Comparison of human pulp response to total-etch and self-etchbonding agentsSeema Nayyar, BDS,a Sanjay Tewari, BDS, MDS,b and Brijbala Arora, MBBS, MD,c Haryana,IndiaGOVERNMENT DENTAL COLLEGE

Objective. To compare the pulp reactions in terms of pulp inflammation, leakage involving bacteria and reactionarydentin formation to total-etch adhesive, Prime & Bond NT and self-etch adhesive, Xeno III in 2 mm deep class Vcavities prepared in human teeth.Study design. 2 mm deep class V cavities were prepared on buccal surface of 40 human premolars. The teeth wererandomly divided into two groups based on the application of a total-etch adhesive system, Prime and Bond NT afteracid etching or a self-etching adhesive system, Xeno III. The teeth were extracted after 7 and 30 days and preparedaccording to routine histological techniques.Results. All teeth were asymptomatic. Both the adhesives showed moderate to severe inflammatory reactions in fewcases initially at seven days, but the response was generally minimal at 30 days. Leakage involving bacteria wasdetected in 42.5% of restorations and 40% of the specimens were associated with various grades of pulpinflammation. The cavity remaining dentin thickness influenced the grade of inflammatory activity. A significantcorrelation was detected between inflammatory cell response and the presence of bacteria. Pulp response to self-etching adhesive, Xeno III was minimally different from that of the total-etch adhesive, Prime & Bond NT (p � .05).Conclusions. A self-etching adhesive system, Xeno III showed similar pulp response and leakage involving bacteria toa total-etch adhesive, Prime & Bond NT. Both adhesives showed an acceptable biological compatibility with human

pulps at 30 days. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:e45-e52)

Dentin adhesive systems have been developed rapidlyover the past 30 years. The introduction of the total-etch technique of both enamel and dentin with phos-phoric acid has revolutionized restorative dentistry byproviding a strong interfacial bond between restorationsand dental tissue.1 However, these multistep total-etchbonding agents have the drawbacks of technique sen-sitivity, risk of overetching dentin, and their inability toascertain optimum dentinal surface conditions forbonding, as well as being time-consuming.2

An alternative approach is based on the use ofnonrinse acidic monomers that simultaneously condi-tion and prime dentin, the so-called “self-etching” ad-hesives.2 Regarding user-friendliness and techniquesensitivity, this approach seems clinically most prom-ising; however, these materials have been reported tobe extremely hydrophilic and may act as a semiperme-

aMDS student, Department of Operative Dentistry and Endodontics,Government Dental College, PGIMS, Rohtak.bProfessor and Head, Department of Operative Dentistry and End-odontics, and Principal, Government Dental College, PGIMS,Rohtak.cProfessor and Head, Department of Pathology, PGIMS, Rohtak.Received for publication Oct 16, 2006; returned for revision Jan 31,2007; accepted for publication Feb 12, 2007.1079-2104/$ - see front matter© 2007 Mosby, Inc. All rights reserved.

doi:10.1016/j.tripleo.2007.02.005

able membrane, which might result in degradation ofbond in due course of time.2

Although high initial bond strength values have beenreported with these bonding agents,2 most of the mic-roleakage studies have concluded that composite resto-rations exhibit microleakage in vitro, but to a differentdegree.3 However, microleakage to tracer moleculesdoes not necessarily imply that leakage involving bac-teria will also occur. Although the space appears to besufficiently large to allow microbial spread with allmaterials, other factors such as availability of nutrientsand the antibacterial properties of the material influencethe leakage involving bacteria, particularly in vivo.3

These observations suggest the importance of in vivoleakage studies involving bacteria in assessing mic-roleakage.

Since the demonstration of the effect of bacteria atthe tooth/filling interface by Brännström and Nyborg,4

there has been mounting evidence to support the viewthat bacterial contamination is the predominant sourceof pulp inflammation.5,6 Severe forms of inflammatoryactivity can often progress to total pulpal necrosis andperiapical lesion development with local bone destruc-tion.7 On the contrary, some studies have not dem-onstrated a good correlation between the presence ofbacteria at the tooth/filling interface and pulp reac-

tions.8-10 Such observations suggest that whereas

e45

OOOOEe46 Nayyar et al. August 2007

bacteria are an important cause of pulp inflammationbeneath cavities filled with different experimentalmaterials, the question concerning the possible influ-ence of the interactions between the variables of cavitypreparation and restoration, such as cavity preparationtrauma,11 chemical activity of restorative materi-als,8,12,13 etching treatment,14 and patient factors, in-cluding aging and treatment history remains, as well asbacterial irritation.15

