covering, x, can also be applied during an acid attack ~s long as the rate of fluoride adsorption is fast compared WIth the demineralization rate. The rate constant for fluoride adsorption has been estimated to be of the order 2 x 104 S-1 m- l around pH 6 (Christoffersen et al., 1984).
Introduction.
Fluoride levels in enamel and caries prevention.
Fluoride and its role in caries prevention have been the subject of numerous papers in the past. There is consensusthat fluori.de in the right amount and in the right location prevents and t~ hibits dental caries and, in particular, enamel caries. In this paper, we limit ourselves first to the nature ~nd role of 100s~ly bound F in enamel caries; second, we will deal only with topical cariostatic effects of fluoride. For the ro~e of systemic fluoride in developing enamel, we refer to Fejerskov et al, (1981) and Hardwick et al, (1982). As far as the effect of fluoride on enamel caries is concerned, frequent low-F applications are more effective than ~he application of high-F doses a few times a year. If t.he fluond.e content is increased in a fluoridating agent (e.g., In a fluoridating dentifrice), the efficacy increases b~t t~e effe~t per u~it of fluoride added decreases. It has been Indicated In the literature that not all but part of the fluoride present during a caries attack may be important in caries prevention te.g., Fejerskov et al., 1981; Arends and Christoffersen, 1986). In this paper we will consider the following aspects: - fluoride levels in enamel and caries prevention, Presented at a Joint IADRJORCA International Symposium on Fluorides: Mechanisms of Action and Recommendations for Use, held March 21-24, 1989, Callaway ~rdens Conference Center, Pine Mountain, Georgia
In sound enamel, a fluoride gradient is created during formation and maturation. The fluoride level in sound matured human enamel is on the order of 1000 ppm in the outer 2 urn. However, when one considers the outermost layer of enamel crystallites the fluoride level is on the order of 6000 ppm total F- (0.6 wiw%) or more (Petersson et al., 1976; Lodding and Odelius, 1978). Despite the presence of this high fluoride level, caries can develop in sound enamel as a result of plaque action. It is also well-known that caries lesions contain substantial amounts of fluoride. Fluoride is in fact very easily deposited in porous lesions. The fluoride co~tent of ",,:hite-spot enam~1 and especially of the surface layer is mu~h higher tha~ that In adjacent sound enamel (Dowse and Jenkins, 1957; Little and Steadman, 1966; Hallsworth and Weatherell, 1969). Furthermore, during lesion formation the high fluo~ide level in ~ound enamel is depleted and increases the fluonde con~ent In the body of the lesion (Clarkson et al., 1981). The fluoride content of white-spot lesions is often from 1000 to 3000 ppm F. Despite the presence of this rather high fluoride content, caries lesions (often) progress. Recently, in situ studies using shark enamel consisting of nearly pure fluorapatite (about 3% w/w F) have shown convincingly that caries develops in thes~ teeth undc:r plaque, despite the very large amounts of fluonde present In the enamel (0gaard et al., 1988). Arends and Christoffersen (1983) show~d, in an in vitro demineralization study at pH 4.5, that bovine 601
602
J Dent Res February 1990
ARENDS & CHRISTOFFERSEN
©
saliva p
Fa
CD ® © ©
With "loosely-bound" fluoride, we mean fluoride adsorbed onto the enamel crystallite surface (F.). These definitions are not ideal, but are the best that can be generated with the physico-chemical knowledge available on fluoride at a mineral interface.
The role of loosely bound fluoride F B in enamel caries. Fs
® Fa
crystallite
Fig. 1 - Schematic indication of fluoride in and around enamel crystallites. F. is F adsorbed to crystallites, FL is F present within enamel in solution, F. is F inside solid enamel crystallites, and Fa is F in outersolution or saliva.
enamel with a F content of about 300 ppm F demineralizes fast without F in solution, thus Fa = O. The demineralization could be inhibited, however, at this pH with Fa"" 30 ppm (1.5 mmollL). Nelson et al, (1983a) demonstrated (in hydroxyapatite powders) that 1000 ppm F present in the powder had much less effect on the apatite dissolution than did 1 ppm F (as Fa) in solution. There is, furthermore, no evidence that caries prevalence is lower in sound enamel with a high F content than in sound enamel with a low F level (Poulsen and Larsen, 1975; van der Merwe et al., 1977; Shem et al., 1977; Schamschula et al., 1979). A similareffect has been observedin rat studies. Larson et al. (1976) observed that 10 ppm F added to the drinking water of rats gave better caries protection than did 300-1700 ppm F in the enamel. The experiments mentioned above provide evidence that the presence of fluoride in the solid enamel, even at very high levels, is not directly responsible for the caries-preventive effect of fluoride. The presence of large amounts of fluoride is not synonymous with caries protection or caries prevention.
What is "loosely-bound" fluoride? Several suggestions have been made in the past that only a part of the fluoride present during a caries attack or that only the regularly applied fluoride is responsible for the caries-preventive effect of fluoride (Fejerskov et al., 1981; Arends et al., 1984; Arends and Christoffersen, 1986; 0gaard et al., 1988). Furthermore, it has been indicatedthat fluoride leaching away from "CaF2-like" material deposited on the enamel surface could be responsible for the prophylactic effect of fluoride ions (Chow, 1977; Nelson et al., 1983b, 1984; 0gaard et al., 1987). Although the three suggestions mentioned above are certainly not identical, they have in common the fact that mobile fluoride in the aqueous phase is most beneficial in caries prevention. In 1986, Arends and Christoffersen published a review paper on this subject and introduced the term "fluoride in the liquid phase" in contrast to fluoride in the solid enamel. In this paper we will use the expression "fluoride in the solid (Fs)", if the fluoride is in the solid enamel crystallite and a few atomic layers under the solid surface (see also Fig. 1).
