You are on page 1of 6

Human pulp responses to in-office tooth bleaching

Carlos Alberto de Souza Costa,a Heraldo Riehl,b João Fernando Kina,c


Nancy Tomoko Sacono,d and Josimeri Hebling,d Araraquara, Brazil
ARARAQUARA SCHOOL OF DENTISTRY, SÃO PAULO STATE UNIVERSITY

Objective. To evaluate and compare the responses of human incisor and premolar pulps after bleaching.
Study design. A bleaching agent with 38% hydrogen peroxide (H2O2) was applied on the buccal surface of 10 sound
lower teeth (G1: 6 premolars; G2: 4 incisors) for 45 minutes. Three premolars and 3 incisors that received only rubber/
pumice prophylaxis were used as control groups G3 and G4, respectively. Two days after the bleaching procedure,
the teeth were extracted and processed for histologic evaluation.
Results. Only in G2 (4 incisors) were any changes in the pulp detected. In the coronal pulp there was a large zone of
coagulation necrosis. The radicular pulp showed mild inflammatory changes manifested as an accumulation of
mononuclear cells around congested and dilated blood vessels. No pulpal damage was seen in either of the control
groups (G3 and G4) or in group G1.
Conclusion. Bleaching with 38% H2O2 for 45 minutes causes irreversible pulp damage in lower incisors but not in
premolars. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2010;109:e59-e64)

Hydrogen peroxide (H2O2) is a thermally unstable ox- in their anterior teeth.10 However, although several in
ygen-derived free radical frequently found within the vitro studies have evaluated the cytotoxicity of bleaching
cells as the result of a series of intracellular reactions agents to culture of cells,8,9 few data are available con-
that occur specifically in the mitochondria.1 High con- cerning the effects of bleaching agents on human pulps.
centrations of this chemical agent, which present oxi- Therefore, the aim of the present study was to evaluate
dative power,2-4 have been used to treat discolored and compare the responses of pulps of anterior (incisors)
teeth.5 The oxidative reactions and consequent cell and posterior (premolars) sound human teeth submitted to
damage caused by free radicals are the main mecha- in-office bleaching using a gel with 38% H2O2.
nisms responsible for the toxicity of peroxide-contain-
ing compounds.6 Although it is known that free radicals MATERIALS AND METHODS
are capable of degrading complex organic molecules Sixteen caries-free human teeth scheduled to be ex-
that are responsible for tooth coloration,4 the exact tracted for orthodontic reasons were selected from
mechanisms by which the teeth are bleached are not young patients. The mean age of the patients was 16.2
completely understood. It has been reported that the years. The parents/guardians as well as the volunteers,
low molecular weight of H2O2 molecules makes them after reading and receiving all necessary explanations
capable of diffusing across enamel and dentin to reach including the experimental rationale, the clinical pro-
the pulpal space.1 Consequently, H2O2 and its degra- cedures, and possible risks, were asked to sign a con-
dation products, which play a role in the tooth bleach- sent form explaining the research protocol, which was
ing, may cause damage to the pulp cells, especially to previously approved by the Ethics Committee.
odontoblasts that underlie dentin.7 It has been reported Sixteen sound teeth were selected for the study and
that ⬃70% of patients submitted to bleaching have divided into 4 groups: bleached premolars (G1; n ⫽ 6),
complained about postoperative sensitivity, particularly bleached incisors (G2; n ⫽ 4), nonbleached premolars
(G3; n ⫽ 3), and nonbleached incisors (G4; n ⫽ 3). The
latter 2 groups served as control groups and received
Supported in part by the Fundação de Amparo à Pesquisa do Estado only rubber/pumice prophylaxis.
de São Paulo—FAPESP (grant nos. 2007/50646-3 and 2008/05890-6)
The teeth were cleaned by rubber/pumice prophy-
and Conselho Nacional de Desenvolvimento Científico e Tecnológico—
CNPq (grant no. 301029/2007-5). laxis, thoroughly washed, and dried with an oil-free
a
Department of Physiology and Pathology. air stream. Then, for groups G1 and G2, a light-cured
b
Private practice. resin-based gingival barrier (Opal Dam; Ultradent
c
Department of Restorative Dentistry. Products, South Jordan, UT) was used to protect the
d
Department of Orthodontics and Pediatric Dentistry.
soft tissues circumjacent to teeth being whitened and
1079-2104/$ - see front matter to protect the adjacent teeth. A 38% H2O2 bleaching
© 2010 Mosby, Inc. All rights reserved. gel (Opalescence Xtra Boost; Ultradent) was handled
doi:10.1016/j.tripleo.2009.12.002 according to the manufacturer’s instructions and was