Surveys indicate that secondary (recurrent) caries isthe most frequent reason for replacing amalgam andresin composite restorations, followed by mechanicalfailure and complications such as hypersensitivity.16

These surveys continue to provide evidence that a highproportion of restored teeth show symptoms that re-quire pulpectomy or extraction, suggesting that restor-ative treatments are not fully optimized.17

Thus, the purpose of this study was to compare thepulp reactions in terms of pulp inflammation, leakageinvolving bacteria, and reactionary dentin formation tothe total-etch adhesive, Prime & Bond NT, and theself-etch adhesive, Xeno III, in 2-mm deep class Vcavities prepared in human teeth.

MATERIAL AND METHODSForty human premolar teeth free of fillings, cervical

abrasion, and caries and scheduled to be removed fororthodontic purposes were selected from patients rang-ing in age from 11 to 20 years old after the patient’sinformed consent. The parents of all the volunteerswere asked to read and sign the consent form explain-ing the research protocol. The patients as well as theirparents or responsible persons received thorough ex-planation concerning the experimental rationale, clini-cal procedures, and possible risks.

Before the start of the operative protocol, each toothwas radiographically examined to exclude presence ofcaries, cervical abrasions, or periapical pathologies.Electric pulp testing was performed to check the pulpalstatus of all the teeth. After applications of rubber dam,all teeth were polished with a rubber cup and prophy-laxis paste at low speed and the surrounding field wascleaned with 70% isopropyl alcohol. Class V cavitieswere made on the cervical third of buccal surfaces ofthe clinical crowns of the premolars using a 012straight-fissure diamond bur (Dentsply De Trey, Kon-stanz, Germany) in an air and water-cooled high-speedhand piece. The bur was replaced after every fifthpreparation to avoid excessive heating. Buccal cavitiesexhibiting defined dimensions of 3.0-mm length,2.0-mm depth, and 1.5-mm width were prepared. Teethwere randomly assigned to 2 experimental groups ac-

cording to different bonding agent application.

Group A (Prime & Bond NT)Cavities were etched with 38% orthophosphoric acid

(DPI tooth-conditioning gel, Dental Products of IndiaLtd, Mumbai, India) for 15 seconds with the help of abrush. Cavities were rinsed with distilled water for 20seconds with a combined air and water spray and air-dried. Then dentin surfaces were rewetted with the helpof a cotton pellet dipped in water and excess removedby touching it on a dry gauze piece leaving it visiblymoist. One coat of bonding agent, Prime & Bond NT(Dentsply) was applied with a bristle brush and leftundisturbed for 20 seconds, thinned for 5 seconds byblowing gently with air from a dental syringe, and thenlight-cured for 10 seconds at an intensity of 600 mW/cm2

(Spectrum 800, Dentsply).

Group B (Xeno III)Equal amounts of liquid A and liquid B of Xeno III

bonding system (Dentsply) was mixed for 5 seconds ona clean mixing well. A generous amount of Xeno IIIwas applied to the cavity to thoroughly wet all cavitysurfaces and was left undisturbed for 20 seconds. Thenit was uniformly spread using a gentle stream of dry airfor 2 seconds followed by light curing for 10 seconds.

The cavities were bulk filled with hybrid composite,Spectrum TPH (Dentsply) and light cured for 40 sec-onds.

Extraction was scheduled after the test period of 7 or30 days. Patients’ symptoms (if any) were recorded.Electric pulp testing was performed as before and theteeth were extracted under local anesthesia; 2% xylo-caine with adrenaline (1:200,000) was used as a localanesthetic. Immediately, their root apices were cut offwith a high-speed hand piece under intensive waterspray to allow penetration of the fixative solution. Theteeth were immersed in freshly prepared 10% neutralbuffered formalin for 48 hours, decalcified in 5% nitricacid, dehydrated, and embedded in paraffin wax.

