Effects of Fluoride on Caries Development and Progression in vivo B. 0GAARD University of Oslo Dental Faculty, Department of Orthodontics, P. O. Box 1109, Blindem, N-0317 Oslo 3, Norway Thedissolution rate ofcalcium-fluoride-like material from the enamel surface in vivo appears to be much slower than previously though~. This could be due to adsorption of phosphate IOns and/or protein molecules to the surface ofthe calcium-fluoride-like particles. During cariogenic challenges, the phosphate/protein coating is released, resulting in increased solubility rate of the calcium-fluoride-like material. Due to this mechanism, calcium-fluoride-like material may be a major aspect of the cariostatic mechanism of topically applie~ fl~. ride. Topically applied neutral fluoride agents are able to inhibit caries development in enamel but not completely stop lesion development. A fluoride solution at low pH has been found to be m~re effective in caries model studies than neutral fluoride a¥ents, w~lch might be due to the formation of a larger depot of calcium fluonde. Data from fluoridated areas indicate that the flu~ride ion as.such has a limited effect on lesion development, and a major mechanism of the cariostatic effect may be reformation of apatite [remineralization}. The product of lesion consolidation (a fluoridated apatite) may have a limited effect, since intra-oral caries model studies show that even pure fluorapatite, in the form of shark enamel, demineralizes. In [ISsures and around orthodontic appliances, conventional fluoride agents appear to have only a small effect.
J Dent Res 69(Spec 155):813-819, February, 1990
Reactions of fluoride with enamel. Theoretical considerations. -The fluoride ion in aqueous solution hasa highaffinity for enamel.The details of this process depend upon such factors as concentration of the fluoride in solution, pH, exposure time, and. the nature of the enamel surface (e.g., sound, etc~ed, or cano~s)(Arends a~~ ten Cate, 1981). Fluoride may be incorporated into the apanttcstructure by ionexchange or through a dissolution/reprecil?itation process. Fluoride may also adsorb onto the enamel mmeral.. Furthermore, it has been suggested that at n~utral pH fluonde reacts with carbonate and acid phosphates 10 the enamel to form a calcium-fluoride-like material (hereafter referred to as CaF2 ) . At low pH, CaF2 may form by decomposition of the enamel (Gerould, 1945; Neuman et al., 1950; McCann, 1953; Leach, 1959; Trautzand Zapanta, 1961; Spinelli et al., 1971; Ramsey et al., 1973; Larsen et al., 1977; Gren, 1977; 0gaard et al., 1988b). It is generally claimed that it does not form fr?m fluoride solutions containing less than 75-100 ppm fluonde (Leach, 1959). Ten Cate and Arends (1977) suggest that.t~e ion-activity product for CaF2 may well exceed the solubility product within caries lesions, even with solutions of low fluoride concentrations. CaF2 material has recently been demonstrated in lesions following useof fluoridated toothpastes (Arends et al., 1988). . . CaF2 is observable in the electron microscope as sphencal globules on the enamel (Fig. 1) (Gerould, 1945; Larsen and Fejerskov, 1978).The amount and size of.the glob~les depend on the pH and concentration of the fluonde solution ~nd the period of exposure (Saxegaard and Rella, 1988).The diameter of the globules has been found to vary from 4-15 nm to 311m, Presented al a Joint IADR/ORCA International Symposium on Fluorides: Mechanisms of Action and Recommendations for Use, held March 21-24, 1989, Callaway Gardens Conference Center, Pine Mountain, Georgia
Fig. 1 - Globules of calcium fluoride-like material on sound enamel after topical treatment with 2% NaF for 24 h. 8000 X.