e59
OOOOE
e60 de Souza Costa et al. April 2010

Table I. Inflammatory cell response Table IV. Absolute frequency (n) observed for each
Score Characterization histopathologic event according to the groups
0 None or a few scattered inflammatory cells present in the Score
pulp area corresponding to the buccal surface of the Histopathologic event Group 0 1 2 3
tooth in which the bleaching gel was applied
Inflammatory cell response G1 6 0 0 0
1 Slight inflammatory cell infiltrate with
G2 0 1 3 0
polymorphonuclear or mononuclear leukocytes
G3 3 0 0 0
2 Moderate inflammatory cell infiltrate
G4 3 0 0 0
3 Severe inflammatory cell infiltrate
Tissue disorganization G1 6 0 0 0
G2 0 0 1 3
Table II. Tissue disorganization
G3 3 0 0 0
Score Characterization G4 3 0 0 0
0 Normal tissue Reactionary dentin formation G1 6 0 0 0
1 Odontoblastic layer disorganized but central pulp normal G2 0 0 1 3
2 Total disorganization of the pulp tissue morphology G3 3 0 0 0
3 Pulp necrosis associated or not with dystrophic calcification G4 3 0 0 0
G1, Bleached premolars (n ⫽ 6); G2, bleached incisors (n ⫽ 4); G3,
Table III. Reactionary dentin formation nonbleached premolars (n ⫽ 3); G4, nonbleached incisors (n ⫽ 3).
Score Characterization
0 Absence
1 Modest hard tissue deposition beneath the buccal surface of
the tooth in which the bleaching gel was applied descriptive analysis according to the criteria presented
2 Moderate hard tissue deposition beneath the buccal surface
in Tables I-III: inflammatory cell response, pulp tissue
of the tooth in which the bleaching gel was applied
3 Intense hard tissue deposition in the pulp tissue disorganization, and reactionary dentin formation.
Using a light microscope (Diastar; Cambridge Instru-
ments, Buffalo, NY) adapted to a video camera (DXC-
107A/107AP; Sony Electronics, Tokyo, Japan), 3 linear
measures from the enamel/dentin junction to the pulp
applied to the buccal surface of the teeth in such a way chamber were carried out to determine the dentin thick-
that a uniform 1-mm layer of material was formed and ness (DT). The video images were loaded into a com-
left undisturbed for 15 min. The bleaching gel was puter and processed using standard software (Mocha;
removed from the tooth surface using a saliva ejector Jondel Scientific, San Rafael, CA). The methodology
with high-power suction, and the bleaching cycle was used in the present study to measure the DT was similar
repeated 2 more times, totaling 3 applications of 15 to that previously used to determine the remaining
minutes each. After the third application, the bleaching dentin thickness between the cavity floor and the sub-
gel was removed from the enamel surface and the tooth jacent pulp tissue.11,12 The 3 readings were averaged to
was copiously rinsed with an air/water spray and gently obtain the mean DT for each tooth. The data regarding
dried with sterile gauze. the DT were submitted to nonparametric Kruskal-Wal-
Two days after the bleaching procedure, the teeth lis test complemented by Mann-Whitney tests at a
were extracted under local anesthesia. The roots were significance level of 5% (SPSS, Chicago, IL).
immediately sectioned midway between the cement/
enamel junction and the root apex with a high-speed RESULTS
handpiece under water spray. The teeth were stored for The scores for each criterion determined by the his-
48 hours in formalin fixative solution at pH 7.2, decal- tologic assessment of the specimens according to
cified in buffered Morse solution (equal volumes of groups are shown in Table IV. The individual and
50% acid formic and 20% sodium citrate), under agi- median DT values are presented in Table V. Regarding
tation, cleared in xylol, vacuum infiltrated with wax the DT, a significant statistical difference was observed
paraffin, and finally embedded in paraffin. Six-micro- between incisors (G1 and G3) and premolars (G2 and
meter-thick serial sections were cut (820; Spencer Mi- G4), with incisors presenting a smaller DT than pre-
crotome, Carson, CA), mounted on glass slides, and molars (P ⬍ .05). No statistical difference was ob-
stained with hematoxylin and eosin (H/E) and Masson served when groups G1 (bleached premolars) and G3
trichrome. Based upon previous in vivo studies,11,12 all (nonbleached premolars) were compared (P ⬎ .05).
sections were evaluated by a calibrated examiner The same lack of difference was seen between G2
blinded to the groups. Using a light microscope (62774; (bleached incisors) and G4 (nonbleached incisors).
Carl Zeiss, Oberköchen, Germany) the following his- The analysis of the radiographs which were taken
topathologic events were evaluated and classified by a immediately before the extractions of all of the teeth
OOOOE
Volume 109, Number 4 de Souza Costa et al. e61

