ORIGINAL ARTICLE

Efficacy of TiF4 and NaF varnish and solution:a randomized in situ study on enamel erosive–abrasive wear

Flávia Mauad Levy & Daniela Rios &Marília A. R. Buzalaf & Ana Carolina Magalhães

Received: 13 January 2013 /Accepted: 21 August 2013 /Published online: 1 September 2013# Springer-Verlag Berlin Heidelberg 2013

AbstractObjectives This in situ/ex vivo study analysed the anti-erosive/abrasive effect of TiF4 and NaF varnish and solutionon enamel wear.Materials and methods Twelve subjects took part in this studywhich was performed in three periods (phases) with the dura-tion of 5 days each. Each two human enamel specimens persubject were pretreated with experimental NaF varnish orsolution (phase A), experimental-TiF4 varnish or solution(phase B) and placebo varnish or untreated control (phaseC). The specimens were worn in palatal appliances; oneenamel specimen, from each treatment, was subjected to ero-sion (ERO; cola soft drink, 4×90 s/day), and the other spec-imen was subjected to erosion plus abrasion (ERO + ABR;tooth brushing, 2×10 s/day). The tooth wear was quantifiedby a contact profilometer (micrometre) and analysed usingtwo-way repeated measures ANOVA and Bonferroni’s test(n =12 subjects, p <0.05).Results All fluoride varnishes and solutions reduced theenamel wear (around 25 %) significantly compared to thecontrol and placebo varnish. There were no significant differ-ences among the fluoride formulations and between the con-ditions ERO and ERO + ABR.Conclusions Therefore, it can be concluded that TiF4 has thesame protective potential as NaF formulations to reduce hu-man enamel wear under this experimental in situ model.

Clinical significance In vitro studies have indicated a betteranti-erosive/abrasive effect of TiF4 compared to NaF varnish.The present in situ study does not support the previous find-ings. Therefore, any of the tested professional fluoride var-nishes in principle could be able to partially reduce enamelwear.

Keywords Enamel . Fluoride . Titanium fluoride . Toothabrasion . Tooth erosion

Introduction

Conventional fluorides whose beneficial effect against cariesis already established [1] have been tested for prevention orcontrol of dental erosion [2, 3]. The potential of conventionalfluorides, such as NaF, to prevent erosive demineralisation ismainly related to the formation of a calcium fluoride (CaF2)-like layer [4, 5]. This layer is assumed to behave as a physicalbarrier hampering the contact of the acid with the underlyingenamel or to act as a mineral reservoir, which is attacked bythe erosive challenge, reducing the demineralisation. After-ward, calcium and fluoride released might increase the satu-ration level with respect to the dental hard tissue in the liquidadjacent to the enamel surface, thus promoting mineralprecipitation.

As the anti-erosive effect of conventional fluorides requiresa very intensive fluoridation regime [6, 7], current studieshave focused on fluoride compounds that might have higherefficacy. In this context, compounds containing polyvalentmetal ions such as stannous fluoride or titanium tetrafluoride(TiF4) were tested.

The potential of TiF4 solution to prevent tooth erosivedemineralisation has been investigated since 1997 [8]. Itsprotective effect may be related to the increased fluorideuptake due to the low pH. Additionally, titanium may also

F. M. Levy :M. A. R. Buzalaf :A. C. Magalhães (*)Department of Biological Sciences, University of São Paulo,Al. Dr. Octávio Pinheiro Brisolla, 9-75, Bauru, SP 17012-901, Brazile-mail: acm@fob.usp.br

D. RiosDepartment of Pediatric Dentistry, Orthodontics and Public Health,Bauru School of Dentistry, University of São Paulo,Al. Dr. Octávio Pinheiro Brisolla, 9-75, Bauru, SP 17012-901, Brazil

Clin Oral Invest (2014) 18:1097–1102DOI 10.1007/s00784-013-1096-y

contribute to the formation of an acid-resistant surface coatingdue to the formation of a new compound (hydrated hydrogentitanium phosphate) or organometallic complexes that mightact as diffusion barrier on enamel [3, 9].