and they appear to form preferentially in prism depressions and focal holes along the perikymata pattern which are more soluble than other parts of the enamel (Nelson et al., 1983, 1984). CaF2 formed on enamel in vivo has chemical properties different from those of chemically pure calcium fluoride (0gaard, 1988b). In the pH range 6.5 to 8.5, phosphate in the form of HPOl- has been found to interfere with its formation (Chander et al., 1982; Christoffersen et al., 1988). In the in vivo situation, caries-susceptible sites (fissures, approximal, and at gingival margins) are covered with pla9ue mostof thetime. The enamel surface is therefore not accessible for an immediate reaction with the fluoride ion. The fluoride concentration in plaque has been found to vary considerably, an accepted level b~ing about 5-10 ppm fluoride f~om low fluoride areas (Jenkins and Edgar, 1977). The fluonde concentrations in plaque fluid are, however, quite low (Moreno and Margolis, 1988). Tatevossian and Gould (1976) showed that plaque fluid contains about 6.5 mmoVL calcium, but the calcium ion activity in plaque fluid has recently been measured as 0.85 mmol/L by use of an ion sl?~cific electrode (Carey ~t al., 1986). Accordingly, the solubility product of CaF2 WIll clearly be exceeded when plaque is exposed to highly concentrated agents like fluoride toothpastes or mouthrinses (0gaard, 1985; R0lla and 0gaard, 1986; Lagerlof et al., 1988a). There is most likely a reservoir of CaF2 in all dental plaque when fluoride toothpastes are used regularly. Carey et al. (1986) reported fluoride ion activities in plaque flui? and noted. i.t to be supersaturated with respect to fluorapa~lte .. I~ addllJ?~, Kaufman and Kleinberg (1973) have also identified apanuc deposits in dental plaque. White and Nancollas (1990).~ave shown that fluorapatite also forms under the same conditions in which CaF2 is formed. Nevertheless, both CaF2 and fluorapatite may liberate fluoride during ~ariogenic ch~llenges .. Clinical data.-A common techmque for fluonde gradient determination of enamel is consecutive acid etchings of thin enamel layers. Caslavska et al. (1975) and Rella and Bowen (1978) reported that a strongly alkaline solution (1 mol/LKOH 813
Downloaded from jdr.sagepub.com at CMU Libraries - library.cmich.edu on October 4, 2015 For personal use only. No other uses without permission. © 1990 International & American Associations for Dental Research
814
0GAARD
J Dent Res February 1990
for 24 h) would dissolve the CaF 2 and adsorbed fluoride as well without affecting the apatitically-bound fluoride. By combining the Caslavska method with the acid-etching technique, information could be obtained of the reaction product formed in or on enamel after topical fluoride treatment (Dijkman, 1982; 0gaard, 1985). Sound ename/.-Figs. 2 and 3 show the results from clinical studies that were conducted to investigate the retention of the reaction products formed on sound enamel after treatment with common fluoride agents (Grobler et al., 1981; 0gaard et al., 1983a, 1984). The amount of fluoride retained on sound enamel after a single application of 2% NaF decreased from an increase of 142% 15 min after treatment to 33% two weeks later. Although a considerable part of the deposited fluoride had leached away, CaF 2 still accounted for the main increase in the total fluoride retentate two weeks later (Grobler et al., 1981). The greatest amount of fluoride was retained after Duraphat® treatment (0gaard et a/., 1984); less fluoride was retained after mouth rinsing (0gaard et al., 1983a). However, in all treatment groups, CaF 2 was present on the enamel surface for at least two weeks. This relatively high stability of CaF 2 in the oral environment is greater than generally thought and
140 120 Mean increase in % 100
m
Total F
80 60
0
CaF2
~
FAp
40
D
20
o
15 min 2 weeks Fig. 2 - Increase in alkali-soluble (CaF 2·like) and alkali-insoluble (FOHap) fluoride in sound enamel 15 min and two weeks after a single 2% NaF application for four min. (From Grobler et al., 1981.)
o
60
in %
Alkali soluble F-CaF2
~ Alkali insoluble F-FAp
Mean increase 40
20
o 2 2% NaF
Daily rinsing 0.05% NaF
Etching no
Duraphat 5% Na F
Fig. 3 - Increase in alkali-soluble (CaF 2-like) and alkali-insoluble (FOHap) fluoride in sound enamel two weeks after a single application with 2% NaF or Duraphat or two weeks after daily rinsing with 0.05% NaF. From Grobler et at. (1981); 0gaard et at. (1983a, 1984).
has also been shown in several other studies (Caslavska et al., 1971; DePaola et al., 1978; Larsen et al., 1981; Dijkman et al., 1983; Dijkman and Arends, 1988; Saxegaard and Rella, 1989). Carious enamel.-In an experiment on retention of fluoride in carious enamel, 21 orthodontic patients participated. Caries lesions were induced on the buccal surfaces of premolars scheduled to be extracted for orthodontic reasons, by use of an orthodontic banding technique (0gaard et al., 1983b, 1984). The caries induction period was four weeks, after which three common fluoride agents were applied to the lesions, the cariogenic milieu re-established, and the teeth extracted after an additional two-week period. The lesions underneath dental plaque contained considerably higher amounts of fluoride, and a larger fraction was retained in an alkali-insoluble form (e.g., as fluorhydroxyapatite) compared with the experiments on sound enamel (Fig. 4) (0gaard et al., 1983b, 1984). Significant uptake of fluoride was also found in subsequent etchings of the enamel lesions. Whereas a steep fluoride gradient was observed in sound enamel, this was less pronounced after topical treatment of demineralized enamel. The plaque covering the enamel lesions may have increased the retention of fluoride by reducing leakage from the surface. The porosity of the lesions may also account for the increased fluoride uptake and for making the fluoride gradient in subsequent layers less pronounced than in sound enamel. The results confirm other reports demonstrating high fluoride content in carious or presoftened enamel (Myers et al., 1952; Dowse and Jenkins, 1957; Hallsworth et al., 1975; Bruun et al., 1983; Larsen et al., 1981).