Next, 5-�m-thick sections were cut with a mic-rotome and stained with hematoxylin and eosin (H&E)for routine histological evaluation and with Brown andBrenn modification of Gram’s staining for detectingmicroorganisms. All serial sections were evaluatedblindly by 2 investigators independently for 4 histo-logic features: inflammatory pulp response, reactionarydentin formation, bacteria, and the remaining dentinthickness (RDT). The intensity of pulp response wasevaluated by light microscope (Olympus, OlympusManufacturing Co Ltd, Tokyo, Japan). Pulpal inflam-mation was categorized from “none” to “severe” inH&E-stained sections according to the published crite-ria.18,19 The criteria are given in Table I. The presenceof reactionary dentin and leakage involving bacteria

were assessed according to the criteria given in Tables

OOOOEVolume 104, Number 2 Nayyar et al. e47

II and III as reported by Hebling and others.13 Theremaining dentin thickness between the cavity floor andpulp tissue was measured for each specimen by imageanalysis. Images provided by a charge device videocamera (Olympus BX51, Olympus Manufacturing CoLtd) coupled with a microscope (Olympus, OlympusManufacturing Co Ltd) at a magnification of �40 werestored on a host computer based on a Pentium IV

Table I. Inflammatory pulp responseInflammatory pulpresponse Characterization

Score 0 None- Normal pulpal structureScore 1 Mild- An increased number of cells in the so-

called cell-free zone and in the adjacentpulpal tissue. The majority of these cellshave morphologic characteristics offibroblasts and undifferentiated cells, but afew inflammatory cells are also involved.Increased number of capillaries are notedand a few extravasated red blood cells maybe found. The response is localized to theaffected dentinal tubules.

Score 2 Moderate- More cells in areas subjacent to theaffected dentin than are associated with aslight reaction. Neutrophilic andmononuclear leukocytes invade theodontoblast–predentin area in proportionsthat depend on whether the reaction ispredominantly acute or chronic.Odontoblasts can not be identified in theirnormal pseudo stratified appearance, butindividual odontoblasts may be discerned.Sometimes odontoblastic nuclei can beobserved in the dentinal tubules. Increasednumber of capillaries are found in theinfiltrated tissue and at its border. Thepulpal reaction is localized. The width ofthe predentin may or may not deviate fromnormal, depending on the duration of thereaction.

Score 3 Severe- Marked cellular infiltration, includingabscess formation is found.Polymorphonuclear and mononuclearleukocytes predominate in the affected area,and the response is well delimited. Theodontoblastic layer cannot be identified as amorphologic entity or as individual cellsshortly after the response is established. Nopredentin is formed, and within days theexisting predentin apparently mineralizesand cannot be distinguished from theadjacent dentin. Odontoblastic nuclei canbe seen in the dentinal tubules provided thechanges do not represent a long-standingreaction. Numerous blood vessels are foundin the tissue surrounding the intensecellular infiltration.

processor with operator system Microsoft Windows XP

through a digital frame grabber and processing wasdone by image analysis software (Micro Image Lite 4.0,Olympus). RDT was measured as the amount of re-maining dentin along dentinal tubules at the thinnestdimension. Sections with the smallest RDT were taken

Table II. Reactionary dentin formationReactionary dentinformation Characterization

Score 0 AbsenceScore 1 Modest hard tissue deposition beneath the

axial wallScore 2 Moderate hard tissue deposition beneath the

axial wallScore 3 Intense hard tissue deposition beneath the

axial wall

Table III. Leakage involving bacteriaBacterialpresence Characterization

Score 0 AbsenceScore 1 Presence of stained bacteria along the cavity lateral

wallsScore 2 Presence of stained bacteria along the cavity lateral

and axial wallsScore 3 Presence of stained bacteria along the cavity walls and

within the cut dentinal tubules

Table IV. Results of histopathological findings7 DaysPrime

& Bond NT Xeno III

30 DaysPrime

& Bond NT Xeno III

Inflammatory pulpresponse

None 4 6 7 7Slight 3 2 3 3Moderate 1 1 0 0Severe 2 1 0 0

Leakage involvingbacteria

None 6 6 5 6Slight 2 3 3 3Moderate 2 1 1 1Severe 0 0 1 0

Reactionary dentinNone 10 10 7 7Slight 0 0 3 3Moderate 0 0 0 0Severe 0 0 0 0

Remaining dentinthickness, mm

Mean 0.507 0.576 0.660 0.587Minimum 0.124 0.179 0.04 0.192Maximum 0.935 0.996 0.996 0.987

*N � 10 for each group.