Table V. Dentin thickness (mm) for bleached and nonbleached (control) human premolars and incisors
Group*
Specimen (tooth) G1 G2 G3 G4
1 2.99 1.90 2.80 1.69
2 3.21 1.73 3.22 2.02
3 3.10 1.79 3.02 1.80
4 2.98 1.86 — —
5 3.10 — — —
6 3.24 — — —
Mean (SD)† 3.10 (0.11) 1.82 (0.08) 3.01 (0.21) 1.84 (0.17)
Median (P25-P75)‡ 3.10 (2.99-3.22)a 1.83 (1.74-189)b 3.02 (2.80-3.18)a 1.80 (1.69-1.94)b
*Group designations as in Table IV.
†Each value represents the mean of 3 measurements taken in the same specimen/tooth.
‡Medians identified with the same letter represent groups that do not differ statistically (Mann-Whitney: P ⬎ .05).

used in the present in vivo study demonstrated no blast layer, cell-free zone, cell-rich zone, and normal cen-
periapical pathology. The volunteers who had their tral part of the pulp tissue (Fig. 2). The medians of the
inferior incisors bleached (G2) reported postoperative dentin thickness for G3 and G4 were 3.02 mm (mean ⫾
pain. Therefore, based upon the IRB recommendations, SD ⫽ 3.01 ⫾ 0.21) and 1.80 mm (mean ⫾ SD ⫽ 1.84 ⫾
all teeth were extracted. Conversely, the volunteers 0.17 mm), respectively.
who had their premolars bleached (G1) did not report
any particular postoperative sensitivity. However, to DISCUSSION
compare the pulp responses occurring in premolars and According to Wataha et al.,13 the risk caused by
incisors, all premolars also were extracted 2 days after dental materials to the dentin-pulp complex depends on
being bleached. the capability of the components of these products to
Overall, the histologic features showed that the pulp diffuse across enamel and dentin to reach the pulp
tissue observed in G1 (bleached premolars) and both tissue. Several studies have demonstrated the diffusion
control groups, G3 (nonbleached premolars) and G4 of H2O2 through enamel and dentin,8,14 which is facil-
(nonbleached incisors), were quite similar. The pulp itated by its low molecular weight as well as its capac-
responses in G2 (bleached incisors) were different from ity to denature tissue proteins.15 In addition, it has been
those observed in G1, G2, and G3. In G1, all premolars reported that the diffusion of H2O2 through enamel and
presented pulp tissue with normal histologic character- dentin increases as the concentration of this unstable
istics, as observed in G3 and G4. Consequently, lack of free radical in the bleaching gel also increases.6 In the
inflammatory response and no tissue disorganization present in vivo study, the authors did not quantify or
was observed for all specimens (Fig. 1, A and B). The identify the products released from the 38% H2O2
median of the dentin thickness for this experimental bleaching gel applied on the enamel in the pulp tissue.
group was 3.10 mm (mean ⫾ SD ⫽ 3.10 ⫾ 0.11 mm). However, it was demonstrated that the coronal pulp of
In G2, the coronary pulp tissue of the 3 bleached most (3 out of 4) of the bleached incisors exhibited
incisors exhibited a wide zone of coagulation necrosis coagulation necrosis (Fig. 1, C and D). In this experi-
(Fig. 1, C and D). Deposition of reactionary dentin was mental group, the mean enamel-dentin junction–pulp
seen in the root pulp of these incisors, which showed a distance was 1.82 ⫾ 0.08 mm. On the other hand, in the
moderate inflammatory response mediated by mononu- control group G4, in which the incisors were not
clear cells among dilated and congested blood vessels bleached, the pulp tissue exhibited normal histologic
(Fig. 1, E and F). In only 1 incisor, intense deposition characteristics (Fig. 2, C-E). Therefore, it may be sug-
of reactionary dentin in the coronal and root pulp tissue gested that toxic components released from the bleach-
was observed. In that tooth, the root pulp exhibited mild ing agent were capable of diffusing through the thin
inflammatory response characterized by a few inflam- enamel and dentin present in the inferior incisors to
matory mononuclear cells. The median of the dentin cause notable damage to the pulp tissue.
thickness for this experimental group was 1.83 mm It has been demonstrated that the application of
(mean ⫾ SD ⫽ 1.82 ⫾ 0.08 mm). In the nonbleached bleaching agents on enamel increases the porosity of
control groups G3 and G4, all premolars and incisors this highly mineralized dental tissue owing to the dis-
exhibited normal pulp tissue. The pulp-dentin complex ruption of matrix protein, which causes loss of struc-
exhibited tubular dentin and predentin, intact odonto- tural components by free radical oxidation.16 In addi-
OOOOE
e62 de Souza Costa et al. April 2010