Possible reasons for a prolonged retention of CaFz in the oral environment. The dissolution rate of CaF 2 from the enamel surface in vivo seems to be much slower than generally assumed. Keyes and Englander (1975) proposed that pellicle proteins soon coat the CaF 2 in vivo and thus inhibit an immediate dissolution. Lindeman (1985) has shown that albumin and glycoproteins adsorb rapidly onto calcium fluoride in vitro. Nelson et at. (1983) observed that CaF 2 in the form of extremely small particles is produced after a single topical fluoride treatment of sound enamel in vitro. They furthermore showed that CaF2 may be deposited in natural enamel defects or prism pits produced by the topical agent, being most pronounced by acidic solutions (Nelson et al., 1984). CaF 2 in these prism depressions, lying parallel to the perikymata, would not be as easily washed away or removed mechanically as CaF 2 deposited on smooth surfaces. Bruun et al. (1983) found the dissolution rate of CaF 2 even in distilled water to be an extremely slow process. Kanaya et al. (1983) observed that calcium and phosphate ions in a remineralizing system inhibited dissolution of CaF 2 , and suggested that an insoluble surface complex was formed. Chander et al. (1982) showed that in the pH range 6.5 to 8.5, phosphate in the form of HPOl- adsorbed onto CaF 2 • They also found that under the experimental condition, part of the CaF 2 may transform into fluorapatite. Rella and 0gaard (1986) demonstrated that CaF z had a limited rate of solubility in saliva which was most likely due to adsorption of phosphate ions and protein molecules onto the surfaces of the CaFz- particles. Reduction in pH from 7 to 5 or 4 gave an increased rate of solubility of phosphate/protein-coated CaF2 in distilled water, as was recently confirmed by Lagerli:if et al, (1988b). It is suggested that phosphate/protein-coated CaF 2 may serve as a pH-controlled depot of fluoride for release during cariogenic
Downloaded from jdr.sagepub.com at CMU Libraries - library.cmich.edu on October 4, 2015 For personal use only. No other uses without permission. © 1990 International & American Associations for Dental Research
EFFECTS OF FLUORIDE ON CARIES
Vol. 69 Special Issue
815
120 100 Mean increase in %
80
D
Alkali soluble F Ca F2
~
Alkali insoluble F FAp
60 40
20
o 2 3 2% NaF
123
1 2 3
Daily rinsing 0.05% NaF
Duraphat 5% NaF
Etching no
Fig. 4 - Increase in alkali-soluble (CaF2-like) and alkali-insoluble (FAp) fluoride in carious enamel underneath dental plaque two weeks after a single application with 2% NaF or Duraphat or two weeks with daily rinsing with 0.05% NaF. From 0gaard et at. (1983b, 1984).
challenges. A possible transformation of part of the CaF 2 into fluorapatite may even further reduce the solubility rate in the oral environment.
Effect of fluoride on enamel caries development. Depending on the eating habits, the plaque-covered enamel surface is subjected to several acid attacks during the day, and mineral is lost on a microlevel. Between the pH drops, the plaque fluid is supersaturated to the enamel mineral, and repair can occur. When the cariogenic challenges progress to such an extent that a situation of unbalance between demineralization and remineralization exists, a visible defect in the enamel will appear in due time. Several recent reports have shown that the initial caries lesion is a surface softening of the enamel (e.g., without a surface layer) and later on develops into a subsurface lesion with a surface layer (for review of the available in vivo data, see Arends and ten Bosch, 1986; Arends and Christoffersen, 1986). It is generally postulated that a fluoridated apatite is the most desirable product of topical treatment and that the effectiveness of a topical agent is proportional to its ability to deposit fluoride as fluorapatite in the enamel. Nearly all presently available fluoride agents are based on this concept. The rationale for this hypothesis is that a high fluoride content in drinking water causes a high fluoride content in surface enamel and a significant reduction in caries (Isaac et at., 1958; Brudevold et al., 1967). However, only a very moderate increase in the firmly bound fluoride is obtained with these agents (Gren, 1977). It might even be possible that the measured increase in the amount of firmly bound fluoride could be CaF 2 "protected" by a layer of fluorapatite, according to the observation by Chander et at. (1982). Such small amounts of apatitically bound fluoride as are deposited in the enamel after topical treatment are unlikely to account for the large cariostatic effect (Fejerskov et al., 1981). Recently, it was found that even shark enamel, containing nearly exclusively fluorapatite, had a limited resistance against caries attacks in an intra-oral human caries model (0gaard et al., 1988c).