In situ studies have shown that TiF4 solution is as effectiveas SnF2 or AmF solution in the prevention of erosion orcombined erosion/abrasion [10, 11], while another in situstudy did not show any protective effect against enamel wear[12]. The efficacy of TiF4 is highly dependent on the pH of theagent. TiF4 at its native pH (pH 1.2) could significantlyreduced enamel erosion but not at a pH buffered to 2.1 [13]or 3.5 [14]. However, the low pH of TiF4 solution does notallow self-application as mouth rinsing by the patient.

In respect to the limitation of the solution, TiF4 was incor-porated in an experimental varnish, indicated for the profes-sional application. In vitro studies have shown higher anti-erosive and anti-abrasive effects of the varnish compared tothe solution [15, 16]. Moreover, the varnish is considered safefor the patient in terms of ingestion and toxicity, as it is directlyapplied on the tooth surface compared to a mouthrinse solution.However, further properly designed in situ or clinical studiesare recommended to better understand the relative differencesin performance of the various fluoride formulations.

Thus, the aim of this in situ study was to analyse the effectof a single application of TiF4 varnish/solution, compared toNaF varnish/solution, to protect human enamel against ero-sion with or without abrasion. The null hypothesis tested wasthat there is no significant difference between NaF and TiF4formulations on the prevention of enamel erosive and erosive–abrasive wear.

Materials and methods

Selection of the subjects and ethical aspects

Twelve healthy adult subjects (eleven females, one male; aged23 to 35 years) who fulfilled the inclusion criteria (physiolog-ical salivary flow rates: stimulated, >1 ml/min and resting,>0.25 ml/min; good oral health: no frank cavities or signifi-cant gingivitis/periodontitis) without violating the exclusioncriteria (systemic illness, pregnancy or breastfeeding, use offixed or removable orthodontic appliances, use of fluoridemouthrinse or professional fluoride application in the last2 months or hyposalivation) were enrolled. Sample size calcu-lation was based on previous in situ study [12]. A sample sizeof 12 subjects was calculated considering α-error level of 5 %and β-error level of 30 % and a relevant difference betweenTiF4 solution and control of 13 % (www.ddsresearch.com).

The research was ethically conducted in accordance withthe Declaration of Helsinki. The ethical approval for the studyinvolving human subjects was granted by the local ethicscommittee (no. 083/2008). The study was planned as a

prospective, single-centre, double-blind and six-cell study incrossover and split-mouth design with three periods/phases(two treatments and two conditions per phase) of 5 days each,with a washout time of 5 days between the periods/phases [12,17]. The subjects received written instructions and a scheduleand were extensively trained for all procedures. Informedconsent was obtained from all subjects prior to the study.Figure 1 shows the study design. The description of the studydesign followed the guideline of the CONSORT statement.

Preparation of specimens and the distributionto the experimental groups

One hundred and forty-four human enamel specimens (4×4×3 mm) were prepared from the labial/lingual surface of thecrowns (two specimens/crown; total, 72 teeth). The impactedthird molars were collected and then stored in 2 % bufferedformaldehyde solution (pH 7.0) at 4 °C for 30 days [12].Thereafter, the specimens were cut in the middle of the sur-faces using an ISOMET low-speed saw (Buehler Ltd., LakeBluff, IL, USA) with two diamond discs (Extec Corp., En-field, CT, USA) separated by a 4-mm-thick spacer. The spec-imens’ surfaces were ground flat using water-cooled siliconcarbide discs (320-, 600- and 1200-grade papers; Buehler,Lake Bluff, IL, USA) and polished using felt paper wet withdiamond spray (1 μm; Buehler). The polishing removedaround 200 μm of enamel. The enamel surface was checkedby using the software of the profilometer.

After polishing, the specimens were cleaned in an ultrason-ic device with deionized water for 2 min. Prior to the exper-iment, two layers of nail varnish were applied on two thirds ofthe enamel surface to maintain reference surfaces for deter-mining tooth wear after the experiment, leaving 1.5 mm ofcentral exposed area. The specimens were maintained in100 % humidity until the in situ experiment was conducted.

