Morphologic effects on L929 fibroblasts of titanium tetrafluoride application Bilge Hakan ~en, DDS, PhD, a Reza B. Kazemi, DMD, b and Larz S. W. Sp~ngberg, DDS, PhD, c Izmir, Turkey, and Farmington, Conn. EGE UNIVERSITY AND UNIVERSITY OF CONNECTICUT HEALTH CENTER

Objective. The aim of this study was to investigate the effect of titanium tetrafluoride solution on L929 fibroblasts by scanning electron microscopy. Titanium tetraf[uoride was then compared with sodium fluoride and acidulated phosphate fluoride. Study design. Cells were treated with fluoride solutions for 1 minute either directly, through a filter membrane with a pore size of 0.4-gm, or indirectly, through dentin disks; they were then investigated at an electron microscopic level. Results. Fluoride application on smeared dentin disks showed fewer cytotoxic effects on fibroblasts than application on nonsmeared dentin disks. Acidulated phosphate fluoride and titanium tetrafluoride appeared to be more cytotoxic than sodium fluoride. Because all fluoride solutions used in this study contained the same fluoride concentration, pH was considered to be the main factor causing the higher toxicity. Conclusion. Because these solutions demonstrated toxicity in vitro, they must be further evaluated under in vivo conditions to ascertain their clinical safety. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;86:341-6)

Dentin is a permeable hard tissue owing to its tubules, which are filled with a plasmalike fluid originating from the dental pulp. 1 Intact enamel and cementum provide dentin with significant protection from damage caused by chemicals, bacteria, and mechan- ical stimuli. When dentin is exposed as a result of grinding, root scaling, or cavity preparation, it becomes susceptible to primary or secondary caries and may become hypersensitive.

A variety of techniques and materials, such as the application of varnishes, liners, and adhesives, has been used to occlude the dentinal tubules and reduce caries formation or dentinal sensitivity. Among these proce- dures, topical fluorides, including sodium fluoride (NaF), acidulated phosphate fluoride (APF), and tin fluoride have gained special interest on account of their anticariogenic properties. 2 Fluoride treatment of dentin cavities aids in decreasing microleakage 3 and therefore in reducing the occurrence of secondary or recurrent caries. 4-5 Experiments in vitro have shown that the resis- tance of inorganic components of dentin to acidic solu- tions is increased by fluoride applications. 6-8 We also found that fluoride-treated smeared dentin had lower permeability and better resistance to citric acid applica-

aAssistant Professor, Department of Restorative Dentistry & Endodontology, School of Dentistry, Ege University, Izmir. bAssistant Professor, Department of Restorative Dentistry & Endodontology, School of Dental Medicine, University of Connecticut Health Center, Farmington. cprofessor and Head, Department of Restorative Dentistry & Endodontology, School of Dental Medicine, University of Connecticut Health Center, Farmington. Received for publication Sept. 15, 1997; returned for revision Nov. 11, 1997; accepted for publication May 11, 1998. Copyright © 1998 by Mosby, Inc. 1079-2104/98/$5.00 + 0 7/15/91733

tion. 9 In addition to commonly used fluoride agents, tita- nium tetrafluoride (TiF4) has demonstrated promising properties. Titanium tetrafluoride reduces enamel solu- bifity, inhibits artificial lesion formation, 10-~2 and forms a resistant coating on dental hard tissues. 13-14 High fluoride uptake and retention in dentin and root surfaces has been well demonstrated both in vitro and in vivo. 15-I6

A smear layer is always formed on dentinal surfaces after cavity preparation. Although removal or condi- tioning of this layer is still debatable, 17-22 it has been shown that TiF 4 demonstrates considerable interaction with both smear layer and intertubular and intratubular dentin. 23 However, there is insufficient knowledge regarding the cytotoxic effect of TiF 4. In order for a material to be used clinically, toxicity of the material must be known; therefore, the material must first be properly evaluated.

The aim of this in vitro study was to investigate, at an electron microscopic level, the effect of TiF 4 applica- tion on the morphology of L929 fibroblasts. This was tested both in the presence and in the absence of dentinal smear layer by comparing the effects of TiF 4 with those of NaF and APF solutions.