We presume that the adsorbed fluoride (Fa) is mainly responsible for caries prevention in enamel as far as mineral loss is concerned. Fa also plays an important role in remineralization. If, due to a high FL value, F. is at its maximum level, further increase of FL does not lead to more fluoride adsorption, and crystal growth becomes important. During a remineralization process, a high value of FL can be important in order for a high value of Fa to be maintained on the growing crystallites. Furthermore, we assume that the caries process is controlled by the kinetics of the enamel mineral dissolution. As shown by Christoffersen et al. (1984), the study of the adsorption of fluoride (Fa) onto the mineral hydroxyapatite (HAP) under acidic conditions is complex. This is because of the fact that, during the kinetic experiments, precipitation of fluorapatite (FAP) as well as rapid dissolution of HAP had to be avoided. Furthermore, the surface layer of the HAP crystals possesses a non-stoichiometric composition, depending on the concentration in the surrounding solution. Important results from the study quoted are: . (a) The amount of fluoride adsorbed to mineral decreases with increasing pH. The decrease is, however, less than expected for a simple one-to-one exchange of hydroxyl ions for FL in the crystal surface, indicating that the adsorption process of Fa is more complex than a simple ion exchange. (b) If one denotes by JjJ F the ratio of dissolution rates without and with fluoride adsorbed, we may write JjJ F = l/(1-0x) = 1 + KKin FL , in which x is the mole fraction of adsorption sites occupied by fluoride ions; 0 is a factor depending on dissolution affinity, KKin is the kinetically determined Langmuir adsorption constant, and FL is the fluoride concentration. The experimentally determined KKin is larger than the equilibrium adsorption constant; this parameter increases with increasing degree of saturation and increases with decreasing pH. (c) The adsorption results can be described by a model in which a dynamic equilibrium between fluoride in solution and the crystal surface exists. This implies that when fluoride ions (see Fig. 2) are not adsorbed in a given region on the crystal surface, the crystals can dissolve locally. Total inhibition of dissolution takes place only when the solutionbecomessaturatedwith respect to FAP and the surface composition of the crystal is close 10 that of FAP (100% coverage). The Fa values in the above study are about 2 umol/rn? at pH - 6. If the liquid in enamel becomes locally supersaturated with respect to FAP, deposition of FAP is possible at this location. If dissolution of enamel at other locations occurs, the total rate of dissolution will appear as inhibited. The above-mentioned kinetic dissolution experiments on HAP cannot be carried out accurately with enamel powders. Therefore, we have to assume that the F adsorption onto hydroxyapatite-like enamel crystallites .can, at least qualitatively, be described by the same model. The model implies that Fa adsorbed in a thin monomolecular layer surrounding the enamel crystallites inhibits dissolution. It partly transforms the crystallite surface to FAP. If partial coverage takes place (x < 1, with the solution being undersaturated with respect to FAP), the crystallites can still dissolve during a caries attack, depending on the degree of undersaturation.
Vol. 69 Special Issue
603
LOOSELY BOUND FLUORIDE IN DENTAL CARIES
If we consider an enamel crystallite in its natural location during an acid attack, the rate of the acid attack depends on the local pH, on Fa' [Cali> and the total phosphate concentration [Plio The values with subindex i depend on the rate of diffusion of the ions into and out of the lesion. From the above, it is obvious that the amount of fluoride in the solid enamel (Fs) is uninteresting; only the value of Fa is important. The value of Fa can be high for at least two reasons: (i) FL is high due to recent fluoride influx. The influx might be due to direct influx from F~ or from fluoride released by, e.g., "CaF2-like" material (see next section). (ii) Fs is high, and part of the crystallites dissolves. The real crystallite surface decreases, and less F is needed to cover the new smaller surface. Phenomenologically, fluoride released by the crystallite enters the liquid and causes FL (and thereby Fa) to increase.
Loosely bound fluoride and the role of "CaF2-like" material. Using an average Ca content in whole saliva on the order of 2.5 mmol/L and a solubility product of about 10- 10 (McCann, 1968), one can easily calculate that fluoride contents of about 10 ppm F or more should form CaF2 in saliva, in plaque, on teeth, or in carious teeth. Globular structures have indeed been identified as "CaF2-like" material on teeth (Nelson et al., 1983a, b, 1984; Rolla and 0gaard, 1986; Grobler et at., 1981; Dijkman et al., 1983; Christoffersen et at., 1988) and in caries lesions in a few cases (Arends et al., 1988). We call the globules "CaF2-like" instead of CaF2 because most likely the material in vivo is highly contaminated by phosphate, proteins, and possibly other compounds (Lagerlof et al., 1988). Christoffersen et at. (1988) observed that in vitro phosphate incorporation in CaF2 resulted in a higher solubility than pure CaF2 , but with a lower rate of dissolution. Both materials, pure CaF2 and "CaF2-like" material, can be very insoluble depending on solution composition, but this does not affect the solubility products. Presently, the chemical composition and solubility product of "CaF2-like" material formed in vivo are, to the authors' knowledge, not known. The only information available is that the material is "CaF2-like" and the solubility is very low. Dijkman and Arends (1988) showed that if CaF 2-like material is formed in vivo by fluoridated varnish action, the globular structures are still observable after three months, indicating that the fluoride leaches away extremely slowly through the pellicle. It is reasonable to presume that the rate of dissolution of "CaF2-like" globules on enamel surfaces is different from that of globules of this material inside lesions. On the enamel surface, the globules are covered by pellicle (and or plaque) and subjected to regular pH fluctuations; inside the enamel lesion, pH fluctuation will be small because of the very large buffer capacity of the enamel mineral. [It is currently not known whether the globules in and on enamel are identical or have different properties. 1 The intriguing question is: Can "CaF2-like" material situated in a lesion or on a tooth surface release fluoride fast enough during an acid attack to prevent caries progress? In numerous papers, this is implicitly assumed to be the case but without evidence. Fluoride released from "CaF2-like" in vivoformed material can of course be as effective as Fa depending on conditions and amount available. As indicated above, in vivo-formed "CaF2-like " material (often called Fon in the literature) can in principle contribute to Fa; however, the value of FL (and therefore Fa) resulting from Fon is not known.
The experiments of 0gaard et al. (1988) on the in situ demineralization of shark enamel show that pure FAP, dissolved by plaque action, is obviously not capable of preventing caries. With shark enamel there is no doubt that the material dissolved is indeed FAP. The actual concentrations inside the shark enamel lesions are not known, but an estimate is possible. In experiments in which shark enamel (FAP) actually dissolves, the solution must be undersaturated with respect to FAP. Since FAP is much less soluble than HAP at low values of pH, we have, for solutions with Ca/P "" 1.67, low values for the concentrations of calcium, phosphate, and fluoride. Possible binding of fluoride (or calcium) onto proteins in the lesion could reduce these concentrations, causing a dramatic effect on the rate of dissolution. This experiment indicates that if [Ca2 + li and [Pl i and the pH are sufficiently low, even FAP and shark teeth do dissolve. Whatever the reason, this experiment shows that under plaque in vivo, FAP dissolved, and the fluoride released obviously did not prevent progress of caries. Another in situ model study has been published by Reintsema and Arends (1988) and Arends et at. (1988), in which on the one hand it is known that quite large amounts of "CaF2 like" material exist in vivo in lesions, but on the other hand lesion progress was not completely inhibited. "CaF2-like" material could dissolve during a caries attack and provide fluoride in the form of Fa. In the experiments of Reintsema et at. (1985), this effect cannot be completely ignored but was obviously small. "CaF2-like" material inside a lesion may also be considered harmful, since the Ca2 + most likely originates from the enamel structure. To the authors' knowledge there is no published study indicating that after a caries attack the amount of "CaF2-like" material in lesions is smaller than before the attack. Summarizing the role of "CaF2-like" material and Fa: - "CaF2-like" material releases fluoride in vivo; an unknown amount could act as Fa.
®
total coverage
@
total coverage after acid attack
partial coverage
@
partial coverage after acid attack
Fig. 2 - Schematic representation of the effect of Fa. A, total coverage, C, total coverage after acid attack, B, partial coverage, and D, partial coverage after acid attack.
J Dent Res February 1990
ARENDS & CHRISTOFFERSEN
604
TABLE FLUORIDE LEVELS IN SOLUTION (AS Fa) NECESSARY TO PREVENT ENAMEL DEMINERALIZATION F level in umol.Lr ' 1500 100 50 25
Conditions pH = 4.5, sink conditions, no Ca or phosphate pH = 5, partially saturated acetate buffer pH = 4.3, partially saturated lactate buffer pH = 5, partially saturated acetate buffer
Reference Arends and Christoffersen (1983) Ten Cate and Duijsters (1983a) Margolis et al. (1986) Borsboom et al. (1985)
"CaF2-like" material in and on enamel might behave differently.