Fig. 1. Pulp-dentin complex of sound lower human premolar (A and B) and incisors (C-F) submitted to in-office bleaching. A,
Note that both pulp horns of the premolar present normal histologic characteristics. H/E, ⫻32. B, High magnification of the pulpal
horn in A related to the tooth surface in which the bleaching gel was applied. Observe the tubular primary dentin, the defined
odontoblast layer (horizontal arrow), cell-free zone (vertical arrow), and the cell-rich zone (oblique arrow). No inflammatory
response, necrosis, or pulpal calcification can be seen. H/E, ⫻125. C, Dentin-pulp complex of a bleached lower incisor (tooth #3).
Observe the thickness of the dentin (mean 1.793 ␮m) below the buccal surface where the bleaching gel was applied. The pulp
horn exhibits a large zone of coagulation necrosis (N). H/E, ⫻32. D, High magnification of C. Note the transition border (arrows)
between the tubular dentin (TD) and the necrotic pulp tissue (N). H/E, ⫻125. E, Root pulp tissue of a bleached lower incisor which
exhibits notable deposition of reactionary dentin matrix. H/E, ⫻125. F, Detail of E. Note the tubular reactionary dentin matrix
(TRDM), which is poorly calcified, the defined predentin (P), and the remaining pulp tissue (PT) with some mononuclear cells
and a few congested blood vessels (arrows). H/E, ⫻250.

tion, the contact time of the bleaching agent with the most frequent adverse effect claimed by the patients
dental tissue influences the transenamel and transden- submitted to in-office vital tooth bleaching.10 There-
tinal diffusion of bleaching components to reach the fore, it may be speculated that the pain claimed by the
pulpal space.3 The increase in hard dental tissue per- volunteers in the present in vivo study after bleaching
meability may also play a role in postoperative sensi- the incisors was caused, at least in part, by the increase
tivity or pain after bleaching therapies, which is the of the permeability in the enamel and dentin. However,
OOOOE
Volume 109, Number 4 de Souza Costa et al. e63

Fig. 2. Control groups. Dentin-pulp complex of unbleached lower sound premolar (A and B) and incisor (C-E). A, The coronal
pulps tissue exhibits normal histologic features. Masson trichrome, ⫻32. B, High magnification of the area indicated in A. Note
the pulp tissue with all defined histological zones. Masson trichrome, ⫻250. C, General view of the coronal pulp tissue. Masson
trichrome, ⫻32. D, High magnification of C. No pulp necrosis, inflammatory pulpal response, or significant change is observed
in this specific connective tissue. Masson trichrome, ⫻64. E, Detail of the coronal pulp tissue in D. Note the continuous
odontoblast layer (arrows) and the subjacent pulp tissue with pulp cells and blood vessels. Masson trichrome, ⫻125.