MATERIAL AND METHODS Preparation of fluoride solutions

Titanium tetrafluoride. One gram of TiF 4 was dissolved in distilled water to yield 100 mL of 1% solution of 0.323 MF (pH = 1.35).

Sodium fluoride. 1.36% solution was prepared by dissolving 1.36 g of NaF in distilled water to yield 100 mL solution of 0.323 MF (pH = 8.45).

Acidulated phosphate fluoride. 1.36 g of NaF was dissolved in approximately 70 mL of distilled water; pH was adjusted to approximately 1.35 by the addition of

34/

342 ~en, Kazemi, and Spdngberg ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY September 1998

Well

Fibrob

Culture Me

Filter

Insert with a 0.4 I~ Culture Treated Filter

Surface Treatment oi Dentin Disk

with APF, NaF or

Ti F 4

Fig 1. Schematic review of experimental device.

85% phosphoric acid. Distilled water was then added to bring the total volume to approximately 100 mL, and pH was exactly adjusted to 1.35 (0.323 MF).

Preparation of dentin disks Fifty noncarious human mandibular first or second

molars extracted for periodontal reasons were used in this study. Dentin disks with a nominal thickness of 600 gm were cut just inside the occlusal dentinoenamel junction by means of a rotating diamond saw (Isomer, Buehler, Mich.) under distilled water irrigation. Only 1 dentin disk was taken from each tooth. Each disk was ground with sandpaper (grit size no:320) on the coronal side to ensure complete removal of enamel and to reduce the thickness to 500 _+ 25 gm, by measurement to the closest 1 Bin with a digimatic micrometer (Mituitoyo, Japan).

To remove smear layer from the dentinal surfaces, the dentin disks were transferred to a flask containing 20 mL of 17% disodium ethylenediaminetetraacetic acid solution. After 3 minutes, the disks were rinsed in distilled water for 30 minutes and subsequently treated with 20 mL of 5.25% sodium hypochlorite for another 3 minutes. All dentin disks were then washed under running water for 30 minutes.

At this stage, 2 samples of treated disks were randomly examined under scanning electron microscope (SEM), and they confirmed the removal of smear layer. During this period, the remaining disks were kept in distilled water. On one half of the dentin disks, the smear layer was re-created on one side by grinding on sandpaper (grit size no:320) for 30 seconds. All disks were then packed individually and sterilized with ethylene-oxide.

Preparation of fibroblast cells L929 mouse fibroblast cells cultured in Dulbecco's

Modified Eagle's Medium (Sigma Chemical Co., St. Louis, Mo.) were used. The medium was supplemented with bovine calf serum (Gimini Bioproducts, Inc., Calabasas, Calif.), 2 mM of L-glutamine, 2.2 mg of sodium bicarbonate, 50 gg of streptomycin, and 100 IU of penicillin per mL. The cells were incubated at 37 ° C in 5% CO 2 in air. The culture medium was changed every other day and on the day before the experiments.

Fig 2. Scanning electron micrograph shows control specimen with untreated dentin disk. Cells are in different periods of growth cycle. Fine fibrillar structures around ceils are well preserved and attached to membrane.

Two-day-old or 3-day-old cultures were used for the experiments.

The experiments were performed in cluster multi- well cell culture plates, each containing 24 wells; each well had an inner diameter of 16 mm (Costar, Cambridge, Mass.). Each cluster well (Fig 1) had a permeable membrane chamber insert (Transwell, Costar) with a diameter of 6.5 mm. The bottom of the insert was at a distance of 1 mm from the bottom of the well. The membrane on the insert had a pore size of 0.4 gm.

The cells were harvested with 0.25% trypsin and suspended in the culture medium to a concentration of 50,000 cells/mL. One mL of cells was dispensed into each insert and incubated for 24 hours at 37 ° C to form a monolayer. The inserts were removed from the culture plates, and dentin disks were attached to the outside bottom of inserts with a nontoxic petroleum jelly (Vaseline, Chesebrough-Pond's Co , Greenwich, Conn.; Fig 1). In every specimen, a smear-free surface was facing the membrane at the bottom of the insert. The assemblies (chamber inserts with dentin disks) were incubated for an additional 24 hours in the culture plates; 1 mL of medium was contained in each well.