Levels of adsorbed fluoride (Fa) necessary in vitro and in vivo for optimum protection. Several investigations indicate that low levels of fluoride in solution reduce mineral loss from enamel. The first paper was from Manly and Harrington (1959), who observed that 0.1 ppm fluoride in solution reduced enamel dissolution substantially. Arends and Christoffersen (1983), ten Cate and Duijsters (1983a, b), Borsboom et al, (1985), Margolis et al, (1986), and Arends and Christoffersen (1986) all indicated that completeor nearly completeinhibition of enamel dissolution could be obtained by fluoride (as FL ) in solution. The levels of fluoride in solution (see Table) necessary for inhibition vary from 5 to 1500 urnol F- .L -1 or "" 0.1 - 30 ppm, depending on the conditions (pH, degreeof saturation, F- level in solution, etc.). Therefore, we can conclude that fluoride, when present in solution in concentrations around 1 ppm (50 urnol.L:"), effectively prevents mineral loss from enamel. The exact level of protection depends strongly on the conditions of pH, etc. Some care is necessary for these data to be extrapolated to the in vivo situation during a cariogenic challenge by plaque. Although looselybound fluoride (Fa) is most likely also acting in vivo, the level neededmay be quite different. One can argue that in the in vivo situation a higher total fluoride content is required to obtain an adequate value of FL and thus of Fa' because fluoride may adsorb, e.g., to the organic matrix of enamel, to the pellicle, to plaque, etc. Furthermore, the time for fluoride penetration into an in vivo lesion can be very long, up to several months (Arends and Gelhard, 1983). The level of FL and thus also of Fa actuallyrequired for caries prevention in vivo will also greatly depend on the formation of insoluble "CaF2-like" material in and on lesions and on the efficiency of the remineralization mechanism. The present information is such that the level Fa of fluoride needed for optimum in vivo protection of enamel against an acid challenge is not really known. An indication of an in vivo level is provided by the study of 0gaard et al, (1987), who showed that a daily rinsing with a 0.2% w/w solution of NaF equivalent to an Fa value of about 1000 ppm F provided total caries protection duringan intensecariogenic challenge.Lower Fa levels might be effective in vivo as well. Summarizing, the literature indicates that: - Fluoride adsorbed onto the enamelmineral as Fa resulting from fluoride in solution is most likely responsible for the caries-preventive effect of fluoride. - The fluoride present inside the solid enamel (Fs) is most likely of lesser importance than fluoride in solution.
-
-
-
-
Fluoride presentas Fs is in fact latent (not effective) until exposed due to crystallite dissolution. It is not certain whether fluoride deposited on or in enamel as "CaF2-like" material provides fluoride that causes an increase in FL and thus in Fa. Fa corresponding to FL values in solution prevents enamel demineralization, if presentin concentrations in the order of 1 ppm (50 umol.Lr '). In vivo levels are presumed to be higher but are presently unknown. Frequent low-level applications of fluoride are more effective than high-dose applications a few times a year, becauseFL and thus Fa are maintained highwith frequent applications. If Fa is at its maximum level, a further increase in FL does not lead to more adsorption; crystal growth processes may require high values of FL in order to form FAP and HAP. REFERENCES
ARENDS, J. and CHRISTOFFERSEN, J. (1983): The Influence of Fluoride Concentration on the Progress of Demineralization in Bovine Enamel at pH = 4.5, Caries Res 17:455-457. ARENDS, J. and CHRISTOFFERSEN, J. (1986): The Nature of Early Caries Lesions in Enamel, J Dent Res 65:2-11. ARENDS, J. and GELHARD, T.B.F.M. (1983): In vivo Remineralization of Human Enamel. In: Demineralisation and Remlneralisation of Teeth, S.A. Leach and W.M. Edgar, Eds., Oxford: IRL Press, pp. 1-16. ARENDS, J.; NELSON, D.G.A.; DIJKMAN, A.G.; and JONGEBLOED, W.L. (1984): Effect of Various Fluorides on Enamel Structure and Chemistry. In: Cariology Today, B. Guggenheim, Ed., Basel: Karger, pp. 231-236. ARENDS, J.; REINTSEMA, H.; and DIJKMAN, T. (1988): Calcium Fluoride-Like Material Formed in Partially Demineralized Human Enamel in vivo, Acta Odontol Scand 46:347-353. BORSBOOM, P.C.F.; VAN DER MEl, H.C.; and ARENDS, J. (1985): Enamel Lesion Formation with and without 0.12 ppm F in Solution, Caries Res 19:396-403. CHOW, L.C. (1977): Chemistry of Topical Fluorides, Caries Res I(SuppI): 191-197. CHRISTOFFERSEN, M.R.; CHRISTOFFERSEN, 1.; and ARENDS, J. (1984): Kinetics of Dissolution of Calcium Hydroxyapatite. VII. The Effect of Fluoride Ions, J Crystal Growth 67:107-
114. CHRISTOFFERSEN, 1.; CHRISTOFFERSEN, M.R.; KIBALCZYC, W.; and PERDOK, W.G. (1988): Kinetics of Dissolution and Growth of Calcium Fluoride and Effects of Phosphate in Solution and Incorporated in the Crystals on these Processes, Acta Odontol Scand 36:325-336. CLARKSON, B.H.; WEFEL, J.S.; and SILVERSTONE, L.M. (1981): Redistribution of Enamel Fluoride during White Spot Lesion Formation, Caries Res 15:158-165. DOWSE, C.M. and JENKINS, G.N. (1957): Fluoride Uptake in vivo in Enamel Defects and its Significance, J Dent Res 36:816. DIJKMAN, A.G. and ARENDS, J. (1988): The Role of CaF 2-like Material in Topical Fluoridation of Enamel in situ, Acta Odontol Scand 46:391-397. DIJKMAN, A.G.; DE BOER, P.; and ARENDS, J. (1983): In vivo Investigations on the F-content in and on Human Enamel After Topical Applications, Caries Res 17:392-402. FEJERSKOV, 0.; THYLSTRUP, A.; and LARSEN, M.J. (1981): Rational Use of Fluorides in Caries Prevention. A Concept Based on the Possible Cariostatic Mechanisms, Acta Odontol Scand 39:241-249. GROBLER, S.R.; 0GAARD, B.; and ROLLA, G. (1981): Uptake and Retention of Fluoride in Sound Dental Enamel in vivo After a Single Application of Neutral 2% Sodium Fluoride. In: Tooth Surface Interactions and Preventive Dentistry, G. Rolla, T. Senju, and G. Embery, Eds., London: Information Retrieval, Inc., pp. 17-25.