Each two enamel specimens per subject were randomlyallocated to each treatment (total, 24 specimens/treatment):experimental NaF varnish and NaF solution (both were locat-ed in phase A, FGM-Dentscare/Brazil, 2.45 % F, pH 5.0),experimental TiF4 varnish and TiF4 solution (both in phase B,FGM-Dentscare/Brazil, 2.45 % F, pH 1.0) and placebo var-nish and untreated control (both in phase C, FGM-Dentscare/Brazil, pH 5.0). The composition of the varnishes was previ-ously reported [15].

In each treatment, half of the specimens were subdividedinto erosion only (ERO, n =12) and erosion plus abrasion(ERO + ABR, n =12). For each phase (A, B and C), 48 newenamel specimens were used.

Enamel specimens were fixed with wax into the recesses ofeach individual acrylic palatal appliance. The treatment (solu-tion or varnish) was randomly divided in rows (at least 1 cmapart from each other), and the conditions (ERO or ERO +ABR) were randomly divided in lines (at least 1 cm apart) for

1098 Clin Oral Invest (2014) 18:1097–1102

each volunteer. This split-mouth palatal appliance design hasbeen previously used for testing the effect of TiF4 on enamelcaries lesion [17].

Treatment and in situ experiment

Five days prior to and throughout the entire experiment, thesubjects brushed their teeth with experimental fluoride-freetoothpaste (Crest, Procter & Gamble, USA) to allow theprotective effect of fluoride to be only due to the solution/varnish treatments. The subjects wore the appliance for 2 h to

allow the formation of a salivary pellicle on the enamelsurface. After the 2-h lead-in period, the treatments wereperformed only once at the beginning of each experimentalphase.

The fluoride solutions were applied on two enamel speci-mens (one row) using a microbrush for 1 min ex vivo. Theexcess of the solution was gently removed with a cotton swab[15, 16, 18, 19]. The fluoride varnishes were applied on thesurface of the two other enamel specimens (the other row)using a microbrush ex vivo. During 6 h, the subjects were notallowed to remove the appliance from the mouth. Thereafter,

Selection of 12 study subjects

Information about the study and written informed consent

Collection of saliva (Salivary flow rate and pH) - Inclusion of subjects (n=12)

Impressions and preparation of the intraoral appliances for each study phase

5 days washout period before starting the study

1st study period: Phase A (n=4, subjects 1-4), phase B (n=4, subjects 5-8) and phase C (n=4, subjects 9-12) (5 days each phase)

5 days washout period

2nd study period: Phase C (n=4, subjects 1-4), phase A (n=4, subjects 5-8) and phase B (n=4, subjects 9-12) (5 days each phase)

5 days washout period

3rd study period: Phase B (n=4, subjects 1-4), phase C (n=4, subjects 5-8) and phase A (n=4, subjects 9-12) (5 days each phase)

Profilometric assessment and statistical analysis (n=12)

Fig. 1 Flowchart of the study

Clin Oral Invest (2014) 18:1097–1102 1099

the varnishes were removed with a scalpel and a cotton swabsoaked in acetone diluted in water (1:1) ex vivo. The varnisheswere removed to allow the chemical rather than mechanicaleffect of the tested products. The nail varnishwas reapplied afterthis procedure to protect the control areas again [15, 16, 20].

At the next day, the erosive and abrasive regimens started.The erosive challenges were performed four times daily (morn-ing, midday, afternoon and evening at least 3 h apart). Forerosion, the subjects were instructed to immerse the appliancecontaining all treated enamel specimens in a cup with 150ml ofcola drink (pH 2.6, 0.32 ppm F, Coca-Cola Company) at roomtemperature for 90 s. Immediately after erosion, the applianceswere washed in tap water and reinserted into the mouth.

Two times a day, after the first and last erosive challenges, thesubjects performed toothbrush abrasion on one line (one enamelspecimen for each varnish and solution treatment). Each speci-menwas brushed ex vivowith an electrical toothbrush (Colgate®MotionsMulti-action) using one drop (around 35 μl) of fluoride-free dentifrice slurry (Experimental Crest, Procter & Gamble,USA; ratio dentifrice: water=1:3, pH 6.8) for 10 s (166 oscilla-tions per second, 1.5 °N) [16]. The erosive and abrasive chal-lenges were repeated daily for 5 days. The subjects were previ-ously trained by the researchers to perform the toothbrushing.

