J Clin Periodonto! 7997; 24: 8i-85Primed in Denmark . Ail rights reserved

Copyright © Munksgaard 7997

dinical periDdontolgjor/SS.V (1303-6979

Studies on stannous fluoridetoothpaste and gel(1). Antimicrobiai properties and staining potentialin vitro

William Wade\ Martin Addy ,̂Julie Hughes ,̂ Susan Mllsom'' andFrances Doherty^Divisions of. ^Restorative Dentistry and. ^OralMedicine. Pathology and Microbiology. DentalSchooi. University o1 Bristol. Bristoi. England.^Oral B Laboratories, USA

Wade W, Addy M, Hughes J, Milsom S. Doherty F: Studies on stannous fluoridetoothpaste and gel, (1). Antimicrobial properties and staining potential in vitro.J Clin Periodontol 1997: 24: 81-85. © Munksgaard, 1997.

Absrraet. Stannous fluoride (SF) in a toothpaste vehicle has the potential to pro-vide anticaries and plaque inhibitory benefits through the fluoride and anti-microbial stannous moieties respectively Dental staining, however, can occur byprecipitation of dietary chromogens onto the tooth surface by stannous ions.These studies in vitro compare the antimicrobiai profile and propensity to causetea staining of a number of stannous fluoride formulations. The formulationsused were 2 SF toothpaste products (SFl, SF2), 2 experimental SF plus stannouspyrophosphate toothpastes (SFSPI, SFSP2), a SF gel (G) and a NaF tooth-paste (C). Maximum inhibitory dilution values against a range of oral bacteriawere determined by agar dilution. Tea staining was measured spectrophotometr-ically on saliva coated clear acrylic blocks exposed to slurries of the paste or gel.All formulations showed antimicrobial activity with the order of greatest activ-ity downwards being C, SF2, SFl, SFSPI, SFSP2 and G.Tea staining at 10 ex-posures was in the following descending order of optical density SFSPI,SFSP2, G.C, SFl, SF2, water control. The antimicrobial profile of G was similarto that of SF, whereas that of the other formulations were varied but similar to adetergent profile. The difference in staining suggested considerable variation inavailability of stannous ions in the formulations. However, the propensity forstannous ions to stain must be balanced against the stain removal propensity ofthe contained detergents in the toothpaste formulations. In conclusion, the vari-ation in antimicrobial activity and more particularly staining activity of the for-mulations suggest the products will vary in activity in vivo .

Key words: stannous fluoride: oral bacteria:extrinsic stain: metal salts: toothpaste

Accepted for publication 12 March 1996

Metal salts have been investigated overmany years for their potential to inhibitplaque formation and thereby preventdental caries and chronic gingivitis (forreviews, see Addy (1986), Kornman(1986), Davey & Embery (1992)). Par-ticular interest has been shown in stan-nous and zinc salts and results haveshown, when used alone or combinedwith antimicrobial agents, inhibition ofplaque and calculus formation (for re-views, see White (1992), (1994) Cum-mins (1992)) Stannous fluoride is of in-terest, since the stannous moiety has thepotential to inhibit plaque and preventgingivitis (Beiswanger et al. 1995, Per-lich et al. 1995) and the fluoride moiety

to prevent dental caries (Tinanoff1995). Research interest in stannousfluoride has waxed and waned presum-ably in part due to the difficulties instabilizing stannous fluoride in oral hy-giene products and because of the den-tal staining side effects of this metalsalt, first reported many years ago(Muhler 1957). Stannous fluoride ora!hygiene products have thus come andgone from the market place of manycountries such that, for a period of sev-eral recent years, stannous fluoride den-tal products were available in very fewcountries for pubhc use.

The major problem with stannousfluoride is hydrolysis and precipitation

in the presence of water (Miller et al.1964). More recently, there has been aresurgence in interest in stannous fluor-ide products for which claims of sta-bility for the stannous fluoride withinthe formulation have been made (Beis-wanger et al. 1995, Perlich et al. 1995).Thus, there are available stannousfluoride gels, toothpastes and, in com-bination with amine fluoride, a mouth-wash preparation. Interest and researchin these products has centred largelyupon their potential to prevent plaqueaccumulation and inhibit gingivitis (forreviews, see White (1992, 1994)), andadditionally, the use of stannous fluor-ide products in the management of den-