Vol. 69 Special Issue
LOOSELY BOUND FLUORIDE IN DENTAL CARIES
HALLSWORTH, A.S. and WEATHERELL, J.A. (1969): The Microdistribution, Uptake, and Loss of Fluoride in Human Enamel, Caries Res 3:109-118. HARDWICK, J.L.; TEASDALE, J.; and BLOODWORTH, G. (1982): Caries Increments over 4 Years in Children aged 12 at the Start of Water Fluoridation, Br Dent J 153:217-222. LAGERLOF, F.; SAXEGAARD, E.; BARKVOLL, P.; and ROLLA, G. (1988): Effects of Inorganic Orthophosphate and Pyrophosphate on the Dissolution of Calcium Fluoride in Water, J Dent Res 67:447-449. LARSON, R.H.; MELLBERG,~·1.R.; ENGLANDER, H.R.; and SENNING, R. (1976): Caries Inhibition in the Rat by Water-borne and Enamel-bound Fluoride, Caries Res 10:321-33l. LITTLE, M.F. and STEADMAN, L.T. (1966): Chemical and Physical Properties of Altered and Sound Enamel IV, Arch Oral Bioi 11:273-278. LODDING, A. and ODELIUS, H. (1978): In Depth Distributions of Elements in Teeth Determined by Ion Probe Analysis, Microscopica Acta (Suppl 2): 367-376. MANLY, R.S. and HARRINGTON, D.P. (1959): Solution Rate of Tooth Enamel in an Acetate Buffer, J Dent Res 38: 910-919. MARGOLIS, H.C. and MORENO, E.C. (1985): Kinetic and ThermodynamicAspects of Enamel Demineralization, Caries Res 19:2235. MARGOLIS, H.C.; MORENO, E.C.; and MURPHY, B.J. (1986): Effect of Low Levels of Fluoride in Solution on Enamel Demineralization in vitro, J Dent Res 65:23-29. McCANN, H.G. (1968): The Solubility of Fluorapatite and its Relationship to that of CaF2, Arch Oral Bioi 13:987-100l. NELSON, D.G.A.; FEATHERSTONE, J.D.B.; DUNCAN, J.F.; and CUTRESS, T.W. (1983a): Effect of Carbonate and Fluoride on the Dissolution Behavior of Synthetic Apatites, Caries Res 17:20021l. NELSON, D.G.A.; JONGEBLOED, W.L.; and ARENDS, J. (1983b): Morphology of Enamel Surfaces Treated with Topical Fluoride Agents: SEM Considerations, J Dent Res 62:1201-1208. NELSON, D.G.A.; JONGEBLOED, W.L.; and ARENDS, J. (1984): Crystallographic Structure of Enamel Surfaces Treated with Topical Fluoride Agents: TEM and XRD Considerations, J Dent Res 63:6-12. 0GAARD, B.; ROLLA, G.; and ARENDS, J. (1988): In vivo Progress of Enamel and Root Surface Lesions under Plaque as a Function of Time, Caries Res 22:302-305.
605
0GAARD, B.; ROLLA, G.; and HELGELAND, K. (1983): Alkali Soluble and Alkali Insoluble Fluoride Retention in Demineralized Enamel in vivo, Scand J Dent Res 9:200-204. 0GAARD, B.; ROLLA, G.; RUBEN, 1.; and ARENDS, 1. (1988): Microradiographic Study of Demineralization of Shark Enamel in a Human Caries Model, Scand J Dent Res 96:209-21l. 0GAARD, B.; ROLLA, G.; and TEN CATE, 1.M. (1987): Calcium Fluoride Protects Enamel against Caries in vivo, Caries Res 21:159. PETERSSON, L.G.; ODELIUS, H.; LODDING, A.; LARSEN, S.J.; and FROSTELL, G. (1976): Ion Probe Study of F Gradients in Outermost Layers of Human Enamel, J Dent Res 55:980-990. POULSEN, S. and LARSEN, M.J. (1975): Dental Caries in Relation to Fluoride Content of Enamel in the Primary Dentition, Caries Res 9:59-65. REINTSEMA, H. and ARENDS, 1. (1988): An in vivo Study of Microhardness and Fluoride Uptake in Partially Demineralized Human Enamel Covered by Plaque, J Dent Res 67:471-473. REINTSEMA, H.; SCHUTHOF, 1.; and ARENDS, J. (1985): An in vivo Investigation of the Fluoride Uptake in Partially Demineralized Human Enamel from Several Different Dentifrices, J Dent Res 64:19-23. ROLLA, G. and 0GAARD, B. (1986): Studies on the Solubility of Calcium Fluoride in Human Saliva. In: Factors Relating to Demineralisation and Remineralisation of the Teeth, S.A. Leach, Ed., Oxford: IRL Press Limited, pp. 45-50. SCHAMSCHULA, R.G.; AGUS, H.; CHARLTON, G.; DUPPENTHALER, J.L.; and UN, P. (1979): Associations between Fluoride Concentration in Successive Layers of Human Enamel and Individual Dental Caries Experience, Arch Oral Bioi 24:847-859. SHERN, R.J.; DRISCOLL, W.S.; and KORTS, D.C. (1977): Enamel Biopsy Results of Children Receiving Fluoride Tablets, JAm Dent Assoc 95:310-314. TEN CATE, J.M. and DUlJSTERS, P.P.E. (1983a): Influence of Fluoride in Solution on Tooth Demineralization.!. Chemical Data, Caries Res 17:193-199. TEN CATE, J.M. and DUlJSTERS, P.P.E. (1983b): Influence of Fluoride in Solution on Tooth Demineralization. II. Microradiographic Data, Caries Res 17:513-519. VAN DER MERWE, E.H.M.; BISCHOFF, J.1.; FATTI, L.P.; RETIEF, D.H.; BARBAKOW, F.H.; and FRIEDMAN, M. (1977): Relationships Between Fluoride in Enamel, DMFT Index and Fluorosis in High- and Low-Fluoride Areas in South Africa, Community Dent Oral Epidemiol 5:61-64.