the diffusion of H2O2 to the pulp tissue may also be defense system of the pulp cells is activated, releasing
involved in postoperative sensitivity or pain caused by several endogenous antioxidant agents, such as peroxi-
the toxic effects of this chemical agent to the pulp dases and catalases, which promote an enzymatic deg-
cells.17 It is known that H2O2 and its degradation prod- radation of H2O2 to avoid excessive tissue damage.24
ucts, such as hydroxyl (OH⫺) ions, are reactive species However, it seems that the outward dentin fluid move-
derived from oxygen, which may cause mutagenesis, ment, the complex network of lymphatic vessels
carcinogenesis, cell membrane damage by lipid peroxi- present in the pulp tissue, as well as the protective role
dation, and protein fragmentation.18,19 Therefore, the of the defense system of the pulp were not enough to
increase in the exogenous levels of these highly reac- prevent coronal pulp necrosis when the bleaching agent
tive free radicals1,18 in contact with cells, as occurs with 38% H2O2 was applied on the buccal surface of
during tooth bleaching, may result in cell death and lower sound incisors. Perhaps, the different pulpal re-
reduced cell proliferation.20-23 These data may explain sponses observed in premolars (Fig. 1, A and B) and
the necrosis occurring in the pulp tissue after bleaching incisors (Fig. 1, C-F) occurred because of the variable
the incisors as well as the postoperative pain reported thickness of enamel and dentin. It has been reported
by the patients. However, future in vivo studies are that the dentin thickness plays an important role in the
needed to clarify these histologic and clinical observa- protection of the pulp tissue against toxic products
tions. released from dental materials.12,14 In the present in
Vital teeth present dentinal fluid flow produced by vivo study, the decalcification of the teeth during the
intrapulpal pressure and cytoplasmatic prolongations of laboratory process resulted in complete enamel loss.
odontoblasts as well as other intratubular components, Consequently, only the dentin thickness was micro-
which may impede the diffusion of the bleaching gel scopically measured. The bleached premolars (G1) and
components through the dentinal tubules. In addition, incisors (G2) presented mean DTs of 3.10 ⫾ 0.11 mm
the pulp tissue presents lymphatic drainage which is and 1.82 ⫾ 0.08 mm, respectively. Therefore, it may be
capable of moving away toxic products that reach this suggested that the protective role of dentin against
connective tissue. Furthermore, owing to the oxidative diffusion of toxic components leached from the bleach-
stress generated by the presence of free radicals, the ing agent with 38% H2O2 used in the present study
OOOOE
e64 de Souza Costa et al. April 2010