From in vitro experiments, it is well-established that low levels of fluoride in acidic buffer solutions may completely inhibit lesion development in enamel (Christoffersen and Arends, 1982; Larsen, 1990). The in vivo situation is much more complex, since it is an open system where plaque fluid may exchange with saliva in periods when acid attacks occur. The optimum fluoride level that would inhibit caries development may therefore vary from individual to individual, and also most likely from site to site within the dentition (e.g., fissures, approximal surfaces). In vivo model studies on the effect of fluoride on lesion development are relatively few. A study with the orthodontic banding model showed that daily rinsing with a neutral 0.2% NaF solution (48 mmol/L F as NaF) reduced lesion depth by a factor of 3 and mineral loss by 60% during a four-week period compared with a control group (0gaard et al., 1986). This supports an investigation by Reintsema et at. (1985) which showed that when the enamel is subjected to a more or less continuous cariogenic challenge, conventional fluoride agents reduce the rate of lesion development but do not completely inhibit it. A more pronounced effect on caries development has been obtained with application of a fluoride solution (0.6% F or 0.32 mol/L F) at pH 1.9. This solution inhibited enamel caries development completely with the orthodontic banding model (0gaard et al., 1988b) (Fig. 5). The low-pH fluoride solution produced twice as much CaF 2 on enamel as did a neutral 2% NaF solution in vitro (0gaard et al., 1988b). Since the major reaction product on enamel after topical treatment with these agents is CaF 2 , this salt clearly has cariostatic properties. It seems that fluoride solutions at low pH which produce large amounts of CaF 2 may have improved clinical properties. The mechanism for the cariostatic effect is not clearly established. Of great importance is probably CaF 2 in local natural enamel defects along the perikymata overlappings, since these spots appear to be the initial sites of acid penetration during lesion development (Nelson et al., 1983). Fluoride released from this reservoir during caries attacks may diffuse into the enamel and promote reformation of apatite. Fluoride may also
Downloaded from jdr.sagepub.com at CMU Libraries - library.cmich.edu on October 4, 2015 For personal use only. No other uses without permission. © 1990 International & American Associations for Dental Research
816
J Dent Res February 1990
0GAARD
adsorb onto the enamel crystallites and inhibit dissolution (Arends and Christoffersen, 1990). It is well-known that fluoride has a catalytic effect by transforming soluble calcium phosphates in the lesion into fluorapatite (Brudevold et al., 1965). However, the product (fluorapatite) probably has a limited value, whereas the remineralization process as such may be crucial. This conclusion supports the view that the fluoride ion activity during the caries process is more important than a high content of fluoride in the enamel (Fejerskov et al., 1981). Clinical data clearly show that the frequency of fluoride applications is more important and that the benefits gradually diminish after fluoride supplementation is stopped (Ericsson, 1977). The clinical effect of sodium fluoride and monofluorophosphate (MFP) is in the same order of magnitude, although differences in action have been proposed since the fluorine of MFP is covalently bound rather than in an ionic form (Ingram, 1972). Several authors have shown that MFP is hydrolyzed by enzymes in plaque (for review, see 0gaard et al., 1985). It is not established whether CaF2 can form in plaque, in lesions, or on the enamel from MFP application. However, fluoride released from MFP would increase the fluoride ion activity and therefore most likely affect the caries process in the same way as fluoride released from CaF2 • In a recent report, topical treatment in the form of mouthrinsing (0.2% NaF) did not inhibit lesion development in shark enamel (3.2% F) in an intra-oral caries model (Fig. S)(0gaard et al., 1989a). This may indicate that fluoride has a limited caries-preventive potential during certain conditions. Bacteria associated with caries (S. mutans and lactobacilli) may lower
1500
the pH in dental plaque below 4.5 (Sandham and Philips, 1976). At this low pH, even fluorapatite will dissolve (McCann, 1968), as was clearly shown in the shark enamel experiments. When plaque fluid is undersaturated with respect to fluorapatite, no redeposition of mineral lost can occur. Although low levels of fluoride in the liquid phase may inhibit dissolution of fluorapatite in vitro due to common ion effect, or by making the liquid phase supersaturatedwith respect to fluorapatite (McCann, 1968), this seems not to take place in vivo, where fluoride ions in plaque fluid are most likely lost due to a sink effect or bind rapidly to plaque components. A low-pH fluoride solution that inhibits lesion development in human enamel has no measurable effect on shark enamel (Fig. 5) (0gaard et al., 1989b). It may be speculated that CaF2 is more rapidly and completely formed on human enamel than on shark enamel containing nearly pure fluorapatite (Trautz and Zapanta, 1961). A massive shell of CaF2 on enamel, acting as an acid-resistant barrier (Gray and Francis, 1962; ten Cate and Duijsters, 1982) and/or a reservoir of fluoride for slow release, may therefore have clinical effects, even when plaque fluid is undersaturated with respect to fluorapatite.