In
Dental Caries
1. ARENDS and J. CHRISTOFFERSEN! Dental School, Laboratory for Materia Technica, University of Groningen, Ant. Deusinglaan 1, 9713 AV Groningen, The Netherlands; and 'Medicinsk-Kemisk Institut, Panum Instituttet, Blegdamsvej 3, DK 2200 Copenhagen N, Denmark This paper discusses loosely bound fluoride and its role in den~al caries and prevention. Loosely bound fluoride (abbr. by FJ IS fluon~e adsorbed onto enamel mineral crystallites. Several recent studies indicate that a high total level offluoride in enamel do~s not guaran~ee protection against caries. This leads to the conclusion that a major part offluoride present in the solid enamel is not active in prevention. The adsorption of F to the mineral under acidic conditions is ~e scribed. Most likely there is a dynamic equilibrium between fluoride in solution and adsorbed F at the crystal surface interface. U1zen the crystallite is completely covered by adsorbed F., there is a maximum inhibition of dissolution. The rate of dissolution of mineral depends on pH, the actual concentrations of calcium and phosphate in the liquid in contact with the crystallites, and on the fraction ofthe surface covered by adsorbed fluoride. The fluoride, F" localized in the inner part of the crystallites is relatively unimportant. "Cali-like" materi~l can be fanned on and in enamel depending on conditions. The In vivo-fanned globular "CaF;rlike" material is not pure CaF] and releases F- ions when dissolving; these ions will also be partly adsorbed as F in and on enamel. Presently, the amount and importance of F originating from in vivo-fanned "Ca F;rlike " material are not known. The level of fluoride, F necessary for strong inhibition of. enamel demineralization invitro is estimated to correspond to a fluoride concentration, FL , in the liquid phase of 1 ppm or 50 }JJ7Iol/L fluoride ions. The concentration, Fu needed in vivo may be substantial?, greater, whereas F. has a maximum value, of the order 2 }JJ7Iol/m , corresponding to the concentration of hydroxyl sites in the surface of the crystallites. The present paper indicates that frequent low-level applications offluoride are most likely more beneficial in caries prevention than few treatments with solutions having a high fluoride concentration. Q
Q
- loosely bound fluoride, - the role of loosely bound fluoride in enamel caries,. - loosely bound fluoride and the role of "CaFl-like" material, levels of loosely bound fluoride necessary for the prevention of mineral loss from enamel in vitro and in vivo. The nomenclature used in this paper (see also Fig. 1) is: Fa-fluoride adsorbed to enamel crystallites, FL -fluoride present within enamel in sol~tion, Fs-fluoride inside (solid) enamel crystallites, and Fo-fluoride in the outer solution surrounding the enamel (saliva or plaque fluid). These symbols are also used to indicate the concentrations of these fluoride species. Assuming a Langmuir adsorption, we have: K = Fj(FowFa)F L = xJ(l-x)F L
Q
Q
Q ,
in which K is the adsorption constant, FOIl is the concentration of adsorption sites, mol/rn-, and x is the fr~ction of.sites.occupied by fluoride ions. Solving the adsorption equation gives x =
F)FolI = KFJ(1 + KFd
J Dent Res 69(Spec 155):601-605, February, 1990
For low values of x, we have x = KFL • The degree of surface covering, x, can also be applied during an acid attack ~s long as the rate of fluoride adsorption is fast compared WIth the demineralization rate. The rate constant for fluoride adsorption has been estimated to be of the order 2 x 104 S-1 m- l around pH 6 (Christoffersen et al., 1984).
Introduction.
Fluoride levels in enamel and caries prevention.
Fluoride and its role in caries prevention have been the subject of numerous papers in the past. There is consensusthat fluori.de in the right amount and in the right location prevents and t~ hibits dental caries and, in particular, enamel caries. In this paper, we limit ourselves first to the nature ~nd role of 100s~ly bound F in enamel caries; second, we will deal only with topical cariostatic effects of fluoride. For the ro~e of systemic fluoride in developing enamel, we refer to Fejerskov et al, (1981) and Hardwick et al, (1982). As far as the effect of fluoride on enamel caries is concerned, frequent low-F applications are more effective than ~he application of high-F doses a few times a year. If t.he fluond.e content is increased in a fluoridating agent (e.g., In a fluoridating dentifrice), the efficacy increases b~t t~e effe~t per u~it of fluoride added decreases. It has been Indicated In the literature that not all but part of the fluoride present during a caries attack may be important in caries prevention te.g., Fejerskov et al., 1981; Arends and Christoffersen, 1986). In this paper we will consider the following aspects: - fluoride levels in enamel and caries prevention, Presented at a Joint IADRJORCA International Symposium on Fluorides: Mechanisms of Action and Recommendations for Use, held March 21-24, 1989, Callaway ~rdens Conference Center, Pine Mountain, Georgia
In sound enamel, a fluoride gradient is created during formation and maturation. The fluoride level in sound matured human enamel is on the order of 1000 ppm in the outer 2 urn. However, when one considers the outermost layer of enamel crystallites the fluoride level is on the order of 6000 ppm total F- (0.6 wiw%) or more (Petersson et al., 1976; Lodding and Odelius, 1978). Despite the presence of this high fluoride level, caries can develop in sound enamel as a result of plaque action. It is also well-known that caries lesions contain substantial amounts of fluoride. Fluoride is in fact very easily deposited in porous lesions. The fluoride co~tent of ",,:hite-spot enam~1 and especially of the surface layer is mu~h higher tha~ that In adjacent sound enamel (Dowse and Jenkins, 1957; Little and Steadman, 1966; Hallsworth and Weatherell, 1969). Furthermore, during lesion formation the high fluo~ide level in ~ound enamel is depleted and increases the fluonde con~ent In the body of the lesion (Clarkson et al., 1981). The fluoride content of white-spot lesions is often from 1000 to 3000 ppm F. Despite the presence of this rather high fluoride content, caries lesions (often) progress. Recently, in situ studies using shark enamel consisting of nearly pure fluorapatite (about 3% w/w F) have shown convincingly that caries develops in thes~ teeth undc:r plaque, despite the very large amounts of fluonde present In the enamel (0gaard et al., 1988). Arends and Christoffersen (1983) show~d, in an in vitro demineralization study at pH 4.5, that bovine 601
602
J Dent Res February 1990
ARENDS & CHRISTOFFERSEN
©
saliva p
Fa
CD ® © ©
With "loosely-bound" fluoride, we mean fluoride adsorbed onto the enamel crystallite surface (F.). These definitions are not ideal, but are the best that can be generated with the physico-chemical knowledge available on fluoride at a mineral interface.
The role of loosely bound fluoride F B in enamel caries. Fs
® Fa
crystallite
Fig. 1 - Schematic indication of fluoride in and around enamel crystallites. F. is F adsorbed to crystallites, FL is F present within enamel in solution, F. is F inside solid enamel crystallites, and Fa is F in outersolution or saliva.
enamel with a F content of about 300 ppm F demineralizes fast without F in solution, thus Fa = O. The demineralization could be inhibited, however, at this pH with Fa"" 30 ppm (1.5 mmollL). Nelson et al, (1983a) demonstrated (in hydroxyapatite powders) that 1000 ppm F present in the powder had much less effect on the apatite dissolution than did 1 ppm F (as Fa) in solution. There is, furthermore, no evidence that caries prevalence is lower in sound enamel with a high F content than in sound enamel with a low F level (Poulsen and Larsen, 1975; van der Merwe et al., 1977; Shem et al., 1977; Schamschula et al., 1979). A similareffect has been observedin rat studies. Larson et al. (1976) observed that 10 ppm F added to the drinking water of rats gave better caries protection than did 300-1700 ppm F in the enamel. The experiments mentioned above provide evidence that the presence of fluoride in the solid enamel, even at very high levels, is not directly responsible for the caries-preventive effect of fluoride. The presence of large amounts of fluoride is not synonymous with caries protection or caries prevention.
What is "loosely-bound" fluoride? Several suggestions have been made in the past that only a part of the fluoride present during a caries attack or that only the regularly applied fluoride is responsible for the caries-preventive effect of fluoride (Fejerskov et al., 1981; Arends et al., 1984; Arends and Christoffersen, 1986; 0gaard et al., 1988). Furthermore, it has been indicatedthat fluoride leaching away from "CaF2-like" material deposited on the enamel surface could be responsible for the prophylactic effect of fluoride ions (Chow, 1977; Nelson et al., 1983b, 1984; 0gaard et al., 1987). Although the three suggestions mentioned above are certainly not identical, they have in common the fact that mobile fluoride in the aqueous phase is most beneficial in caries prevention. In 1986, Arends and Christoffersen published a review paper on this subject and introduced the term "fluoride in the liquid phase" in contrast to fluoride in the solid enamel. In this paper we will use the expression "fluoride in the solid (Fs)", if the fluoride is in the solid enamel crystallite and a few atomic layers under the solid surface (see also Fig. 1).