82 fVade et al

tine hypersensitivity (for review, seeAddy & Dowell (!983)), Early studieson stannous fluoride solutions andtoothpaste products supported efficacyagainst plaque accumulation and thedevelopment of gingivitis which ap-peared to be via an antimicrobial action(for review; see Tinanoff (1995)). Cer-tainly, stannous fluoride appears sub-stantive to the tooth surface in a man-ner similar to chiorhexidine (Dowe!l &Addy 1984) which may explain both itsplaque inhibitory effects and the ten-dency to cause extrinsic staining. Thus,the divalent metal ion appears to bindto pellicle coated dental surfaces and iscapable of interacting with negativelycharged moieties such as bacteria and;::.-'imogenic dietary compounds (Addyet al. 1985). As formulations arise con-taining stannous fluoride it would seempertinent and reasonable to evaluatesuch formulations for activity and effi-cacy, particu!arly when the stannousfluoride is contained within complexformulations such as toothpaste. Theaim of this, the first of two studies, wasto screen a number of stannous fluorideformulations for their antimicrobial ef-fects and potential to cause dietarystaining in vitro. In particular, by com-paring the propensity to cause stainingin vitro availability of stannous fluoridewitliin a number of formulations couldbe measured and compared.

Material and Methods

The toothpaste formuiations used inthe tests in vitro together with their ac-tive ingredients and manufacturer areshown in Tab!e !.

Antimicrobial tests in vitro

The agar dilution method was used todetermine the maximum inhibitory di-!ution of the stannous fluoride and con-trol products, A panel of 20 test bac-

teria (Table 2) were chosen as represen-tative of the oral flora in health anddisease and known to show differencesin susceptibility to agents included intoothpaste formulations. Each tooth-paste or gel was double diluted in steriledistilled water and 2 ml of each dilutionimmediately mixed with 18 ml of mol-ten fastidious anaerobe agar base (LabM, Bury, UK) to give final dilutions ofl;20 to l;5l20and poured into 9 cm pe-tri dishes. The test bacteria were grownon fastidious anaerobe agar for 48 h at37°C under anaerobic conditions. Asuspension was made in phosphate buf-fered sa!ine and the turbidity adjustedto the Macfarlane 0.5 standard. Plateswere inocttlated with the bacteria by amultipoint inoculator and incubated for96 h at 37°C under anaerobic con-ditions. The results were expressed asMID (maximum inhibitory dilution)which was taken as the highest dilutionw hich yielded no growth, one colony ora fine visible haze. The experitnentswere performed in triplicate and themedian MID calculated.

Tea staining in vitro

The method is similar to that describedto study the aetiology of staining as-sociated with chiorhexidine (Prayitno etal. 1979) and later that associated withmetal salts including tin (Addy et al.1985). The substrate for the develop-ment of staining was optically clear po-lymethyl methacrylate (Perspex, ICI.Macclefield, England) cut into blocks30 mmXlO mmX5 mm to fit the speci-men chamber of a UV/visible doublebeam spectrophotometer. The dietarychromogen was a tea solution producedby boiling 8 g of branded tea in 800 mlwater for 2 min, filtering through gauzeand allowing to coo! to room tempera-ture. Groups of 3 blocks were placedinto pooled human saliva for 2 min, re-moved and washed and then p!aced for

Table I. Formulations employed in the antimicrobial and tea staining studies invitro

ProductType

toothpastetoothpastetoothpaste

toothpaste

toothpastegel

Ingredients

sodium fluoride0.4% stannous fluoride0,4% stannous fluoride1% stannous pyrophosphate0.4% stannous fluoride2% stannous pyrophosphate0.4% stannous fluoride0.4% stannous fluoride

Manufacturer

Procter & Gamble, USAProcter & Gamble, USAOral B Co., USA

Oral B Co., USA

Oral B Co., USAColgate Palmolive, USA

Textabbreviation

CSFl

SFSPl

SFSP2

SF2G

2 min into the toothpaste/gel slurries,water was used as the control. Slurrieswere produced by mixing 5 g of tooth-paste/ge! in 20 m! water and thoroughlymixing. After 2 min exposure to the re-spective slurry, specimens were re-moved, washed in water and thenplaced in standard tea solutions for 60min, removed, washed and finally theoptical density read at the lambdamaximum for tea namely 395 nm. Theregimen was repeated for 10 cycles.