The inserts, with or without dentin disk, were placed upside down, rinsed with 3 mL of distilled water, and dried with sterile gauze. The dentin surface of each group was treated with 1 of the 3 fluoride solutions, which was applied for 1 minute (Table I). The treated surface of each disk was again washed with 3 mL of distilled water and dried with sterile gauze. One group of disks was immediately removed after fluoride application to observe the rapid effect of fluoride solutions; the remaining disks were left in place to observe the prolonged effect (Table I). The

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY ~en, Kazemi, and Spgmgberg 343 Volume 86, Number 3

Fig 3. Fibroblasts after rapid NaF treatment of smeared dentin disk. All cells have developed surface blebs but seem to retain their original shapes.

Fig 4. Cells after prolonged NaF treatment of nonsmeared dentin disk; most have been detached from filter membrane. Microvilli and ruffles of attached cells appear severely injured.

inserts were then placed back in the culture plates and incubated for 4 hours.

Control groups For each group, additional inserts having the cells on

the internal side and with or without dentin disks were prepared like the previous groups. The membranes of 3 inserts and 6 dentin disks (3 with smear layer, 3 without smear layer) attached to the inserts did not receive any treatment as positive control groups. The membranes of 9 inserts were directly subjected to the fluoride solutions (3 inserts for each solution/group) for 1 minute as negative control groups.

Scanning electron microscopic procedures At the end of the incubation period, the cells in the

inserts were washed with 5 mL of sterile saline for 30 seconds and then fixed in 5% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer for 30 minutes. After the cells were rinsed in 0.1 mol/L sodium cacodylate buffer for 30 minutes, they received a postfixation procedure with 2% osmium tetroxide in 0.1 mol/L sodium cacodylate buffer. After all specimens were rinsed twice with distilled water, 5 minutes each time, the specimens were dehydrated through ascending ethanol solutions (35%, 70%, 85%, 95%, and 100%), 5 minutes each time, and then immedi- ately dried with hexamethyldisilazane solution (Electron Microscopy Sciences, Ft. Washington, Pa.) for 5 minutes. After air-drying for 30 minutes, the membranes of the inserts were cut out with a sharp scalpel, placed on aluminum stubs, and coated with gold to a thickness of 20 nmol/L. The samples were then viewed with an SEM (JSM-35CF, Jeol, Tokyo, Japan).

Evaluation The SEM photographs were analyzed for cellular

Table I. Results with fluoride solutions on different dentin disk groups

Fluoride solutions Dentin disks Application mode* Results

NaF With smear Rapid +

NaF With smear Prolonged ++

NaF Without smear Rapid ++

NaF Without smear Prolonged +++

NaF No dentin disk'~ Rapid +

APF With smear Rapid ++

APF With smear Prolonged +++

APF Without smear Rapid +++

APF Without smear Prolonged ++++

APF No dentin disk? Rapid ++++

TiF 4 With smear Rapid ++

TiF 4 With smear Prolonged +++

TiF 4 Without smear Rapid +++

TiF 4 Without smear Prolonged ++++

TiF 4 No dentin d i sk t Rapid ++++

No applicat ions With smear

No application:) Without smear

No application:) No dentin d i sk t

*In these groups, dentin disk was either removed after fluoride application (rapid effect) or incubated with cells (prolonged effect). "Hn these groups, fluoride solutions were applied directly on filter membranes. ~Control groups.

damage. In each sample, the surface of the membrane was carefully scanned from the center to the outer side and scored according to the following scale:

(-) No reaction: no change in morphology of the cells; dense in population (several cell layers; Fig 2).

(+) Mild: presence of few cells having minor morphologic changes, such as presence of surface blebs; normal in population (at least one continuous cell layer; Figs 3 and 8).

(++) Moderate: detachment of cells from the surface, loss of microvilli, presence of collapsed cells; distur- bance in cell layer (Fig 4).

344 ~en, Kazemg and Spdmgberg ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY September 1998

Fig 5. Higher magnification view shows specimen after APF treatment through filter membrane. Cell walls are exten- sively damaged.

Fig 6. Fibroblasts after prolonged APF treatment of non- smeared dentin disk. Cells are severely damaged and have lost their microvilli and pseudopodia.

(+++) Severe: presence of fragmented cell bodies, loss of pseudopodia; destruction of cell layer (Figs 5 and 7).