We presume that the adsorbed fluoride (Fa) is mainly responsible for caries prevention in enamel as far as mineral loss is concerned. Fa also plays an important role in remineralization. If, due to a high FL value, F. is at its maximum level, further increase of FL does not lead to more fluoride adsorption, and crystal growth becomes important. During a remineralization process, a high value of FL can be important in order for a high value of Fa to be maintained on the growing crystallites. Furthermore, we assume that the caries process is controlled by the kinetics of the enamel mineral dissolution. As shown by Christoffersen et al. (1984), the study of the adsorption of fluoride (Fa) onto the mineral hydroxyapatite (HAP) under acidic conditions is complex. This is because of the fact that, during the kinetic experiments, precipitation of fluorapatite (FAP) as well as rapid dissolution of HAP had to be avoided. Furthermore, the surface layer of the HAP crystals possesses a non-stoichiometric composition, depending on the concentration in the surrounding solution. Important results from the study quoted are: . (a) The amount of fluoride adsorbed to mineral decreases with increasing pH. The decrease is, however, less than expected for a simple one-to-one exchange of hydroxyl ions for FL in the crystal surface, indicating that the adsorption process of Fa is more complex than a simple ion exchange. (b) If one denotes by JjJ F the ratio of dissolution rates without and with fluoride adsorbed, we may write JjJ F = l/(1-0x) = 1 + KKin FL , in which x is the mole fraction of adsorption sites occupied by fluoride ions; 0 is a factor depending on dissolution affinity, KKin is the kinetically determined Langmuir adsorption constant, and FL is the fluoride concentration. The experimentally determined KKin is larger than the equilibrium adsorption constant; this parameter increases with increasing degree of saturation and increases with decreasing pH. (c) The adsorption results can be described by a model in which a dynamic equilibrium between fluoride in solution and the crystal surface exists. This implies that when fluoride ions (see Fig. 2) are not adsorbed in a given region on the crystal surface, the crystals can dissolve locally. Total inhibition of dissolution takes place only when the solutionbecomessaturatedwith respect to FAP and the surface composition of the crystal is close 10 that of FAP (100% coverage). The Fa values in the above study are about 2 umol/rn? at pH - 6. If the liquid in enamel becomes locally supersaturated with respect to FAP, deposition of FAP is possible at this location. If dissolution of enamel at other locations occurs, the total rate of dissolution will appear as inhibited. The above-mentioned kinetic dissolution experiments on HAP cannot be carried out accurately with enamel powders. Therefore, we have to assume that the F adsorption onto hydroxyapatite-like enamel crystallites .can, at least qualitatively, be described by the same model. The model implies that Fa adsorbed in a thin monomolecular layer surrounding the enamel crystallites inhibits dissolution. It partly transforms the crystallite surface to FAP. If partial coverage takes place (x < 1, with the solution being undersaturated with respect to FAP), the crystallites can still dissolve during a caries attack, depending on the degree of undersaturation.
Vol. 69 Special Issue
603
LOOSELY BOUND FLUORIDE IN DENTAL CARIES
If we consider an enamel crystallite in its natural location during an acid attack, the rate of the acid attack depends on the local pH, on Fa' [Cali> and the total phosphate concentration [Plio The values with subindex i depend on the rate of diffusion of the ions into and out of the lesion. From the above, it is obvious that the amount of fluoride in the solid enamel (Fs) is uninteresting; only the value of Fa is important. The value of Fa can be high for at least two reasons: (i) FL is high due to recent fluoride influx. The influx might be due to direct influx from F~ or from fluoride released by, e.g., "CaF2-like" material (see next section). (ii) Fs is high, and part of the crystallites dissolves. The real crystallite surface decreases, and less F is needed to cover the new smaller surface. Phenomenologically, fluoride released by the crystallite enters the liquid and causes FL (and thereby Fa) to increase.
Loosely bound fluoride and the role of "CaF2-like" material. Using an average Ca content in whole saliva on the order of 2.5 mmol/L and a solubility product of about 10- 10 (McCann, 1968), one can easily calculate that fluoride contents of about 10 ppm F or more should form CaF2 in saliva, in plaque, on teeth, or in carious teeth. Globular structures have indeed been identified as "CaF2-like" material on teeth (Nelson et al., 1983a, b, 1984; Rolla and 0gaard, 1986; Grobler et at., 1981; Dijkman et al., 1983; Christoffersen et at., 1988) and in caries lesions in a few cases (Arends et al., 1988). We call the globules "CaF2-like" instead of CaF2 because most likely the material in vivo is highly contaminated by phosphate, proteins, and possibly other compounds (Lagerlof et al., 1988). Christoffersen et at. (1988) observed that in vitro phosphate incorporation in CaF2 resulted in a higher solubility than pure CaF2 , but with a lower rate of dissolution. Both materials, pure CaF2 and "CaF2-like" material, can be very insoluble depending on solution composition, but this does not affect the solubility products. Presently, the chemical composition and solubility product of "CaF2-like" material formed in vivo are, to the authors' knowledge, not known. The only information available is that the material is "CaF2-like" and the solubility is very low. Dijkman and Arends (1988) showed that if CaF 2-like material is formed in vivo by fluoridated varnish action, the globular structures are still observable after three months, indicating that the fluoride leaches away extremely slowly through the pellicle. It is reasonable to presume that the rate of dissolution of "CaF2-like" globules on enamel surfaces is different from that of globules of this material inside lesions. On the enamel surface, the globules are covered by pellicle (and or plaque) and subjected to regular pH fluctuations; inside the enamel lesion, pH fluctuation will be small because of the very large buffer capacity of the enamel mineral. [It is currently not known whether the globules in and on enamel are identical or have different properties. 1 The intriguing question is: Can "CaF2-like" material situated in a lesion or on a tooth surface release fluoride fast enough during an acid attack to prevent caries progress? In numerous papers, this is implicitly assumed to be the case but without evidence. Fluoride released from "CaF2-like" in vivoformed material can of course be as effective as Fa depending on conditions and amount available. As indicated above, in vivo-formed "CaF2-like " material (often called Fon in the literature) can in principle contribute to Fa; however, the value of FL (and therefore Fa) resulting from Fon is not known.