for grading the pulpal responses. Results of the inflam-

OOOOEe48 Nayyar et al. August 2007

matory pulp response, reactionary dentin deposition,and bacterial staining were statistically analyzed usingKruskal-Wallis nonparametric analysis followed byMann-Whitney U test to evaluate differences betweenthe experimental groups at a significance level of P lessthan .05. The correlations among various variables likeinflammatory pulp response, remaining dentin thick-ness, bacterial presence, and reactionary dentin forma-tion were investigated by Spearman’s Rho correlationcoefficient test at a significance level of P less than .05.The variance of the remaining dentin thickness wasanalyzed using 1-way analysis of variance (ANOVA)

Fig. 1. A, Prime & Bond NT at 7 days showing severeinflammation. RDT in this tooth was 0.124 mm. Dense ac-cumulation of inflammatory cells is evident directly oppositethe cut dentinal tubules. Cavity is on the right. (Originalmagnification �40; hematoxylin and eosin [H&E] stained.)B, Higher magnification of pulp-predentin area of Fig. 1, A.There are numerous exudative cells consisting of polymor-phonuclear and mononuclear leukocytes. Odontoblastic layeris completely destroyed. Some odontoblastic nuclei have beendisplaced into the dentinal tubules. (Original magnification�400; H&E stained.)

with a statistical significance of P less than .05.

RESULTSThe electric pulp testing performed before the cavity

preparation and before extraction did not demonstrateany differences among the control and experimentalgroups. The patients reported no particular symptomsduring the experimental time period. Periapical radio-graphs demonstrated that all teeth used in the presentstudy had no periapical pathosis before the clinicalprocedure and extraction.

Table IV shows a summary of the histological find-ings and the thickness of remaining dentin. No statis-tically significant differences of remaining dentin thick-ness were found among the groups (1-way ANOVA,P � .724) at a significance level of .05.

Inflammatory pulp responseIn the Prime & Bond NT, at 7 days, 2 severe, 1

moderate, and 3 slight inflammatory pulp responseswere detected. The RDTs of the 2 specimens, whichshowed severe inflammatory response, were 0.173 and0.124 mm. In these specimens, several congested bloodvessels and numerous inflammatory cells were foundadjacent to the cavity floor. Odontoblastic layer disrup-tion was observed. A few cells were aspirated into thedentinal tubules (Fig. 1, A and B). These 2 specimensalso showed presence of bacteria along the cavity walls(Fig. 2). The RDT of the specimen showing moderatereaction was 0.168 mm. The remaining 7 specimensexhibited either normal pulpal architecture or mild in-flammation. The RDTs of these specimens ranged from0.283 mm to 0.935 mm. At 30 days, only 3 of the 10specimens showed slight inflammatory cell infiltrationin the presence of bacteria. The remaining 7 specimens

Fig. 2. Brown and Brenn–stained section; Bacteria are evi-dent on the thin hybrid layer formed on the lateral cavity wall.(Original magnification �400.)

showed normal pulpal architecture (Fig. 3, A and B).

OOOOEVolume 104, Number 2 Nayyar et al. e49

In the Xeno III group at 7 days, slight, moderate, andsevere reactions were observed in 2, 1, and 1 cases,respectively. The specimen that showed severe reactionhad an RDT of 0.179 mm and was associated withpresence of bacteria. In this tooth, severe inflammatoryresponse with neutrophils and macrophages associatedwith odontoblast layer disruption was observed. Twoslight reactions were observed in specimens havingRDTs of 0.242 mm and 0.682 mm (Fig. 4, A and B). At30 days, only 3 slight inflammatory cell reactions weredetected in the presence of bacteria similar to the Prime& Bond NT group (Fig. 5, A and B). The remaining 7specimens were free of inflammation. In total, 16 of 40cases (40%) showed varying degrees of pulpal inflam-matory activity; 45% of cases in the Prime & Bond NTgroup and 35% of cases in the Xeno III group exhibitedpulpal inflammation. There was no significant differ-ence in the incidence of inflammatory cell responseamong all the groups (P � .05). Inflammatory cellresponse appeared to be highly correlated to both bac-terial presence (Spearman’s Rho P � .001) and thecavity’s remaining dentin thickness (Spearman’s RhoP � .003).