Results

The MID results are shown in Table 2.All of the test pastes showed good activ-ity against the bacteria tested except Gwhich was active against 10 of the 19bacteria which grew on the test me-dium. The pastes were ranked for theiractivity against each individual test bac-teria; the mean ranking is shown inTable 2. This shows that C and SF2 dis-played the highest activity followed inrank order by SFl. SPSFl, SPSF2 andG.

Tea staining in vitro

The mean optical density readings ofthe groups of specimens at the termin-ation of the 10 cycles are shown in Fig.1. The order of staining potential is ap-parent from the bar chart with the SF2showing no more staining than achievedwith water. Staining with the C and SFlproducts was essentially similar andgreater than that observed with water.G produced more staining than the for-mer products but most staining wasseen with the SFSPl and SFSP2 formu-lations.

Discussion

The antimicrobial properties in vitro ofthe conventional fluoride toothpastewas similar to that seen in previousstudies using this methodology (Moranet al. 1988). Given the pattern of anti-microbial effect in vitro, it has beenconcluded previously that the activitywas mainly derived from the anionic de-tergent content of the formulation (Mo-ran & Addy 1984, Moran et al. 1988).Sodium lauryl sulphate (SLS) is a com-mon detergent employed in toothpasteagainst which many oral micro-organ-isms are sensitive. Interestingly, as re-ported previously, (Wade & Addy 1992)Veillonella species show reduced sensi-tivity to detergents such as SLS and this

Stannous fluoride, baeteria and staining 83

Table 2. Maximum inhibitory dilutions of toothpastes and gel against test bacteria

Species

Gram negative bacteriaAetinobaeillus actinomyeetemeomitansAetinobacillus actinomyeetemeomitansCampylobacter reetusCapynoeytophagci sputigenaCapynocytophaga speciesFusobacterium nueleatumPorphyromonas gingivalisPorpltvronwnas gingivalisFrevotella nigrescensFrevotella intermediaVeitlonella purvalaVeillonella species

Gram positive bacteriaActinomyees viscosusAetinomyees odontolytieusEubacterium timidumPeptostreptocwcus anaerobiusPeptostreptococcus microsStreptocoeciis constellatusStreptococcus mutansStreptococcus oral'ts

mean ranking

Strain no.

W1425Y4ATCC 33238NCTC 11654W287NCTC 10562ATCC 33277W1335NCTC 9336W1436NCTC 11463W1459

NCTC 1095!NCTC 9935ATCC 33093NCTC 11460W1980W362NCT C 10832NCTC 7864

C

160(1)160(1)320(1)320 (2)320 (1)640(1)

1280(1)NG

640(1)1280(1)

80(1)20(2)

320(1)320(1)640(1)320(1)320 (2)320(1)320 (1)220(1)

1.2

Maximum inhibitory dilution

SFl

160(1)80(2)

160(3)320 (2)320(1)320 (3)640(2)

NG320(3)640(2)

80(1)20(2)

160(3)160(3)320 (3)320 (1)320 (2)160(3)320(1)320(1)

2.!

SPSFI

160(1)80(2)

160(3)320 (2)160 (4)160(4)160 (4)

NG160 (4)160(4)40(4)40(1)

160 (3)80(4)

160 (4)80 (4)

160 (4)80(4)80(4)

160(4)

3.4

SPSF2

80(6)80(2)80(5)

320 (2)160 (4)160 (4)160 (4)

NG80(5)80(5)40(4)20(2)

160 (3)80(4)80(5)80(4)

160 (4)80(4)40 (5)80(5)

4.1

SF2

160(1)80 (2)

320(1)640(1)320(1)640(1)640(2)

NG640(1)640(2)

80(1)20(2)

320(1)320(1)640(1)320(1)640(1)320(1)320(1)320(1)

1.2

G

160(1)80(2)40(6)20(6)

<20 (6)40(6)80 (6)NG

<20 (6)20(6)

<20 (6)<20(6)

<20 (6)<20 (61

20(6)<20 (6)

40(6)<20 (6)<20 (6)

20 (6)

5.5

Ranking ( )

is clearly apparent from the table of re-sults. G showed the antimicrobial pro-file of stannous fluoride as has been de-termined previously (unpublished ob-servations). The remaining pastesdisplayed profiles typical of detergentssuch as SLS (Wade & Addy 1992). Thedetergent contents of these pastes areunknown but it is interesting that pasteSFl containing stannous fluoride wasless active than C, a conventional non-stannous fluoride containing paste. Thepastes containing stannous pyrophos-

2M T

phate showed further reduced activityand SPSF2 containing 2% stannouspyrophosphate was less active thanSPSFI containing 1% stannous pyro-phosphate. These results might suggestthat stannous ions are interacting withdetergents present in the formulationthus reducing the detergent's anti-microbial activity. However, the stainingdata show that the pastes containingthe stannous pyrophosphate causedmore staining than the stannous fluor-ide only pastes. It is possible, therefore.