(++++) Very severe: lysis of cells, loss of cell outlines; no cell layer (Figs 6 and 9).

RESULTS Findings are presented in Table I. The sections that

follow describe the results of the SEM analysis in detail.

Positive controls In all positive control specimens, with or without

dentin disks, fibroblasts appeared as flat or round mature cells depending on the stage of the growth cycle and firmly attached to the substrate. Whereas fiat cells had numerous microvilli well distributed across

Fig 7. TiF 4 t r e a t m e n t through filter membrane has totally destroyed cells, which present a fibrillar structure.

the cell surface and pseudopodia, mature cells were uniformly covered with surface nodules. Rinsing all control specimens with distilled water had not altered the cells (Fig 2).

Application of 1.36% NaF solution Rapid effect. After direct application of NaF on the

filter membrane (negative control) or a smeared dentin disk for 1 minute, the fibroblasts still appeared dense in population but with numerous surface blebs of varying size (Fig 3). With nonsmeared dentin, most cells also appeared to be undamaged. However, in some areas, the cells collapsed with pitted surfaces. The injured cells frequently formed irregular clusters.

Prolonged effect. When the cells were incubated for 4 hours after treatment of dentin disks in the presence of smear layer, the cells appeared scarce in some areas but mostly preserved their structure with microvilli and pseudopodia. In smear-free specimens, it was apparent that fibroblast monolayers were disturbed and that the cells were contracted. Most cells lost their microvilli and were detached from the substratum (Fig 4).

Application of APF solution Rapid effect. After APF application to the filter

membrane for 1 minute (negative control), there was extensive damage to the cell layer. Although clusters of collapsed cells were predominant, cells with defects on the cellular membrane were also present (Fig 5). After a rapid application of APF on smeared or nonsmeared dentin disks, undamaged fiat cells were observed in isolated areas. Damaged cells were mostly rounded, with surface blebs. Particularly in specimens with nonsmeared dentin disks, the cells were detached from the surface and formed dense clusters.

Prolonged effect. After incubation with smeared dentin disks for 4 hours, the cells were detached,

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY ~en, Kazemi, andSpdmgberg 345 Volume 86, Number 3

Fig 8. Fibroblasts after rapid TiF 4 treatment of dentin disk with smear layer. Fibrillar extensions and raffles of some cells are injured.

, O . . . . e © ,

i'r,~,~'+ ' - - + s .o;,f'";~, { 2

:1~ > L "~ 4:'

Fig 9. After prolonged TiF 4 t rea tment of nonsmeared dentin disk, cell layer was present; however, cells have lost their outlines and appear as an amorphous irregular mass.

fragmented, and spread over sparse healthy cells. Specimens with nonsmeared dentin disks demon- strated very few atrophic/scarce cell bodies. Most of the cells were heavily damaged and had non-nodular surfaces. They also had an irregular and collapsed shape (Fig 6).

Application of 1% TiF 4 Rapid effect. When 1% TiF 4 solution was applied to

the membrane (negative control), a layer of the cells was still present. However, all of the cells lost their outlines, forming clusters and being transformed into a fibrillar mass (Fig 7). The rapid application of 1% TiF 4 solution on smeared dentin disks caused different patterns of cell morphology. Although a group of healthy cells was observed in some areas, clusters of rounded and collapsed cells with surface blebs were observed in other areas (Fig 8). Cell membranes appeared disturbed. In specimens with nonsmeared dentin disks, there were healthy cells but also many other cells that were frag- mented or collapsed. In some areas, the fibroblasts were totally destroyed and fiddled or shriveled.

Prolonged effect. After 4 hours of incubation with TiF4-treated smeared dentin disks, the cells mostly lost their microvill i and pseudopodia. They also appeared to be embedded in an extracellular matrix. In specimens with nonsmeared dentin disks, cell outlines were not observed and the cells were trans- formed into islands of irregular, gelatinous mass (Fig 9).

DISCUSSION Morphologic changes such as loss of microvilli and

pseudopodia, formation of blebs on the cell membrane, alterations in cell shape, and cell detach- ment from the substrate may occur during cytotoxi-

city testing of materials. 24-26 In our study the most common observations were cell detachment from the filter membrane because of loss of cellular extensions and formation of clusters of rounded cells. It has been reported that cultured fibroblasts become rounded as the cell processes are shortened up and that blebs occur on the cell surfaces. According to A1-Nazhan and Spgmgberg, 26 bleb formation on injured cells may be due to cytoplasmic shrinkage. It was observed in our study that in NaF-treated specimens, bleb forma- tion, spongious appearance, and cluster formation were the dominant morphologic characteristics in addition to the normal appearance of healthy cells.