Leakage involving bacteriaThe bacterial leakage ratings of the 2 groups, Prime

& Bond NT and Xeno III, showed no significant dif-ference at 7 and 30 days (P � .05). Stained bacteriawere found in 17 specimens (42.5%) in this study.Prime & Bond NT and Xeno III demonstrated bacterial

Fig. 3. A, Prime & Bond NT at 30 days. The pulp tissue isnormal without inflammation. RDT in this tooth was 0.786mm. Cavity is on the right. Original magnification �40; H&Estained. B, Higher magnification of pulp-predentin area ofFig. 3 A. No disruption of odontoblastic layer is seen. Cell-free zone is present. Predentin is normal in width. Originalmagnification �400; H&E stained.

presence in 45% and 40% cases, respectively.

Irritation dentin formationAt 7 days, in both groups, there was no indication of

any irritation dentin, nor were there any reorganizedodontoblast cells below the cut tubules of the remainingdentin. At 30 days, only a small amount of irritationdentin formation was seen in 3 of 10 specimens in bothgroups. There were no statistically significant differ-ences in irritation dentin formation within any of the 2groups at 7 and 30 days (P � .05). Irritation dentinformation was found to be highly correlated with re-maining dentin thickness (Spearman’s Rho correlationP � .001).

DISCUSSIONThe present study has been carried out in human

teeth as human study probably represents the closest tothe real clinical situation. The results obtained with invivo animal studies cannot be directly extrapolated tohuman clinical conditions because of the different heal-ing potential of the animal and human pulp.9 The foodhabits as well as local environment affecting the be-havior of the restorative material is different in animaland human teeth.20

This study showed that pulpal response to the self-etching adhesive, Xeno III, was minimally differentfrom the total-etch adhesive, Prime & Bond NT. Six-teen of 40 cases (40%) showed varying degrees ofpulpal inflammatory activity; 45% cases of the Prime &Bond NT group and 35% cases of the Xeno III exhib-ited pulpal inflammation. Prime & Bond NT and XenoIII groups detected stainable bacteria in 45% and 40%of specimens, respectively. These findings were corrob-orated by the previous study of Murray et al.,10 whoobserved bacterial microleakage in 33% of teeth andpulp inflammation in 22% of the teeth restored withcertain types of adhesives. Costa et al.,12 using a self-etching primer (Clearfil Liner Bond 2, Kuraray, Osaka,Japan), reported that 43% of their specimens had vary-ing degrees of bacterial penetration in a direct pulp-capping study performed in human teeth.

In agreement with previous studies, it was observedthat with every variable measured, the mean categoryof pulpal inflammation always increased with the pres-ence of bacteria at a short time period of 7 days.6,21

Nevertheless, not all pulpal inflammation was associ-ated with bacteria. Three specimens in the Prime &Bond NT group and 2 specimens in the Xeno III group,at 7 days, had mild to moderate degrees of inflamma-tory cell infiltration in the absence of stainable bacteria.These data suggest the pulpal irritation from the chem-ical activity of the adhesive systems or signify thelimitations of the staining procedures such as loss of

bacteria during histological processing,22 number

OOOOEe50 Nayyar et al. August 2007

of sections examined from one specimen,3 difficulty ofstaining gram-negative organisms,22 and so forth.

The increasing severity of pulpal inflammatory ac-tivity with a decreasing cavity RDT in the presence orabsence of bacteria agrees with previous investiga-tions8,13 reporting that RDT thinner than 300 �m maynot be enough to prevent pulpal damage when a total-etch technique is associated with the application of abonding agent in deep cavities. The greater inflamma-

Fig. 4. A, Xeno III at 7 days demonstrating mild reaction. ROriginal magnification �40; H&E stained. B, Higher magnificnuclei have been displaced into the dentinal tubules. The cdentinal tubules and comprise of neutrophilic and mononuOriginal magnification �400; H&E stained.

Fig. 5. A, Xeno III at 30 days showing mild pulpal response.Original magnification �40; H&E stained. B, Higher magnifibut it has more capillaries than normal adjacent to it. Incrmesenchymal cells are seen in the area of cell-free zone. An inare also seen. Original magnification �400; H&E stained.

tory activity observed with Prime & Bond NT in spec-

imens with decreased RDT may be partly attributable todentin etching before application of this bonding agent.