The erosive and abrasive protocols were applied according tothe Guideline for Erosion Studies to appropriately simulate theclinical condition [21]. The appliances were worn day and night(20 h/day) and were stored in humidity during the main mealsand oral hygiene procedures (four times daily, 1 h each, total of4 h). The subjects were advised not to eat or to drink while theappliances were into the mouth. A minimum of 30 min elapsedbetween individual oral hygiene and the experiment.

Wear analysis

Enamel wear (micrometer) was defined as outcome parameter.After the end of each experimental period/phase, the enamelspecimens were removed from the appliances. The nail var-nish on the reference surfaces was carefully removed withacetone/water-soaked cotton swab [15, 16, 20]. The speci-mens were stored in 100 % humidity until the analysis. Allspecimens were analysed at the end of the study. The wear wasquantitatively determined by a contact profilometer (HommelTester T1000, VS, Schwenningen, Germany). The diamondstylus (2 μm in diameter; angulation, 90°) moved 2.5 mmfrom the first reference across the exposed area onto the otherreference area. Four profile measurements were randomlyperformed in the centre of each specimen. The vertical dis-tance between the midpoints of regression lines on the refer-ences and experimental areas was defined as tooth wear usingthe software of the device. The values were averaged(micrometre) and submitted to statistical analysis. The detec-tion limit for this contact profilometer is about 0.5 μm ofenamel wear. The standard deviation of repeated analysis

(the reproducibility of the method) of a given enamel samplefrom the control group, without removing it from the holder,was 0.2 μm (mean wear of 2.5 μm).

Statistical analysis

The assumptions of equality of variances and normal distri-bution of errors were checked for all the variables tested usingthe Bartlett and Kolmogorov–Smirnov tests (GraphPad Instatfor Windows version 4.0, San Diego, CA, USA), respectively.Since the assumptions were satisfied, two-way repeated mea-sures, ANOVA and Bonferroni post hoc test, were used(GraphPad Prism 4 version 4.0 for Windows, GraphPad Soft-ware, San Diego, CA, USA). Thereby, the different treatments(fluoride) were considered as dependent and the conditions(ERO and ERO+ABR) as independent variables and viceversa. The subjects were considered as statistical unit (n =12), and the significance level was set at 5 %.

Results

All participants completed the study, and all specimens could bemeasured profilometrically. No specimens were lost and the nailvarnish was still protecting the control areas duringwhole in situphase. No participants reported adverse events or side effects.

Two-way repeated measure ANOVA revealed no signifi-cant difference between the conditions ERO and ERO + ABR(p >0.05). In respect to the treatments, all fluoride varnishesand solutions significantly reduced (around 25 %) the enamelwear when compared to the untreated control and placebovarnish (p <0.001), which in turn did not significantly differfrom each other. There were no significant differences amongthe fluoride formulations against enamel erosion and erosionplus abrasion as well (Table 1). There was no interactionbetween the factors (p >0.05).

Discussion

The null hypothesis tested was accepted. In the present study,all fluoridated varnishes and solutions were able to significantlyreduce enamel erosion and erosion plus abrasion compared toplacebo varnish and control. Differently from previous resultsobtained from in vitro studies [15, 16], TiF4 has the sameprotective potential as NaF to reduce human enamel wear insitu. Considering the same fluoride salt, there was also nosignificant difference between varnish and solution, showingthat the distinct contact times between the vehicles (solution,1 min vs. varnish, 6 h) seem not to be an important factor in thepresent model. It is important to discuss that no cross-effectbetween solution and varnish treatments could be expected byusing the present experimental design (split mouth appliance).

1100 Clin Oral Invest (2014) 18:1097–1102

Attin et al. [22] have shown that fluoride acquisition could notbe detected in specimens located at a distance of 1 cm from thespecimens treated with fluoride varnish.