Effect of fluoride on enamel caries progression. The established, visible caries lesion (white spot or cavity) is a result of a fast or slow intermittent process which depends on the individual's fluoride supplementation and cariogenic challenge (e.g., sucrose consumption, oral hygiene, salivary properties).
--
.
SHARK ENAMEL 32000 ppm F
Mineral loss
--
.1 Z=Vol%.l.lm
,....-
HUMAN ENAMEL
J Dent Res 69(Spec 155):813-819, February, 1990
Reactions of fluoride with enamel. Theoretical considerations. -The fluoride ion in aqueous solution hasa highaffinity for enamel.The details of this process depend upon such factors as concentration of the fluoride in solution, pH, exposure time, and. the nature of the enamel surface (e.g., sound, etc~ed, or cano~s)(Arends a~~ ten Cate, 1981). Fluoride may be incorporated into the apanttcstructure by ionexchange or through a dissolution/reprecil?itation process. Fluoride may also adsorb onto the enamel mmeral.. Furthermore, it has been suggested that at n~utral pH fluonde reacts with carbonate and acid phosphates 10 the enamel to form a calcium-fluoride-like material (hereafter referred to as CaF2 ) . At low pH, CaF2 may form by decomposition of the enamel (Gerould, 1945; Neuman et al., 1950; McCann, 1953; Leach, 1959; Trautzand Zapanta, 1961; Spinelli et al., 1971; Ramsey et al., 1973; Larsen et al., 1977; Gren, 1977; 0gaard et al., 1988b). It is generally claimed that it does not form fr?m fluoride solutions containing less than 75-100 ppm fluonde (Leach, 1959). Ten Cate and Arends (1977) suggest that.t~e ion-activity product for CaF2 may well exceed the solubility product within caries lesions, even with solutions of low fluoride concentrations. CaF2 material has recently been demonstrated in lesions following useof fluoridated toothpastes (Arends et al., 1988). . . CaF2 is observable in the electron microscope as sphencal globules on the enamel (Fig. 1) (Gerould, 1945; Larsen and Fejerskov, 1978).The amount and size of.the glob~les depend on the pH and concentration of the fluonde solution ~nd the period of exposure (Saxegaard and Rella, 1988).The diameter of the globules has been found to vary from 4-15 nm to 311m, Presented al a Joint IADR/ORCA International Symposium on Fluorides: Mechanisms of Action and Recommendations for Use, held March 21-24, 1989, Callaway Gardens Conference Center, Pine Mountain, Georgia
Fig. 1 - Globules of calcium fluoride-like material on sound enamel after topical treatment with 2% NaF for 24 h. 8000 X.
and they appear to form preferentially in prism depressions and focal holes along the perikymata pattern which are more soluble than other parts of the enamel (Nelson et al., 1983, 1984). CaF2 formed on enamel in vivo has chemical properties different from those of chemically pure calcium fluoride (0gaard, 1988b). In the pH range 6.5 to 8.5, phosphate in the form of HPOl- has been found to interfere with its formation (Chander et al., 1982; Christoffersen et al., 1988). In the in vivo situation, caries-susceptible sites (fissures, approximal, and at gingival margins) are covered with pla9ue mostof thetime. The enamel surface is therefore not accessible for an immediate reaction with the fluoride ion. The fluoride concentration in plaque has been found to vary considerably, an accepted level b~ing about 5-10 ppm fluoride f~om low fluoride areas (Jenkins and Edgar, 1977). The fluonde concentrations in plaque fluid are, however, quite low (Moreno and Margolis, 1988). Tatevossian and Gould (1976) showed that plaque fluid contains about 6.5 mmoVL calcium, but the calcium ion activity in plaque fluid has recently been measured as 0.85 mmol/L by use of an ion sl?~cific electrode (Carey ~t al., 1986). Accordingly, the solubility product of CaF2 WIll clearly be exceeded when plaque is exposed to highly concentrated agents like fluoride toothpastes or mouthrinses (0gaard, 1985; R0lla and 0gaard, 1986; Lagerlof et al., 1988a). There is most likely a reservoir of CaF2 in all dental plaque when fluoride toothpastes are used regularly. Carey et al. (1986) reported fluoride ion activities in plaque flui? and noted. i.t to be supersaturated with respect to fluorapa~lte .. I~ addllJ?~, Kaufman and Kleinberg (1973) have also identified apanuc deposits in dental plaque. White and Nancollas (1990).~ave shown that fluorapatite also forms under the same conditions in which CaF2 is formed. Nevertheless, both CaF2 and fluorapatite may liberate fluoride during ~ariogenic ch~llenges .. Clinical data.-A common techmque for fluonde gradient determination of enamel is consecutive acid etchings of thin enamel layers. Caslavska et al. (1975) and Rella and Bowen (1978) reported that a strongly alkaline solution (1 mol/LKOH 813