SF1 VtaterSPSFI SPSF2 SF2 O

Products

Fig. 1. Mean optical density of product treated acrylic specimens after 10 cycles through tea.

that it is the presence of the pyrophos-phate which is reducing the antibac-terial activity of the detergents.

The agar dilution method employedhere although able to detect activity dueto stannous fluoride in the G formu-lation, which is presumably free of an-ionic detergents, was not able to detectactivity due to stannous ions in the re-maining pastes, possibly because thiswas masked by the activity of the deter-gents.

The dietary staining investigation in-dicated that at least two of the productsnamely SFl and SF2, exhibited littlepropensity to cause staining greaterthan that seen with the controls ofwater and the standard fluoride tooth-paste. Given that stannous fluoride,when available, is known to react insolution and on the surfaces to precipi-tate dietary chromogens (Addy et al.1985), these findings suggest limitedavailability of stannous fluoride fromthe latter pastes. On the other hand, themethod is not only measuring the pro-pensity of stannous fluoride to stain butof the toothpastes detergent to removestain (Addy et al. 1991). The stainmgproduced finally must reflect the netoutcome of these opposing actions. In-deed, careful balancing of stannousfluoride and detergent levels could the-oretically enable effective antiplaque

84 Wade et al.

products to be formulated but whichhave little or no staining. The G geland, more particularly, the SFSPl andSFSP2 formulations demonstrated in-creased staining in vitro supportive ofconsiderably increased availability ofthe contained stannous ions. In the caseof the pyrophosphate toothpastes pre-sumably, the stannous staining poten-tial considerably exceeded the stain re-moval action of the detergents. Thepossible action of the detergents is sup-ported also by the G data. This productstained considerably more than the twostannous fluoride products, SFl andSF2, yet contains the same concen-tration of stannous fluoride but no de-tergents.

In conclusion, the data suggest thatif stannous salts do possess plaque in-hibitory properties, the stannous fluor-ide./stannous pyrophosphate formu-lations studied here would be expectedto exhibit plaque inhibitory action overand above the other formulations testedin these experiments. Such conclusions,however, have to be tempered facedwith the known plaque inhibitory ac-tion of conventional SLS detergentbased toothpastes (Addy et al. 1983).Thus, as stated, the antimicrobial dataindicate that the statmous fluoride for-mulations have antimicrobial activity.but this is probably derived from thedetergents. The antimicrobial activitywas less than noted with a conventionalSLS product suggesting some adverseingredient interactions. The stainingdata indicates modest stannous fluorideactivity in the SFl and SF2 productsbut increased stannous activity associ-ated with the stannous fluoride/stan-nous pyrophosphate formulations andthe stannous fluoride gel.

Zusammenfassung

Studien iiber eine zinnftuoridhaltige Zahnpasteund ein zinnfluoridhahiges Gel < I). Anti-mikrobielle Eigenscfwften und Verfarbungser-zeugung in vitroZinnfiuorid (SF) in Zahnpasten hesitzt durchdas Fluorid bzw. die Zinnanteile karies- undplaquehemmende Eigenschaften. Durch Zin-nionen kann es jedoch zu Zahnverfarbungenhervorgerufen durch Ausfallungen von Nah-rungschromogenen auf die Zahnoberflachekommen. Diese in vitro Studie vergleicht heieiner Reihe von Zinnfluorld-Formulierungendie antimikrobielle Wirkung und die Nei-gung zu Teeverfarbungen. Die verwendetenFormulierungen waren zwei SF-Zahnpasta-produkte (SFl, SF2), zwei experimentelle SFplus Zinnpyrophosphat-Zahnpasten (SFSPl,SFSP2), ein SF-Gel (G) und eine NaF-Zahn-