Prolonged incubation with APF or TiF 4 and direct application through the filter membrane showed more dramatic changes on cell morphology. In most cases, cells were seriously damaged to a degree beyond identification, which means total cell lysis. However, the rapid effect of the same acidic solu- tions through smeared dentin disks appeared to be better tolerated than the effects with treated non- smeared dentin disks. It is likely that acidic fluoride solutions were buffered by smear layer and plugs in disks with lower permeability; this mode of action may decrease the cytotoxic effect.

TiF 4 interacts with and modifies the smear layer and plugs by forming a resistant coat on dentinal walls. 23 However, the toxic effect of APF and TiF 4 was more pronounced in comparison with NaF. Considering that all fluoride solutions used in our study had the same fluoride concentration, the low pH of APF and TiF 4 might explain increased cytotox- icity. Several concentrations of NaF solutions (1% to 4%) were tolerated well by the pulp. 27-29 In contrast, application of acidic stannous fluoride solution for 5 minutes produced local necrosis, but 30-second

346 ~en, Kazemi, and Spgmgberg ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY September 1998

application of the same solution had no harmful effects.27,30

Although NaF application appeared to be well toler- ated by cultured cells in our study, its application on dentin in a previous study 9 did not reduce the perme- ability of dentin, and the NaF-treated smear layer was not resistant to citric acid challenge. Its cfinical usage may not be efficient in terms of occluding dentinal tubules. Furthermore, lowered concentrations of TiF 4 (0.1%, 0.5%) were found to be as effective as 1% TiF 4 and APF in terms of acid resistance. 9 In another study, 13 a 10-second application of 1% TiF 4 solution was demonstrated to be sufficient to deposit a substan- tial amount of fluoride ions on dentin and root surfaces. When these data are evaluated together, it may be proposed that either lowering the concentration or decreasing the application period of TiF 4 solutions may cause less cytotoxic effect.

The rapid effect of all fluoride solutions was better tolerated than the prolonged effect. Possibly, the concentration of fluoride during the 4 hours of incu- bation after fluoride application would increase in the limited amount of medium because of release from surface and tubular content. Therefore, the cytotoxic effect would be sustained on both previ- ously injured and undamaged cells that had survived the rapid application.

Although cell culture cytotoxicity testing is poten- tially useful in providing a sensitive and low-cost assay system, the results must be interpreted with care. In situ, dentinal fluid has a distal flow that may neutralize some effects of toxic materials placed on dentin surface. In addition, there is a clearing function of pulp tissue through the vascular system. Despite these limi- tations, our experiments show that TiF 4 is toxic to cells and should be further evaluated in vivo before being used in routine dental care.

REFERENCES 1. Pashley DH. Dentin permeability: Theory and practice. In:

Sp~ngberg LSW, editor. Experimental endodontics. Boca Raton: CC Press; 1990. p. 19-49.

2. Melberg JR, Ripa LW. Fluoride in preventive dentistry: theory and clinical applications. Chicago: Quintessence; 1983.

3. Barkmeier WW, Kelsey WP, Cooley RL, Berry TD. The effect of topical fluoride on microleakage. General Dentistry 1983;31:452-6.

4. Alexander WE, McDonald RE, Stookey GK. Effect of stannous fluoride on recurrent caries: results after 24 months. J Dent Res 1973;52:1147.

5. Cooley RL, Barkmeier WW. Reducing recurrent caries with topical stannous fluoride treatment of cavity preparations. General Dentistry 1979;27:30-3.

6. Selvig KA. Effect of fluoride on the acid solubility of human dentin. Arch Oral Biol 1968;13:1297-310.

7. Shannon IL, Buchanan WE Jr, Mahan CJ. In vitro treatment of human root surfaces with fluorides. J Public Health Dent 1976;36:201-6.

8. Derand T, Lodding A, Petersson LG. Effect of topical F solu- tions on caries-like lesions in root surfaces. Caries Res 1989; 23:135 -40.