The experiments of 0gaard et al. (1988) on the in situ demineralization of shark enamel show that pure FAP, dissolved by plaque action, is obviously not capable of preventing caries. With shark enamel there is no doubt that the material dissolved is indeed FAP. The actual concentrations inside the shark enamel lesions are not known, but an estimate is possible. In experiments in which shark enamel (FAP) actually dissolves, the solution must be undersaturated with respect to FAP. Since FAP is much less soluble than HAP at low values of pH, we have, for solutions with Ca/P "" 1.67, low values for the concentrations of calcium, phosphate, and fluoride. Possible binding of fluoride (or calcium) onto proteins in the lesion could reduce these concentrations, causing a dramatic effect on the rate of dissolution. This experiment indicates that if [Ca2 + li and [Pl i and the pH are sufficiently low, even FAP and shark teeth do dissolve. Whatever the reason, this experiment shows that under plaque in vivo, FAP dissolved, and the fluoride released obviously did not prevent progress of caries. Another in situ model study has been published by Reintsema and Arends (1988) and Arends et at. (1988), in which on the one hand it is known that quite large amounts of "CaF2 like" material exist in vivo in lesions, but on the other hand lesion progress was not completely inhibited. "CaF2-like" material could dissolve during a caries attack and provide fluoride in the form of Fa. In the experiments of Reintsema et at. (1985), this effect cannot be completely ignored but was obviously small. "CaF2-like" material inside a lesion may also be considered harmful, since the Ca2 + most likely originates from the enamel structure. To the authors' knowledge there is no published study indicating that after a caries attack the amount of "CaF2-like" material in lesions is smaller than before the attack. Summarizing the role of "CaF2-like" material and Fa: - "CaF2-like" material releases fluoride in vivo; an unknown amount could act as Fa.
®
total coverage
@
total coverage after acid attack
partial coverage
@
partial coverage after acid attack
Fig. 2 - Schematic representation of the effect of Fa. A, total coverage, C, total coverage after acid attack, B, partial coverage, and D, partial coverage after acid attack.
J Dent Res February 1990
ARENDS & CHRISTOFFERSEN
604
TABLE FLUORIDE LEVELS IN SOLUTION (AS Fa) NECESSARY TO PREVENT ENAMEL DEMINERALIZATION F level in umol.Lr ' 1500 100 50 25
Conditions pH = 4.5, sink conditions, no Ca or phosphate pH = 5, partially saturated acetate buffer pH = 4.3, partially saturated lactate buffer pH = 5, partially saturated acetate buffer
Reference Arends and Christoffersen (1983) Ten Cate and Duijsters (1983a) Margolis et al. (1986) Borsboom et al. (1985)
"CaF2-like" material in and on enamel might behave differently.
Levels of adsorbed fluoride (Fa) necessary in vitro and in vivo for optimum protection. Several investigations indicate that low levels of fluoride in solution reduce mineral loss from enamel. The first paper was from Manly and Harrington (1959), who observed that 0.1 ppm fluoride in solution reduced enamel dissolution substantially. Arends and Christoffersen (1983), ten Cate and Duijsters (1983a, b), Borsboom et al, (1985), Margolis et al, (1986), and Arends and Christoffersen (1986) all indicated that completeor nearly completeinhibition of enamel dissolution could be obtained by fluoride (as FL ) in solution. The levels of fluoride in solution (see Table) necessary for inhibition vary from 5 to 1500 urnol F- .L -1 or "" 0.1 - 30 ppm, depending on the conditions (pH, degreeof saturation, F- level in solution, etc.). Therefore, we can conclude that fluoride, when present in solution in concentrations around 1 ppm (50 urnol.L:"), effectively prevents mineral loss from enamel. The exact level of protection depends strongly on the conditions of pH, etc. Some care is necessary for these data to be extrapolated to the in vivo situation during a cariogenic challenge by plaque. Although looselybound fluoride (Fa) is most likely also acting in vivo, the level neededmay be quite different. One can argue that in the in vivo situation a higher total fluoride content is required to obtain an adequate value of FL and thus of Fa' because fluoride may adsorb, e.g., to the organic matrix of enamel, to the pellicle, to plaque, etc. Furthermore, the time for fluoride penetration into an in vivo lesion can be very long, up to several months (Arends and Gelhard, 1983). The level of FL and thus also of Fa actuallyrequired for caries prevention in vivo will also greatly depend on the formation of insoluble "CaF2-like" material in and on lesions and on the efficiency of the remineralization mechanism. The present information is such that the level Fa of fluoride needed for optimum in vivo protection of enamel against an acid challenge is not really known. An indication of an in vivo level is provided by the study of 0gaard et al, (1987), who showed that a daily rinsing with a 0.2% w/w solution of NaF equivalent to an Fa value of about 1000 ppm F provided total caries protection duringan intensecariogenic challenge.Lower Fa levels might be effective in vivo as well. Summarizing, the literature indicates that: - Fluoride adsorbed onto the enamelmineral as Fa resulting from fluoride in solution is most likely responsible for the caries-preventive effect of fluoride. - The fluoride present inside the solid enamel (Fs) is most likely of lesser importance than fluoride in solution.
-
-
-
-
Fluoride presentas Fs is in fact latent (not effective) until exposed due to crystallite dissolution. It is not certain whether fluoride deposited on or in enamel as "CaF2-like" material provides fluoride that causes an increase in FL and thus in Fa. Fa corresponding to FL values in solution prevents enamel demineralization, if presentin concentrations in the order of 1 ppm (50 umol.Lr '). In vivo levels are presumed to be higher but are presently unknown. Frequent low-level applications of fluoride are more effective than high-dose applications a few times a year, becauseFL and thus Fa are maintained highwith frequent applications. If Fa is at its maximum level, a further increase in FL does not lead to more adsorption; crystal growth processes may require high values of FL in order to form FAP and HAP. REFERENCES
ARENDS, J. and CHRISTOFFERSEN, J. (1983): The Influence of Fluoride Concentration on the Progress of Demineralization in Bovine Enamel at pH = 4.5, Caries Res 17:455-457. ARENDS, J. and CHRISTOFFERSEN, J. (1986): The Nature of Early Caries Lesions in Enamel, J Dent Res 65:2-11. ARENDS, J. and GELHARD, T.B.F.M. (1983): In vivo Remineralization of Human Enamel. In: Demineralisation and Remlneralisation of Teeth, S.A. Leach and W.M. Edgar, Eds., Oxford: IRL Press, pp. 1-16. ARENDS, J.; NELSON, D.G.A.; DIJKMAN, A.G.; and JONGEBLOED, W.L. (1984): Effect of Various Fluorides on Enamel Structure and Chemistry. In: Cariology Today, B. Guggenheim, Ed., Basel: Karger, pp. 231-236. ARENDS, J.; REINTSEMA, H.; and DIJKMAN, T. (1988): Calcium Fluoride-Like Material Formed in Partially Demineralized Human Enamel in vivo, Acta Odontol Scand 46:347-353. BORSBOOM, P.C.F.; VAN DER MEl, H.C.; and ARENDS, J. (1985): Enamel Lesion Formation with and without 0.12 ppm F in Solution, Caries Res 19:396-403. CHOW, L.C. (1977): Chemistry of Topical Fluorides, Caries Res I(SuppI): 191-197. CHRISTOFFERSEN, M.R.; CHRISTOFFERSEN, 1.; and ARENDS, J. (1984): Kinetics of Dissolution of Calcium Hydroxyapatite. VII. The Effect of Fluoride Ions, J Crystal Growth 67:107-
114. CHRISTOFFERSEN, 1.; CHRISTOFFERSEN, M.R.; KIBALCZYC, W.; and PERDOK, W.G. (1988): Kinetics of Dissolution and Growth of Calcium Fluoride and Effects of Phosphate in Solution and Incorporated in the Crystals on these Processes, Acta Odontol Scand 36:325-336. CLARKSON, B.H.; WEFEL, J.S.; and SILVERSTONE, L.M. (1981): Redistribution of Enamel Fluoride during White Spot Lesion Formation, Caries Res 15:158-165. DOWSE, C.M. and JENKINS, G.N. (1957): Fluoride Uptake in vivo in Enamel Defects and its Significance, J Dent Res 36:816. DIJKMAN, A.G. and ARENDS, J. (1988): The Role of CaF 2-like Material in Topical Fluoridation of Enamel in situ, Acta Odontol Scand 46:391-397. DIJKMAN, A.G.; DE BOER, P.; and ARENDS, J. (1983): In vivo Investigations on the F-content in and on Human Enamel After Topical Applications, Caries Res 17:392-402. FEJERSKOV, 0.; THYLSTRUP, A.; and LARSEN, M.J. (1981): Rational Use of Fluorides in Caries Prevention. A Concept Based on the Possible Cariostatic Mechanisms, Acta Odontol Scand 39:241-249. GROBLER, S.R.; 0GAARD, B.; and ROLLA, G. (1981): Uptake and Retention of Fluoride in Sound Dental Enamel in vivo After a Single Application of Neutral 2% Sodium Fluoride. In: Tooth Surface Interactions and Preventive Dentistry, G. Rolla, T. Senju, and G. Embery, Eds., London: Information Retrieval, Inc., pp. 17-25.