pasta (C). Die Werte fiir die maximale inhi-bierende Verdunnung einer Reihe von oralenBakterien gegenuber wurden mit der Agar-verdunnungsmethode bestimmt. Die Teever-farbungen wurden spektroskopisch auf mitSpeichel bedeckten Acrylblocken. die den Pa-sten- oder Gelaufschlammungen ausgesetztwaren gemessen. AUe Formulierungen eineantimikrobielle Wirkung die von links nachrechts abnahm; C, SF2, SFl; SFSPl, SFSP2und G. Die Teeverfarbungen bei 20 Exposi-tionen nahmen folgendermaCen ab; SFSPl,SFSP2, G, C, SFl, SF2, Wasserkontrolle.Das antimikrobielle Profil von G war ahnlichzu dem von SF, wahrend das der anderenFormulierungen unterschiediich waren, je-doch dem Profil emes Deiergentiums ahnelte.Die unterschiede hinsichtlich der Verfarbun-gserzeugung laBt eine bedeutende Variationin der Verftigbarkeii von Zinnionen in derFormulierungen annehmen. Jedoch muB dieNeigung von Zinnionen zu Verfarbung mitder Neigung der vorhandenen Detergentienzur Entfernung von Verfarbungen ausbaian-ciert werden. Die SchluBfolgerung ist, daGdie Variation der antimikrobiellen Wirkungund lnsbesondere der Verfarbungscrzeugungder Formulierungen wahrscheinlich eine un-terschiedhche Wirkung der Produkte in vivozur Folge hat.

Resume

Etudes sur des pates dentifrices et gels auftuorure stanneux 11). Proprietes antimicro-biennes et poientiel de coloration in vitroLa presence de fluorure stanneux (SF) dansTexcipient d'une pate dentifrice peut y offrirI avantage d'effets anti-carie et d'effetsd'inhibition de la plaque, apportes respecti-vement par la fraction fluorure et par la frac-tion stanneuse, antimicrobienne. Cependant.les ions stanneux peuvent provoquer des co-lorations dentaires par precipitation deschromogenes alimentaires sur la surface den-taire. Les etudes presentees ici comparent invitro le profil amimicrobien d'un certainnombre de formuies au fluorure stanneux etleur tendance a causer I'apparition d'une co-loration par le the. Les formules utiliseesetaient 2 pates dentifrices au SF du commer-ce (SFl ei SF2), 2 pates experimentales auSF plus pyrophosphate stanneux (SFSPl,SFSP2), un gel au SF (G) et une pate denti-frice au NaF (C). Les valeurs des dilutionsinhibitrices maximales ont ete determineespour une gamme de bacteries buccales pardilution dans I'agar. La coloration par le thea ete mesaree par spectrophotometrie sur desblocs de resine acrylique transparente enduitsde salive et exposes a Sa pate ou an gel reduitsen bouillie. Les formules presentaient toutesune activite antimicrobienne; par ordre des-cendant, C, SF2, SFl , SFSPl, SFSP2 et G.La coloration par Ie the pour 20 expositionsetait. par ordre descendant de densite opti-que; SFSPl, SFSP2, G, C, SFl. SF2. contro-le a i'eau. Le profil antimicrobien de G etaitsemblable a celui du SF, tandis que celui desautres formules variait mais etait semblable

au profil d'un detergent. Les differences decoloration semblaient indiquer une variationconsiderable dans la disponibilite des ionsstanneux dans les differentes formules. Ce-pendant, il convient de faire I'equilibre entrela tendance a provoquer une coloration quepossedent ies ions stanneux et la tendance aeliminer les colorations que possedent les de-tergents contenus dans ies formules de patedentifrice. En conclusion, les variations del'activite antimicrobienne et surtout les varia-tions de l'activite de coloration dans les for-muies font supposer que l'activite des pro-duits in vivo presentera des variations.

References

Addy, M. (1986) Chiorhexidine comparedwith other locally delivered antimicrobials.A short review. Journal of Clinical Period-ontology 13. 957-964.

Addy. M. & Dowell, P. (1983) Dentme Hy-persensitivity. A Review (II). Clinical andin vitro evaluation of treatment agents.Journal of Clinical Periodontology 10, 351-363.

Addy, M., Moran, J., Griffiths, A. & Wills-Wood, N. J. (1985) Extrinsic tooth dis-colouration by metals and chiorhexidine.I; Surface protein denaturation dietaryprecipitation? Briti.^h Dental Journal 159,331-334.