Topical fluoridation induces the formation of a protectivelayer on dental hard tissue, which is composed of CaF2-likelayer (in case of conventional fluorides as NaF) or of metal-rich surface precipitates (in case of TiF4). From the results ofthe present study, it can be speculated that the CaF2-like layerproduced by NaF was as effective as the metal-rich surfaceprecipitates created by the application of TiF4 formulations insitu. CaF2-globules behave as a physical barrier inhibiting thecontact of the acid with enamel and/or acting as a fluoridereservoir. On the other hand, TiF4-treated enamel samplespresented a visible coating on the surface rich in fluorideand titanium as confirmed, using SEM and microprobe anal-ysis, in a previous study from our group [15]. The effect ofTiF4 is related to both fluoride (as it happens for NaF) andtitanium [23]. It is hypothesized that titanium might complexwith phosphate groups (e.g. hydrated hydrogen titanium phos-phate) producing an acid-resistant surface coating [8, 9].

Some factors have shown to influence the efficacy offluoride products against demineralisation such as fluorideconcentration, pH, and frequency of application [4]. In ourstudy, the formulations presented the same fluoride concen-tration; however, NaF presented a higher pH (5.0) comparedto TiF4 (1.0). From this view, we would expect a better effectof TiF4 compared to NaF not only due to the titanium but alsoto the low pH that could increase the enamel fluoride uptake.

Contrasting the expectations, there were no differencesbetween the fluoride salts and the vehicles. The present resultsare in disagreement with an in situ study, in which 4 % TiF4solution was unable to reduce human permanent and decidu-ous enamel erosive wear [12], and from some in vitro studiesin that 4 % TiF4 solution could not reduce bovine enamelerosive wear as well [15, 20]. However, in the cited works, thespecimens were not covered by salivary pellicle before thetreatments. We speculate that this factor could explain thedifferences among the studies.

We allowed 2 h of pellicle formation according to previousstudies about the formation and composition of the salivarypellicle [24, 25]. The presence of human saliva in situ allowsthe formation of salivary pellicle, which in turn might have a

significant impact on the reaction between fluoride and toothsurface. There are few studies dealing with the effect of fluorideon pellicle-covered tooth. Wiegand et al. [26] showed that theefficacy of the tetrafluoride solutions was positively influencedby the presence of the salivary pellicle in vitro. On the otherhand, two studies performed by Hove et al. [27, 28] pointed outthat TiF4 solution provides protection against acid attack, re-gardless of the presence of the pellicle. Therefore, there is stillsome uncertainty if salivary pellicle modulates the anti-erosiveeffect of fluoride, especially in case of TiF4, which may be alsoinfluenced by the time of application, fluoride concentration aswell as to the following erosive challenges.

The human saliva might also allow better stability to theCaF2 or metal-rich layer (in case of TiF4 solution applied onlyfor 1 min) on enamel compared to F-treated enamel samplesexposed only to the artificial saliva in vitro [15, 20]. It isalready known that the stability of the CaF2-like layer on toothin vitro is substantially reduced compared to in situ condition.Human salivary proteins and phosphate may stabilize thefluoride precipitation on enamel [5].

Another important point that might explain the differentresults is the type of substrate (bovine vs. human enamel).According to Hove et al. [28], TiF4 solution (0.5 M F) mightbetter react with bovine compared to human enamel. To ourknowledge, there are no other studies comparing the anti-erosive effect of fluoride on bovine and human enamel direct-ly. Our studies do not allow such comparison, but from theprevious in vitro studies using bovine enamel, at least in thecase of 4 % TiF4 solution, it is hypothesized that this formu-lation might react better with human than bovine enamel,refuting previous findings [28]. However, this hypothesisshould be better addressed in further studies.

In the present study, the ERO + ABR condition showedsimilar enamel wear mean values compared to the ERO only.The presence of human saliva might also have reduced theprogression of tooth wear when erosion was associated toabrasion. Additionally, this result may be a consequence ofthe experimental protocol, in which the abrasion procedure(2×10 s/day) was less aggressive than the erosive challenges(4×90 s/day) [29]. Using this protocol, the toothbrushingabrasion could not have adversely affected the progressionof tooth erosive wear differently from the results found by

Table 1 Mean enamel wear (micrometre) ± SD (95 % confidence limit) for the different treatments and enamel conditions