Vol. 69 Special Issue
LOOSELY BOUND FLUORIDE IN DENTAL CARIES
HALLSWORTH, A.S. and WEATHERELL, J.A. (1969): The Microdistribution, Uptake, and Loss of Fluoride in Human Enamel, Caries Res 3:109-118. HARDWICK, J.L.; TEASDALE, J.; and BLOODWORTH, G. (1982): Caries Increments over 4 Years in Children aged 12 at the Start of Water Fluoridation, Br Dent J 153:217-222. LAGERLOF, F.; SAXEGAARD, E.; BARKVOLL, P.; and ROLLA, G. (1988): Effects of Inorganic Orthophosphate and Pyrophosphate on the Dissolution of Calcium Fluoride in Water, J Dent Res 67:447-449. LARSON, R.H.; MELLBERG,~·1.R.; ENGLANDER, H.R.; and SENNING, R. (1976): Caries Inhibition in the Rat by Water-borne and Enamel-bound Fluoride, Caries Res 10:321-33l. LITTLE, M.F. and STEADMAN, L.T. (1966): Chemical and Physical Properties of Altered and Sound Enamel IV, Arch Oral Bioi 11:273-278. LODDING, A. and ODELIUS, H. (1978): In Depth Distributions of Elements in Teeth Determined by Ion Probe Analysis, Microscopica Acta (Suppl 2): 367-376. MANLY, R.S. and HARRINGTON, D.P. (1959): Solution Rate of Tooth Enamel in an Acetate Buffer, J Dent Res 38: 910-919. MARGOLIS, H.C. and MORENO, E.C. (1985): Kinetic and ThermodynamicAspects of Enamel Demineralization, Caries Res 19:2235. MARGOLIS, H.C.; MORENO, E.C.; and MURPHY, B.J. (1986): Effect of Low Levels of Fluoride in Solution on Enamel Demineralization in vitro, J Dent Res 65:23-29. McCANN, H.G. (1968): The Solubility of Fluorapatite and its Relationship to that of CaF2, Arch Oral Bioi 13:987-100l. NELSON, D.G.A.; FEATHERSTONE, J.D.B.; DUNCAN, J.F.; and CUTRESS, T.W. (1983a): Effect of Carbonate and Fluoride on the Dissolution Behavior of Synthetic Apatites, Caries Res 17:20021l. NELSON, D.G.A.; JONGEBLOED, W.L.; and ARENDS, J. (1983b): Morphology of Enamel Surfaces Treated with Topical Fluoride Agents: SEM Considerations, J Dent Res 62:1201-1208. NELSON, D.G.A.; JONGEBLOED, W.L.; and ARENDS, J. (1984): Crystallographic Structure of Enamel Surfaces Treated with Topical Fluoride Agents: TEM and XRD Considerations, J Dent Res 63:6-12. 0GAARD, B.; ROLLA, G.; and ARENDS, J. (1988): In vivo Progress of Enamel and Root Surface Lesions under Plaque as a Function of Time, Caries Res 22:302-305.
605
0GAARD, B.; ROLLA, G.; and HELGELAND, K. (1983): Alkali Soluble and Alkali Insoluble Fluoride Retention in Demineralized Enamel in vivo, Scand J Dent Res 9:200-204. 0GAARD, B.; ROLLA, G.; RUBEN, 1.; and ARENDS, 1. (1988): Microradiographic Study of Demineralization of Shark Enamel in a Human Caries Model, Scand J Dent Res 96:209-21l. 0GAARD, B.; ROLLA, G.; and TEN CATE, 1.M. (1987): Calcium Fluoride Protects Enamel against Caries in vivo, Caries Res 21:159. PETERSSON, L.G.; ODELIUS, H.; LODDING, A.; LARSEN, S.J.; and FROSTELL, G. (1976): Ion Probe Study of F Gradients in Outermost Layers of Human Enamel, J Dent Res 55:980-990. POULSEN, S. and LARSEN, M.J. (1975): Dental Caries in Relation to Fluoride Content of Enamel in the Primary Dentition, Caries Res 9:59-65. REINTSEMA, H. and ARENDS, 1. (1988): An in vivo Study of Microhardness and Fluoride Uptake in Partially Demineralized Human Enamel Covered by Plaque, J Dent Res 67:471-473. REINTSEMA, H.; SCHUTHOF, 1.; and ARENDS, J. (1985): An in vivo Investigation of the Fluoride Uptake in Partially Demineralized Human Enamel from Several Different Dentifrices, J Dent Res 64:19-23. ROLLA, G. and 0GAARD, B. (1986): Studies on the Solubility of Calcium Fluoride in Human Saliva. In: Factors Relating to Demineralisation and Remineralisation of the Teeth, S.A. Leach, Ed., Oxford: IRL Press Limited, pp. 45-50. SCHAMSCHULA, R.G.; AGUS, H.; CHARLTON, G.; DUPPENTHALER, J.L.; and UN, P. (1979): Associations between Fluoride Concentration in Successive Layers of Human Enamel and Individual Dental Caries Experience, Arch Oral Bioi 24:847-859. SHERN, R.J.; DRISCOLL, W.S.; and KORTS, D.C. (1977): Enamel Biopsy Results of Children Receiving Fluoride Tablets, JAm Dent Assoc 95:310-314. TEN CATE, J.M. and DUlJSTERS, P.P.E. (1983a): Influence of Fluoride in Solution on Tooth Demineralization.!. Chemical Data, Caries Res 17:193-199. TEN CATE, J.M. and DUlJSTERS, P.P.E. (1983b): Influence of Fluoride in Solution on Tooth Demineralization. II. Microradiographic Data, Caries Res 17:513-519. VAN DER MERWE, E.H.M.; BISCHOFF, J.1.; FATTI, L.P.; RETIEF, D.H.; BARBAKOW, F.H.; and FRIEDMAN, M. (1977): Relationships Between Fluoride in Enamel, DMFT Index and Fluorosis in High- and Low-Fluoride Areas in South Africa, Community Dent Oral Epidemiol 5:61-64.