After dentin etching with an acidic gel, the highpermeability of dentin close to the pulp is markedlyincreased, which may interfere with an intradentin per-meation of fluid resin into the intertubular dentin.23

This seems to result in an unprotected collagen fibrousnetwork between the resin dentin interdiffusion zone,which becomes vulnerable to hydrolysis and future

this tooth was 0.682 mm. Cavity preparation is on the left.f the mild reaction observed in Fig. 4 A. Some odontoblasticsubodontoblastic region are well delimited to the exposed

eukocytes as well as fibroblasts and undifferentiated cells.

in this tooth was 0.363 mm. Cavity preparation is on the left.of mild response of Fig. 5 A. Odontoblastic layer is presentumber of cells resembling fibroblasts and undifferentiated

d number of capillaries and a few extravasated red blood cells

DT ination oells inclear l

RDTcation

eased ncrease

bacterial attack.24 There is also incomplete polymeriza-

OOOOEVolume 104, Number 2 Nayyar et al. e51

tion of primer and\or adhesive resin in deep dentin atthe pulpal floor of cavities.25 Because of heat generatedduring polymerization of these adhesives and compos-ite resin, the inwards dentin fluid movement may alsocarry unpolymerized resin fragments through dentinaltubules to reach the pulp tissue.13

In the Xeno III group, one severe inflammatory re-action in a specimen with RDT 0.179 mm may beattributed to its strong acidic and hydrophilic nature.There is a possibility that more highly acidic and hy-drophilic resin monomers may deeply penetrate notonly intertubular dentin but also water-filled tubules.The water may interfere with acidic monomer polymer-ization.26 The present study found only mild inflamma-tory response with Xeno III in a specimen having anRDT of 0.242 mm. These data suggest that applicationof Xeno III is safer in deeper dentin. As only 2 samplesexhibited RDT less than 300 �m in this study, furtherresearch is warranted to clarify doubts concerning thepulpal response of this adhesive system in very deepdentin.

Secretion of irritation dentin has been observed be-neath more than 50% of restorations.15,21 The currentstudy observed slight deposition of irritation dentin in30% of the specimens at 30 days. The shorter timeframe of the present study would have provided lesstime for the dentin repair process to be fully accom-plished. Other explanations for these differences in-clude a variation between patient samples, includingthe possible effects of treatment history, and otherfactors such as age.27

Both the Prime & Bond NT and Xeno III groupsshowed the favorable response, i.e., number of speci-mens having moderate to severe reaction decreasedwith time despite of presence of continuous bacterialirritation. These data suggest that both adhesives arebiocompatible to the pulp, as the presence of bacteriathat had an important role in initial pulp responses wasnot found to be related with long-term pulpal inflam-mation. It seems that the hybrid layer and the variousdefensive reactions of the pulp would have providedadequate sealing of the dentinal tubules, thus prevent-ing the access of microorganisms to the dentin-pulpcomplex.

The present study evaluated the pulp responses inhealthy human teeth scheduled to be removed for orth-odontic purposes. Consequently, we may speculate thatin clinical conditions in carious teeth in which bacteriaand their products are associated with the cytotoxiceffects of the dentin-bonding agent, the pulp responsewill be more severe than reported in the present study.Although moderate to severe inflammatory response

was not observed at 30 days, still a long-term evalua-

tion is warranted to assess the response of Prime &Bond NT and Xeno III on unexposed human pulp.

CONCLUSIONSPulpal response to the self-etching adhesive, Xeno

III, was minimally different from the total-etch adhe-sive, Prime & Bond NT. Both adhesives, Prime & BondNT and Xeno III, provided for successful pulp healingfollowing non–exposed pulp cavity restoration. How-ever 42% of specimen with leakage involving bacteriaand 40% of pulp with inflammation remain a matter ofconcern.

REFERENCES1. Fusayama T, Nakamura M, Kurosaki N, Iwaku M. Non-pressure

adhesion of a new adhesive restorative resin. J Dent Res1979;58(4):1364-70.

2. Van Meerberk B, De Munck J, Yoshida Y, Inoue S, Vargas M,Vijay P, et al. Adhesion of enamel and dentin: current status andfuture challenges. Oper Dent 2003;28 (3):215-35.

3. Browne RM, Tobias RS. Microbial microleakage and pulpalinflammation: a review. Endod Dent Traumatol 1986;2:177-83.

4. Brännström M, Nyborg H. Cavity treatment with a microbicidalfluoride solution. Growth of bacteria and effect on the pulp. JProsthet Dent 1973;30(3):303-10.

5. Akimoto N, Momoi Y, Kohno A, Suzuki S, Otsuki M, Sujuki S,et al. Biocompatibility of Clearfil Liner Bond 2 and ClearfilAP-X system on non-exposed and exposed primate teeth. Quin-tessence Int 1998;29(3):177-88.