Downloaded from jdr.sagepub.com at CMU Libraries - library.cmich.edu on October 4, 2015 For personal use only. No other uses without permission. © 1990 International & American Associations for Dental Research
814
0GAARD
J Dent Res February 1990
for 24 h) would dissolve the CaF 2 and adsorbed fluoride as well without affecting the apatitically-bound fluoride. By combining the Caslavska method with the acid-etching technique, information could be obtained of the reaction product formed in or on enamel after topical fluoride treatment (Dijkman, 1982; 0gaard, 1985). Sound ename/.-Figs. 2 and 3 show the results from clinical studies that were conducted to investigate the retention of the reaction products formed on sound enamel after treatment with common fluoride agents (Grobler et al., 1981; 0gaard et al., 1983a, 1984). The amount of fluoride retained on sound enamel after a single application of 2% NaF decreased from an increase of 142% 15 min after treatment to 33% two weeks later. Although a considerable part of the deposited fluoride had leached away, CaF 2 still accounted for the main increase in the total fluoride retentate two weeks later (Grobler et al., 1981). The greatest amount of fluoride was retained after Duraphat® treatment (0gaard et a/., 1984); less fluoride was retained after mouth rinsing (0gaard et al., 1983a). However, in all treatment groups, CaF 2 was present on the enamel surface for at least two weeks. This relatively high stability of CaF 2 in the oral environment is greater than generally thought and
140 120 Mean increase in % 100
m
Total F
80 60
0
CaF2
~
FAp
40
D
20
o
15 min 2 weeks Fig. 2 - Increase in alkali-soluble (CaF 2·like) and alkali-insoluble (FOHap) fluoride in sound enamel 15 min and two weeks after a single 2% NaF application for four min. (From Grobler et al., 1981.)
o
60
in %
Alkali soluble F-CaF2
~ Alkali insoluble F-FAp
Mean increase 40
20
o 2 2% NaF
Daily rinsing 0.05% NaF
Etching no
Duraphat 5% Na F
Fig. 3 - Increase in alkali-soluble (CaF 2-like) and alkali-insoluble (FOHap) fluoride in sound enamel two weeks after a single application with 2% NaF or Duraphat or two weeks after daily rinsing with 0.05% NaF. From Grobler et at. (1981); 0gaard et at. (1983a, 1984).
has also been shown in several other studies (Caslavska et al., 1971; DePaola et al., 1978; Larsen et al., 1981; Dijkman et al., 1983; Dijkman and Arends, 1988; Saxegaard and Rella, 1989). Carious enamel.-In an experiment on retention of fluoride in carious enamel, 21 orthodontic patients participated. Caries lesions were induced on the buccal surfaces of premolars scheduled to be extracted for orthodontic reasons, by use of an orthodontic banding technique (0gaard et al., 1983b, 1984). The caries induction period was four weeks, after which three common fluoride agents were applied to the lesions, the cariogenic milieu re-established, and the teeth extracted after an additional two-week period. The lesions underneath dental plaque contained considerably higher amounts of fluoride, and a larger fraction was retained in an alkali-insoluble form (e.g., as fluorhydroxyapatite) compared with the experiments on sound enamel (Fig. 4) (0gaard et al., 1983b, 1984). Significant uptake of fluoride was also found in subsequent etchings of the enamel lesions. Whereas a steep fluoride gradient was observed in sound enamel, this was less pronounced after topical treatment of demineralized enamel. The plaque covering the enamel lesions may have increased the retention of fluoride by reducing leakage from the surface. The porosity of the lesions may also account for the increased fluoride uptake and for making the fluoride gradient in subsequent layers less pronounced than in sound enamel. The results confirm other reports demonstrating high fluoride content in carious or presoftened enamel (Myers et al., 1952; Dowse and Jenkins, 1957; Hallsworth et al., 1975; Bruun et al., 1983; Larsen et al., 1981).