9. Kazemi RB, ~en BH, Sphngberg LSW. Permeability changes of dentin treated with titanium tetrafluoride. J Dent. In press 1998.

10. Shrestha BM, Mundorff SA, Bibby BG. Enamel dissolution, I: effects of various agents and titanium tetrafluoride. J Dent Res 1972;51:1561-6.

11. Wefel JS, Harless JD. The effect of several topical fluoride agents on artificial lesion formation. J Dent Res 1982;61: 1169-71.

12. Biiyfikyilmaz T, Tangugsorn V, Ogaard B, Arends J, Ruben J, R011a G. The effect of titanium tetrafluoride (TiF4) application around orthodontic brackets. Am J Orthod Dentofac Orthop 1994;105:293-6.

13. Wefel JS, Harless JD. The effect of topical fluoride uptake and surface morphology. J Dent Res 1981 ;60:1842-8.

14. Bfiytikyilmaz T, ~en BH, Ogaard B. Retention of titanium tetra- fluoride used as a fissure sealant on human decidous molars. Acta Odontol Scand 1997;5573-8.

15. Skartveit L, Tveit AB, Totdal B, Selvig KA, Effects of TiF 4 solutions on root surfaces in vitro after different application periods. Acta Odontol Scand 1989;47:25-30.

16. Skartveit L, Tveit AB, Klinge B, Totdai B, Selvig KA. In vivo uptake and retention of fluoride after a brief application of TiF 4 to dentin. Acta Odontol Scand 1989;47:65-8.

17. Pashley DH. Smear layer: physiological considerations. Operative Dentistry Supplement 1984;3:13-29.

18. Van Meerbeek B, Perdig~o J, Lambrechts P, Vanherle G. The clinical performance of adhesives. J Dent 1998;26:1-20.

19. Hayakawa T, Nemoto K, Horie K. Adhesion of composite to polished dentin retaining its smear layer. Dental Materials 1995; 11:218-22.

20. Inokoshi S, Burrow MF, Takatsu T. In vivo adhesive interface between resin and dentin. Operative Dentistry 1995;20:204-10.

21. Konuma M. Effects of smeared layer on luting strength of dental cements. Journal of Stomatological Society, Japan 1996;63:231- 42.

22. Love RM, Chandler NE Jenkinson HE Penetration of smeared or nonsmeared dentine by Streptococcus gordonii. International Endodontics Journal 1996;29:2-12.

23. ~en BH, Bfiytikyilmaz T. The effect of 4% titanium tetrafluoride solution on root canal walls: a preliminary investigation. Journal of Endodontics 1998;24:239-43.

24. Meryon S, Stephen P, Brown R. A comparison of the in vitro cytotoxicity of two glass ionomer cements. J Dent Res 1983;62:769-73.

25. Zislis T, Martin SA, Cerbas E, Heath JR III, Mansfield JL, Hollinger JO. A scanning electron microscopic study of in vitro toxicity of ethylene-oxide-sterilized bone repair materials. Journal of Oral Implantology 1989;15:41-6.

26. A1-Nazhan S, Spfingberg L. Morphological cell changes due to chemical toxicity of a dental material: an electron microscopic study on human periodontal ligament fibroblasts and L929 cells. Journal of Endodoutics 1990; 16:129-34.

27. Bannstr6m M, Nyborg H. Pulp reaction to fluoride solution applied to deep cavities: an experimental histological study. J Dent Res 1971;50:1548-52.

28 Furseth R, Mj6r IA. Pulp studies after 2 per cent sodium fluoride treatment of experimentally prepared cavities. Oral Surg Oral Med Oral Pathol 1973;36:109-14.

29. Walton RE, Leonard LA, Sharawy M, Gangarosa LP. Effects on pulp and dentin of iontophoresis of sodium fluoride on exposed roots in dogs. Oral Surg Oral Med Oral Patho] 1979;48:545-57.

30. Massler M, Evans JA. Absence of pulpal response to stannous fluoride applied to freshly cut dentin. J Dent Res.1967;46:1469.

Reprint requests: Reza Kazemi, DMD Department of Restorative Dentistry and Endodontology University of Connecticut Health Center School of Dental Medicine 263 Farmington Ave. Farmington, CT 06030-1715