6. White KC, Cox CF, Kanka J 3rd, Dixon DL, Farmer JB, SnuggsHM. Pulpal response to adhesive resin systems applied to acid-etched vital dentin: damp versus dry primer application. Quin-tessence Int 1994;25(4):259-68.

7. Bergenholtz G. Pathogenic mechanisms in pulpal disease. JEndod 1990;16(2):98-101.

8. Costa CAS, Lopes do Nascimento AB, Teixeira HM. Responseof human pulps following acid conditioning and application of abonding agent in deep cavities. Dent Mater 2002;18:543-51.

9. Ersin NK, Eronat N. The comparison of a dentin adhesive withcalcium hydroxide as a pulp-capping agent on the exposed pulpsof human and sheep teeth. Quintessence Int 2005;36:271-80.

10. Murray PE, Windsor LJ, Hafez AA, Stevenson RG, Cox CF.Comparison of pulp responses to resin composites. Oper Dent2003;28(3):242-50.

11. Murray PE, Windsor LJ, Smyth TW, Hafez AA, Cox CF. Anal-ysis of pulpal reactions to restorative procedures, materials, pulpcapping and future therapies. Crit Rev Oral Biol Med 2002;13(6):509-20.

12. Costa CAS, Lopes do Nascimento AB, Teixeira HM, FontanaUF. Response of human pulps capped with a self-etching adhe-sive system. Dent Mater 2001;17:230-40.

13. Hebling J, Giro EMA, Costa CAS. Human pulp response after anadhesive system application in deep cavities. J Dent 1999;27:557-64.

14. Macko DJ, Rutberg M, Langeland K. Pulpal response to theapplication of phosphoric acid to dentin. Oral Surg Oral MedOral Pathol Oral Radiol Endod 1978;45(6):930-46.

15. Murray PE, About I, Lumley PJ, Smith G, Franquin JC, SmithAJ. Postoperative pulpal and repair responses. J Am Dent Assoc2001;131:321-9.

16. Mjor IA, Jokstad A, Quist V. Longevity of posterior restorations.

Int Dent J 1990;40:11-7.

OOOOEe52 Nayyar et al. August 2007

17. Zollner A, Gaengler P. Pulp reactions to different preparationtechniques on teeth exhibiting periodontal disease. J Oral Rehab2000;27:93-102.

18. Heyeraas KJ, Sveen OB, Mjor IA. Part 3: Pulpal inflammationand its sequelae. Quintessence Int 2001;32:611-25.

19. Stanley HR. Design for a human pulp study. I. Oral Surg OralMed Oral Pathol Oral Radiol Endod 1968;25:633-47.

20. Goldberg M, Smith AJ. Cells and extra cellular matrices ofdentin and pulp: a biological basis for repair and tissue engineer-ing. Crit Rev Oral Biol Med 2004;15(1):13-27.

21. Murray PE, About I, Lumley PJ, Franquin JC, Remusat M, SmithAJ. Human odontoblast cell numbers after dental injury. J Dent2000;28:277-85.

22. Cox CF, Suzuki S, Suzuki SH, Cox LK II. Histological evalua-tion of direct pulp capping with various adhesive systems. Pro-ceedings of the international conference on dentin/pulp complex.Quintessence Int 1995;Special issue:P.209-16.

23. Costa CAS, Hebling J, Hanks CT. Current status of pulp cappingwith dentin adhesive systems: a review. Dent Mater 2000;

16(3):188-97.

24. Turki MAL, Akpata ES. Penetrability of dentinal tubules inadhesive-lined cavity walls. Oper Dent 2002;27:124-31.

25. Rueggeberg RA, Margeson DH. The effect of oxygen inhibitionon an unfilled/filled composite system. J Dent Res 1990;69:1652-8.

26. Wang Y, Spencer P. Continuing etching of an all-in-one adhesivein wet dentin tubules. J Dent Res 2005;84(4):350-4.

27. Murray PE, Smith AJ, Windsor LJ, Mjor IA. Remainingdentine thickness and human pulp responses. Int Endod J2003;36:33-43.

Reprint requests:

Seema Nayyar, BDSDepartment of Operative Dentistry and EndodonticsGovernment Dental CollegePGIMS, RohtakHaryana 124001, India

tewarisanjayrohtak@yahoo.co.in