Possible reasons for a prolonged retention of CaFz in the oral environment. The dissolution rate of CaF 2 from the enamel surface in vivo seems to be much slower than generally assumed. Keyes and Englander (1975) proposed that pellicle proteins soon coat the CaF 2 in vivo and thus inhibit an immediate dissolution. Lindeman (1985) has shown that albumin and glycoproteins adsorb rapidly onto calcium fluoride in vitro. Nelson et at. (1983) observed that CaF 2 in the form of extremely small particles is produced after a single topical fluoride treatment of sound enamel in vitro. They furthermore showed that CaF2 may be deposited in natural enamel defects or prism pits produced by the topical agent, being most pronounced by acidic solutions (Nelson et al., 1984). CaF 2 in these prism depressions, lying parallel to the perikymata, would not be as easily washed away or removed mechanically as CaF 2 deposited on smooth surfaces. Bruun et al. (1983) found the dissolution rate of CaF 2 even in distilled water to be an extremely slow process. Kanaya et al. (1983) observed that calcium and phosphate ions in a remineralizing system inhibited dissolution of CaF 2 , and suggested that an insoluble surface complex was formed. Chander et al. (1982) showed that in the pH range 6.5 to 8.5, phosphate in the form of HPOl- adsorbed onto CaF 2 • They also found that under the experimental condition, part of the CaF 2 may transform into fluorapatite. Rella and 0gaard (1986) demonstrated that CaF z had a limited rate of solubility in saliva which was most likely due to adsorption of phosphate ions and protein molecules onto the surfaces of the CaFz- particles. Reduction in pH from 7 to 5 or 4 gave an increased rate of solubility of phosphate/protein-coated CaF2 in distilled water, as was recently confirmed by Lagerli:if et al, (1988b). It is suggested that phosphate/protein-coated CaF 2 may serve as a pH-controlled depot of fluoride for release during cariogenic
Downloaded from jdr.sagepub.com at CMU Libraries - library.cmich.edu on October 4, 2015 For personal use only. No other uses without permission. © 1990 International & American Associations for Dental Research
EFFECTS OF FLUORIDE ON CARIES
Vol. 69 Special Issue
815
120 100 Mean increase in %
80
D
Alkali soluble F Ca F2
~
Alkali insoluble F FAp
60 40
20
o 2 3 2% NaF
123
1 2 3
Daily rinsing 0.05% NaF
Duraphat 5% NaF
Etching no
Fig. 4 - Increase in alkali-soluble (CaF2-like) and alkali-insoluble (FAp) fluoride in carious enamel underneath dental plaque two weeks after a single application with 2% NaF or Duraphat or two weeks with daily rinsing with 0.05% NaF. From 0gaard et at. (1983b, 1984).
challenges. A possible transformation of part of the CaF 2 into fluorapatite may even further reduce the solubility rate in the oral environment.
Effect of fluoride on enamel caries development. Depending on the eating habits, the plaque-covered enamel surface is subjected to several acid attacks during the day, and mineral is lost on a microlevel. Between the pH drops, the plaque fluid is supersaturated to the enamel mineral, and repair can occur. When the cariogenic challenges progress to such an extent that a situation of unbalance between demineralization and remineralization exists, a visible defect in the enamel will appear in due time. Several recent reports have shown that the initial caries lesion is a surface softening of the enamel (e.g., without a surface layer) and later on develops into a subsurface lesion with a surface layer (for review of the available in vivo data, see Arends and ten Bosch, 1986; Arends and Christoffersen, 1986). It is generally postulated that a fluoridated apatite is the most desirable product of topical treatment and that the effectiveness of a topical agent is proportional to its ability to deposit fluoride as fluorapatite in the enamel. Nearly all presently available fluoride agents are based on this concept. The rationale for this hypothesis is that a high fluoride content in drinking water causes a high fluoride content in surface enamel and a significant reduction in caries (Isaac et at., 1958; Brudevold et al., 1967). However, only a very moderate increase in the firmly bound fluoride is obtained with these agents (Gren, 1977). It might even be possible that the measured increase in the amount of firmly bound fluoride could be CaF 2 "protected" by a layer of fluorapatite, according to the observation by Chander et at. (1982). Such small amounts of apatitically bound fluoride as are deposited in the enamel after topical treatment are unlikely to account for the large cariostatic effect (Fejerskov et al., 1981). Recently, it was found that even shark enamel, containing nearly exclusively fluorapatite, had a limited resistance against caries attacks in an intra-oral human caries model (0gaard et al., 1988c).