Addy, M., Goodfield, S. & Harrison, A.(1991) The use of acrylic to compare theabrasivity and stain removal properties oftoothpastes. Clinical Materials 7, 2)9-225.

Addy, M., Willis, L. & Moran, J. (1983) Theeffect of toothpaste and chiorhexidinerinses on plaque accumulation during a 4day period. Journal of Clinical Periodonto-logy 10. 89-98.

Beiswanger. B. B.. Doyle. P. M., Jackson, R.D , Mallatt, M. E.. Mau, M. S., Boilmer,B. W., Crisanti. M M., Guay C B., Lan-zalaco, A. C , Lukacovic, M. F, Majeti.S. & MeClanahan, L. K. (1995) The clin-ical effect of dentifrices containing stabil-ised stannous fluoride on plaque forma-tion and gingivitisa six-month study withad libitum brushmg. Journal of ClinicalDentistry 6, 46-53.

Cummins, D. (1992) Mechanism of action ofclinically proven anti-plaque agents. In;Embery G. & Rolla, G. (eds); Clinical andbiological a.ipects of dentifrices. Oxford,UK; Oxford Medical Publications, pp.205-228.

Davey, H. & Embery, G. (1992) Metal ionsin oral hygiene products. In; Embery, G. &Rolla, G. (eds). Clinical and Biological As-peels of dentifrices. Oxford. UK; OxfordMedical Publications, pp. 165-172.

Dowell. P. & Addy, M. (1984) Dentine Hy-perseusitivity. A quantitive comparison ofthe uptake of metal salts and fluoride bydentine and hydroxyapatite. Journal ofPeriodtmtal Research 19, 530-539.

Kornman, K. S. (i9S6) Antimicrobial agents.In; Loe, H. & Kleinman, D. V (eds); Den-tal plaque control measures and oral hy-

Slartnous fluoride, bacteria and staining 85

giene practices. Oxford: IRL Press, pp.121-142.

Miller. J. T. Shannon. I.. Kiigore. W &Bookman. J. (1969) Use of water free stan-nous fluoride containing gei in the controlof dentine hypersensitivity. Journal ql Peri-odontologv 40. 490-491.

Moran. J., Addy, M. & Wade. W. G. (1988)Determination of minimum inhibitoryconcentration of commercial toothpastesusing an agar dilution method. Journal ofDentistry 16, 27-^31.

Moran. J. & Addy. M. (1984) The antibac-terial properties of some commerciallyavailable toothpastes in vitro. British Den-tat Journal 156. i75-178.

Muhler. J. C. (1957) Effect on gingiva andoccurrence of pigmentation of teeth fol-lowing the topical application of stannousfluoride or stannous chlorofluoride.Journal of Periodontology 28. 281-286

Pterlich. M. A.. Bacca. L. A.. Bollmer, B. W.Lanzalaco. A. C . McClanahan. L. K., Se-wak. L. K.. Beiswanger. B. B.. Eichold. W.A.. Hull. J. R.. Jackson. R. D. & Mau. M.S. (1995) The clinical effect of dentifricescontaining stabilised stannous fluoride onplaque formation and gingivitisand gingi-val bleeding-a six-month study. Journal ofCiinicai Dentistry 6. 54-58.

Prayitno, S.. Taylor. L.. Cadogan. S. & Addy.M.. (1979) An m vitro study of dietary fac-tors in the aetiology of tooth staining as-sociated with the use of chlorhexidine.Journal of Periodontal Research 14. 411-417.

Tinanoff. N. (1995) Progress regarding theuse of stannous fluoride in clinical den-tistry Journal of Clinical Dentistry 6. 37-40 '

Wade. W. & Addy. M. (1992) Antibacterialactivity of some triciosan containing

toothpastes and their ingredients. Journalof Periodomology 63. 280-282.

White. D. J. (1992) Tartar control dentifrices:current status and future prospects. In:Embery. G. & Rolla, G. (eds): Clinical andbiological aspects of dentifrices, Oxford,UK: Oxford Medical Publications, pp.277-292.

White. D. J. (1995) A return to stannousfluoride dentifrices. Journal of ClinicalDentistry 6, (spec, iss.), 29-36.

Address:

Martin AddyDivision of Restorative DentistryDental SchoolLower Maudlin StreetBristol BSl 2LYUK