NaF varnisha NaF solutiona TiF4 varnisha TiF4 solution

a Placebo Varnishb Controlb

EROA 1.1±0.5 1.3±0.4 1.2±0.5 1.2±0.7 1.8±0.8 1.8±0.8

(0.74–1.42) (1.03–1.57) (0.87–1.51) (0.83–1.63) (1.31–2.30) (1.33–2.30)

ERO + ABRA 1.5±0.6 1.6±0.6 1.2±0.5 1.5±0.7 2.1±0.6 2.2±0.8

(1.09–1.82) (1.19–1.95) (0.91–1.58) (1.02–1.88) (1.70–2.48) (1.72–2.70)

Similar upper case letters indicate no difference between the enamel conditions. Distinct lower case letters indicate significant differences among thefluoride treatments and the control

Clin Oral Invest (2014) 18:1097–1102 1101

Wiegand et al. [11] in that abrasion was performed 2×30 s/day. The erosive and abrasive protocols applied in the presentstudy follows the guideline for erosion research published in aspecial issue of Caries Research [21, 29]. The cola drink waschosen as one of the most consumed soft drink by the popu-lation and has been shown to be highly erosive [12, 20].

Under the conditions of the present study, it can be con-cluded that TiF4 varnish has similar protective potential as thesolution and NaF formulations to reduce human enamel ero-sion with or without abrasion in situ. Future in situ studiesmust test the durability of the effect of fluoride varnishes ontooth wear using longer experimental periods [3], which inturn shall allow a satisfactory subjects’ compliance. Otherimportant point that should be further addressed is the clinicalrelevance of the results (25 % reduction by the application ofprofessional fluoride products) compared to other preventivemeasures, as daily application of fluoridemouthrinse. It wouldbe interesting to test the independent effect of professional andhome-care fluoride applications as well as the combination ofboth methods.

Acknowledgments This study was supported by The São Paulo Re-search Foundation - FAPESP (Proc. 2008/07105-4 for the last author,2008/03727-0 for the first author). The funders had no role in the studydesign, data collection and analysis, decision to publish, or preparation ofthe manuscript.

Conflict of interest The authors declare that they have no conflict ofinterest.

References

1. ten Cate JM (1997) Review on fluoride, with special emphasis oncalcium fluoride mechanisms in caries prevention. Eur J Oral Sci105:461–465

2. Wiegand A, Attin T (2003) Influence of fluoride on the prevention oferosive lesions—a review. Oral Health Prev Dent 1:245–253

3. Magalhães AC, Wiegand A, Rios D, Buzalaf MA, Lussi A (2011)Fluoride in dental erosion. Monogr Oral Sci 22:158–170

4. Saxegaard E, Rolla G (1988) Fluoride acquisition on and in humanenamel during topical application in vitro. Scand J Dent Res 96:523–535

5. Ganss C, Schlueter N, Klimek J (2007) Retention of KOH-solublefluoride on enamel and dentine under erosive conditions. A compar-ison of in vitro and in situ results. Arch Oral Biol 52:9–14

6. Ganss C, Klimek J, Schaffer U, Spall T (2001) Effectiveness of twofluoridation measures on erosion progression in human enamel anddentine in vitro. Caries Res 35:325–330

7. Ganss C, Klimek J, Brune V, Schürmann A (2004) Effects of twofluoridation measures on erosion progression in human enamel anddentine in situ. Caries Res 38:561–566

8. Büyükyilmaz T, Øgaard B, Rølla G (1997) The resistance of titaniumtetrafluoride-treated human enamel to strong hydrochloric acid. Eur JOral Sci 105:473–477

9. Ribeiro CC, Gibson I, Barbosa MA (2006) The uptake of titaniumions by hydroxiapatite particles—structural changes and possiblemechanisms. Biomaterials 27:1749–1761

10. Hove LH, Holme B, Young A, Tveit AB (2008) The protective effectof TiF4, SnF2 and NaF against erosion-like lesions in situ. Caries Res42:68–72

11. Wiegand A, Hiestand B, Sener B, Magalhães AC, Roos M, Attin T(2010) Effect of TiF(4), ZrF(4), HfF(4) and AmF on erosion anderosion/abrasion of enamel and dentin in situ. ArchOral Biol 55:223–228