From in vitro experiments, it is well-established that low levels of fluoride in acidic buffer solutions may completely inhibit lesion development in enamel (Christoffersen and Arends, 1982; Larsen, 1990). The in vivo situation is much more complex, since it is an open system where plaque fluid may exchange with saliva in periods when acid attacks occur. The optimum fluoride level that would inhibit caries development may therefore vary from individual to individual, and also most likely from site to site within the dentition (e.g., fissures, approximal surfaces). In vivo model studies on the effect of fluoride on lesion development are relatively few. A study with the orthodontic banding model showed that daily rinsing with a neutral 0.2% NaF solution (48 mmol/L F as NaF) reduced lesion depth by a factor of 3 and mineral loss by 60% during a four-week period compared with a control group (0gaard et al., 1986). This supports an investigation by Reintsema et at. (1985) which showed that when the enamel is subjected to a more or less continuous cariogenic challenge, conventional fluoride agents reduce the rate of lesion development but do not completely inhibit it. A more pronounced effect on caries development has been obtained with application of a fluoride solution (0.6% F or 0.32 mol/L F) at pH 1.9. This solution inhibited enamel caries development completely with the orthodontic banding model (0gaard et al., 1988b) (Fig. 5). The low-pH fluoride solution produced twice as much CaF 2 on enamel as did a neutral 2% NaF solution in vitro (0gaard et al., 1988b). Since the major reaction product on enamel after topical treatment with these agents is CaF 2 , this salt clearly has cariostatic properties. It seems that fluoride solutions at low pH which produce large amounts of CaF 2 may have improved clinical properties. The mechanism for the cariostatic effect is not clearly established. Of great importance is probably CaF 2 in local natural enamel defects along the perikymata overlappings, since these spots appear to be the initial sites of acid penetration during lesion development (Nelson et al., 1983). Fluoride released from this reservoir during caries attacks may diffuse into the enamel and promote reformation of apatite. Fluoride may also
Downloaded from jdr.sagepub.com at CMU Libraries - library.cmich.edu on October 4, 2015 For personal use only. No other uses without permission. © 1990 International & American Associations for Dental Research
816
J Dent Res February 1990
0GAARD
adsorb onto the enamel crystallites and inhibit dissolution (Arends and Christoffersen, 1990). It is well-known that fluoride has a catalytic effect by transforming soluble calcium phosphates in the lesion into fluorapatite (Brudevold et al., 1965). However, the product (fluorapatite) probably has a limited value, whereas the remineralization process as such may be crucial. This conclusion supports the view that the fluoride ion activity during the caries process is more important than a high content of fluoride in the enamel (Fejerskov et al., 1981). Clinical data clearly show that the frequency of fluoride applications is more important and that the benefits gradually diminish after fluoride supplementation is stopped (Ericsson, 1977). The clinical effect of sodium fluoride and monofluorophosphate (MFP) is in the same order of magnitude, although differences in action have been proposed since the fluorine of MFP is covalently bound rather than in an ionic form (Ingram, 1972). Several authors have shown that MFP is hydrolyzed by enzymes in plaque (for review, see 0gaard et al., 1985). It is not established whether CaF2 can form in plaque, in lesions, or on the enamel from MFP application. However, fluoride released from MFP would increase the fluoride ion activity and therefore most likely affect the caries process in the same way as fluoride released from CaF2 • In a recent report, topical treatment in the form of mouthrinsing (0.2% NaF) did not inhibit lesion development in shark enamel (3.2% F) in an intra-oral caries model (Fig. S)(0gaard et al., 1989a). This may indicate that fluoride has a limited caries-preventive potential during certain conditions. Bacteria associated with caries (S. mutans and lactobacilli) may lower
1500
the pH in dental plaque below 4.5 (Sandham and Philips, 1976). At this low pH, even fluorapatite will dissolve (McCann, 1968), as was clearly shown in the shark enamel experiments. When plaque fluid is undersaturated with respect to fluorapatite, no redeposition of mineral lost can occur. Although low levels of fluoride in the liquid phase may inhibit dissolution of fluorapatite in vitro due to common ion effect, or by making the liquid phase supersaturatedwith respect to fluorapatite (McCann, 1968), this seems not to take place in vivo, where fluoride ions in plaque fluid are most likely lost due to a sink effect or bind rapidly to plaque components. A low-pH fluoride solution that inhibits lesion development in human enamel has no measurable effect on shark enamel (Fig. 5) (0gaard et al., 1989b). It may be speculated that CaF2 is more rapidly and completely formed on human enamel than on shark enamel containing nearly pure fluorapatite (Trautz and Zapanta, 1961). A massive shell of CaF2 on enamel, acting as an acid-resistant barrier (Gray and Francis, 1962; ten Cate and Duijsters, 1982) and/or a reservoir of fluoride for slow release, may therefore have clinical effects, even when plaque fluid is undersaturated with respect to fluorapatite.
Effect of fluoride on enamel caries progression. The established, visible caries lesion (white spot or cavity) is a result of a fast or slow intermittent process which depends on the individual's fluoride supplementation and cariogenic challenge (e.g., sucrose consumption, oral hygiene, salivary properties).
--
.
SHARK ENAMEL 32000 ppm F
Mineral loss
--
.1 Z=Vol%.l.lm
,....-
HUMAN ENAMEL