increased as a function of the thickness of this tubular 8. Benetti AR, Valera MC, Mancini MNG, Miranda CB, Baldicci I.
hard tissue. Considering the tooth bleaching procedures In vitro penetration of bleaching agents into the pulp chamber.
Int Endod J 2004;37:120-4.
commonly used in esthetic dentistry, it may also be 9. Camargo SEA, Valera MC, Camargo CHR, Mancini MNG,
suggested that the intensity of pulp responses is in- Menezes MM. Penetration of 38% hydrogen peroxide into the
versely related to the thickness of the enamel. Because pulp chamber in bovine and human teeth submitted to office
the enamel is thicker in premolars than in incisors, it bleach technique. J Endod 2007;33:1074-7.
may be speculated in the present investigation that the 10. Buchalla W, Attin T. External bleaching therapy with activation
by heat, light or laser—a systematic review. Dent Mater
thickness of enamel plays also an important role in 2007;23:586-96.
preventing H2O2 and other products leached by the 11. Costa CAS, Teixeira HM, Nascimento ABL, Hebling J. Biocom-
bleaching gel from reaching the pulp tissue. However, patibility of resin-based materials applied as liners in deep cav-
this hypothesis needs further investigations. No pulp ities prepared in human teeth. J Biomed Mater Res B Appl
damage in bleached premolars was reported by Rob- Biomater 2007;81:175-84.
12. de Souza Costa CA, Hebling J, Randall RC. Human pulp re-
ertson and Melfi.25 Those authors demonstrated that sponse to resin cements used to bond inlay restorations. Dent
disastrous pulp response in bleached premolars was Mater 2006;22:954-62.
observed only when the bleaching agent was heated. 13. Wataha JC, Hanks CT, Strawn SE, Fat JC. Cytotoxicity of
It is known that the continuous deposition of second- components of resins and other dental restorative materials.
ary and intratubular dentin over time increases the J Oral Rehabil 1994;21:453-62.
14. Gökay O, Müjdeci A, Algin E. In vitro peroxide penetration into
thickness of this hard tissue and reduces the inner pulp chamber from newer bleaching products. Int Endod J
diameter of the dentinal tubules. Both physiologic 2005;38:516-20.
mechanisms, which decrease the dentin permeability, 15. Goldstein CE, Goldstein RE, Feinman RA, Garber DA. Bleach-
may impede the diffusion of bleaching agent compo- ing vital teeth: state of the art. Quintessence Int 1989;20:729-37.
nents across dentin to reach the pulp tissue. Therefore, 16. Kwon YH, Huo MS, Kim KH, Kim SK, Kim YJ. Effects of
hydrogen peroxide on the light reflectance and morphology of
further in vivo studies are needed to evaluate if similar bovine enamel. J Oral Rehabil 2002;29:473-77.
pulpal damage, as demonstrated in the present investi- 17. Haywood VB. Current status of nightguard vital bleaching. Com-
gation for sound bleached incisors of young patients, pend Contin Educ Dent 2000;28:S10-7.
might also take place in teeth of older patients submit- 18. Martindale JL, Holbrook NJ. Cellular response to oxidative stress:
ted to in-office bleaching. signaling for suicide and survival. J Cell Physiol 2002;192:1-15.
19. Sies H. Strategies of antioxidant defense. Eur J Biochem
1993;215:213-9.
CONCLUSION 20. Allen RG, Tresini M. Oxidative stress and gene regulation. Free
Radic Biol Med 2000;28:463-99.
Based on the methodology used in this in vivo study, 21. Slater AF, Stefan C, Nobel I, van den Dobbelsteen DJ, Orrenius
it may be concluded that tooth bleaching with 38% S. Signalling mechanisms and oxidative stress in apoptosis.
H2O2 for 45 minutes causes irreversible pulp damage in Toxicol Lett 1995;82:149-53.
lower incisors but not in premolars. 22. Trindade FZ, Ribeiro APD, Sacono NT, Oliveira CF, Lessa FCR,
Hebling J, Costa CAS. Trans-enamel and trans-dentinal cytotoxic
effects of a 35% H2O2 bleaching agent on an odontoblast cell line
REFERENCES after consecutive applications. Int Endod J 2009;42:516-24.
1. Shackelford RE, Kaufmann WK, Paules RS. Oxidative stress and 23. Dias Ribeiro AP, Sacono NT, Lessa FCR, Nogueira I, Coldebella
cell cycle checkpoint function. Free Radic Biol Med 2000;28: CR, Hebling J, de Souza Costa CA. Cytotoxic effect of a 35%
1387-404. hydrogen peroxide bleaching gel on odontoblast-like MDPC-23
2. Arwill T, Myrberg N, Soremark R. Penetration of radioactive iso- cells. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
topes through enamel and dentin. II. Transfer of 22Na in fresh and 2009;108:458-64.
chemically treated dental tissues. Odontol Revy 1969;20:47-54. 24. Esposito P, Varvara G, Murmura G, Terlizzi A, Caputi S. Ability
3. Hanks CT, Fat JC, Wataha JC, Corcoran JF. Cytotoxicity and of healthy and inflamed human dental pulp to reduce hydrogen
dentin permeability of carbamide peroxide and hydrogen-perox- peroxide. Eur J Oral Sci 2003;111:454-6.
ide vital bleaching materials, in vitro. J Dent Res 1993;72:931-8. 25. Robertson WD, Melfi RC. Pulp response to vital bleaching
4. Tredwin CJ, Naik S, Lewis NJ, Scully C. Hydrogen peroxide procedures. J Endod 1908;6:645-9.
tooth-whitening (bleaching) products: review of adverse effects
and safety issues. Br Dent J 2006;200:371-6. Reprint requests:
5. Zantner C, Beheim-Schwarzbach N, Neumann K, Kielbassa AM. Prof. Dr. Carlos Alberto de Souza Costa
Surface microhardness of enamel after different home bleaching Departamento de Fisiologia e Patologia
procedures. Dent Mater 2007;2:243-50. Faculdade de Odontologia de Araraquara—UNESP
6. Li Y. Biological properties of peroxide-containing tooth whiten- Rua Humaitá, 1680 —Centro
ers. Food Chem Toxicol 1996;34:887-904. CEP: 14801-903
7. Goldberg M, Smith AJ. Cells and extracellular matrices of dentin Araraquara, SP
and pulp: a biological basis for repair and tissue engineering. Crit Brasil
Rev Oral Biol Med 2004;15:13-27. casouzac@foar.unesp.br