12. Magalhães AC, Rios D, Honório HM, Jorge AM Jr, Delbem AC,Buzalaf MA (2008) Effect of 4 % titanium tetrafluoride solution ondental erosion by a soft drink: an in situ/ex vivo study. Arch Oral Biol53:399–404

13. Hove LH, Holme B, Stenhagen KR, Tveit AB (2011) Protectiveeffect of TiF(4) solutions with different concentrations and pH ondevelopment of erosion-like lesions. Caries Res 45:64–68

14. Wiegand A, Waldheim E, Sener B, Magalhães AC, Attin T (2009)Comparison of the effects of TiF4 andNaF solutions at pH 1.2 and 3.5on enamel erosion in vitro. Caries Res 43:269–277

15. MagalhãesAC, KatoMT, Rios D,WiegandA,Attin T, BuzalafMAR(2008) The effect of an experimental 4 % TiF4 varnish compared toNaF varnishes and 4 % TiF4 solution on dental erosion in vitro.Caries Res 42:269–274

16. Levy FM, Magalhães AC, Gomes MF, Comar LP, Rios D, BuzalafMAR (2012) The erosion and abrasion-inhibiting effect of TiF4 andNaF varnishes and solutions on enamel in vitro. Int J Paediatr Dent22:11–16

17. Comar LP, Wiegand A, Moron BM, Rios D, Buzalaf MA, BuchallaW,Magalhães AC (2012) In situ effect of sodium fluoride or titaniumtetrafluoride varnish and solution on carious demineralization ofenamel. Eur J Oral Sci 120:342–348

18. Tezel H, Ergucu Z, Onal B (2002) Effects of topical fluoride agentson artificial enamel lesion formation in vitro. Quintessence Int 33:347–352

19. van RijkomH, Ruben J, Vieira A, HuysmansMC, Truin GJ,Mulder J(2003) Erosion-inhibiting effect of sodium fluoride and titaniumtetrafluoride treatment in vitro. Eur J Oral Sci 111:253–257

20. Magalhães AC, Stancari FH, Rios D, Buzalaf MAR (2007) Effect ofan experimental 4 % titanium tetrafluoride varnish on dental erosionby a soft drink. J Dent 35:858–861

21. Shellis RP, Ganss C, Ren Y, Zero DT, Lussi A (2011) Methodologyand models in erosion research: discussion and conclusions. CariesRes 45:69–77

22. Attin T, Lennon AM, Yakin M, Becker K, Buchalla W, Attin R,Wiegand A (2007) Deposition of fluoride on enamel surfaces re-leased from varnishes is limited to vicinity of fluoridation site. ClinOral Investig 11:83–88

23. Wei SH, Soboroff DM, Wefel JS (1976) Effects of titanium tetraflu-oride on human enamel. J Dent Res 155:426–431

24. Siqueira WL, Zhang W, Helmerhorst EJ, Gygi SP, Oppenheim FG(2007) Identification of protein components in in vivo human ac-quired enamel pellicle using LC-ESI-MS/MS. J Proteome Res 6:2152–2160

25. Siqueira WL, Oppenheim FG (2009) Small molecular weightproteins/peptides present in the in vivo formed human acquiredenamel pellicle. Arch Oral Biol 54:437–444

26. Wiegand A, Meier W, Sutter E, Magalhães AC, Becker K, Roos M,Attin T (2008) Protective effect of different tetrafluorides on erosionof pellicle-free and pellicle-covered enamel and dentine. Caries Res42:247–254

27. Hove LH, Holme B, Young A, Tveit AB (2007) The erosion-inhibiting effect of TiF4, SnF2, and NaF solutions on pellicle-covered enamel in vitro. Acta Odontol Scand 65:259–264

28. Hove LH, Young A, Tveit AB (2007) An in vitro study on the effectof TiF(4) treatment against erosion by hydrochloric acid on pellicle-covered enamel. Caries Res 41:80–84

29. Wiegand A, Attin T (2011) Design of erosion/abrasion studies-insights and rational concepts. Caries Res 45:53–59

1102 Clin Oral Invest (2014) 18:1097–1102