Tubule Density and Diameter in Coronal Dentin fromPrimary and Permanent Human Teeth

Tathiane L. Lenzi,1 Camila de Almeida B. Guglielmi1, Victor E. Arana-Chavez,2

and Daniela P. Raggio1,*

1Department of Pediatric Dentistry, School of Dentistry, Universidade de São Paulo, Av. Lineu Prestes, 2227,Cidade Universitária, São Paulo 05508-000, Brazil

2Department of Biomaterials and Oral Biology, School of Dentistry, Universidade de São Paulo, Av. Lineu Prestes,2227, Cidade Universitária, São Paulo 05508-000, Brazil

Abstract: This study compared dentinal tubule density and diameter of human primary and permanent teethat different depths of the coronal dentin. Crowns of eight primary second molars and eight permanentthird molars were serially sectioned into three disks of ;0.5 mm thickness ~superficial, middle, and deeplayers!, perpendicular to the long axis. Tubule density and diameter were evaluated in 2,000� and3,000� magnifications by scanning electron microscopy. Data obtained were subjected to two-way repeatedmeasures ANOVA and Tukey’s post hoc test ~a � 0.05!. Tubule density was greater in primary teethcompared with permanent ones, regardless of depth ~primary: 124,329 6 43,594 mm2; permanent: 45,972 621,098 mm2!. In general, the tubule density increased as the dentin depth increased, except to thesuperficial and middle layers from permanent teeth. Tubule diameter was larger in the dentin layer close tothe pulp chamber ~superficial: 2.4 6 0.07 mm; middle: 3.70 6 0.06 mm; deep: 4.28 6 0.04 mm!. No dif-ference was observed between primary ~3.48 6 0.81 mm! and permanent teeth ~3.47 6 0.73 mm!. The tubulediameter increases as the dentin depth increases for primary and permanent teeth; however, the tubule densityis higher in primary teeth.

Key words: dentin, human tooth, microstructure, scanning electron microscopy

INTRODUCTION

Dentin is the collagen-based mineralized tissue that formsthe bulk of teeth. Formed by odontoblasts, which leave theircell processes and branch into the matrix during dentino-genesis, dentin is crossed by thousands of dentinal tubulesand canaliculae. The dentin formed by the secretion ofcollagen and noncollagenous proteins is called intertubulardentin. Subsequent to the formation of mantle dentin, eachodontoblast secretes additional noncollagenous compo-nents that mineralize rapidly between the previously formedintertubular dentin and the odontoblast process for estab-lishing the peritubular dentin. Peritubular dentin is hyper-mineralized in relation to intertubular dentin and forms thewall of the dentinal tubule ~Arana-Chavez & Massa, 2004!.

Bonding to dentin is a challenge owing to its anatomicaland compositional heterogeneity ~Marshall et al., 1997!. Thesecharacteristics are even more pronounced when differenttypes of dentin substrates are compared, such as sound ver-sus caries-affected dentin ~Erhardt et al., 2008! or dentin ofprimary versus permanent teeth ~Ricci et al., 2010!.

Bond strength and other parameters used for in vitroevaluations of adhesive material performance are influ-enced by structural characteristics such as diameter and thenumber of dentinal tubules per mm2, as well as the relativeamount of peritubular and intertubular dentin ~Marshallet al., 1997; Pashley & Carvalho, 1997!. These values vary at

different locations in the same tooth and through the den-tin thickness ~Pashley, 1989; Schilke et al., 2000!.

Studies comparing bond strength of dentin of primaryand permanent teeth are not in agreement, and divergencesare found with regard to the bonding efficacy of differentadhesive systems to these substrates. Although some studieshave shown lower bond strength to dentin of primary teeth~Senawongse et al., 2004; Uekusa et al., 2006!, others havefound similar ~Burrow et al., 2002; Soares et al., 2005; Ricciet al., 2010! or even superior performance of the adhesivesin primary teeth ~Hosoya et al., 1997!.

Chemical and micromorphological differences betweenprimary and permanent teeth are usually considered asfactors related to lower bond strength values of adhesivesystems reported in dentin of primary teeth ~Senawongseet al., 2004; Uekusa et al., 2006!. However, the characteris-tics of dentin morphology still need scientific investigation,since the information regarding the comparison of thetubule diameter and density between these substrates is verylimited and controversial. Although Sumikawa et al. ~1999!demonstrated greater tubule density and diameter for pri-mary teeth compared with permanent teeth, the resultsobtained in a study conducted by Koutsi et al. ~1994!suggest the opposite. Furthermore, there is a paucity ofstudies that directly compare potential differences in themicrostructure of dentin from primary and permanentteeth ~Schilke et al., 2000!.

Similarly, the occlusal surface has been most suitablefor standardized measurement of bonding performance to

Received March 19, 2013; accepted May 31, 2013*Corresponding author. E-mail: danielar@usp.br

Microsc. Microanal. Page 1 of 5doi:10.1017/S1431927613012725 Microscopy AND

Microanalysis© MICROSCOPY SOCIETY OF AMERICA 2013

dentin. Some studies have evaluated the occlusal tubuledensity ~Reis et al., 2012! and diameter ~Lopes et al., 2009!from permanent teeth. Nevertheless, to the best of ourknowledge, no data are available on the morphology ofhuman dentin from primary teeth, including different oc-clusal coronal depths, for in vitro investigation of bondstrength.

Therefore, this study aimed to compare dentinal tubuledensity and diameter of human primary and permanentteeth at different depths of the coronal dentin by scanningelectron microscopy ~SEM!.

MATERIAL AND METHODS

Tooth Selection and PreparationSixteen human teeth, consisting of eight primary secondmolars and eight permanent third molars, were collectedafter the patients’ informed consent had been obtainedunder a protocol reviewed and approved by the Institu-tional Ethics Board. The primary molars were obtainedfrom children aged 9–12 years, and permanent teeth wereobtained from individuals aged 18–23 years; they weremainly extracted for orthodontic reasons in both groupsand were caries free and unrestored.

The teeth were disinfected in 0.5% aqueous chloramineand stored in distilled water at 48C until use. The roots weresectioned ;1 mm apical to the cementum–enamel junctionusing a water-cooled diamond saw in a cutting machine~Labcut 1010, Extec Co., Enfield, CT, USA!. The pulp tissueswere removed with endodontic files.

The occlusal enamel was ground with 320-grit siliconcarbide paper under running water to obtain flat dentinsurfaces. The specimens were carefully examined under astereomicroscope at 30� magnification to confirm the ab-sence of enamel islets.

Next, the crown was serially sectioned with a water-cooled diamond saw into disks of ;0.5 mm thickness,perpendicular to the long axis. Three disks were obtainedfrom each tooth at progressively smaller distances from thepulp ~superficial, middle, and deep layers!.

The smear layer was removed ultrasonically by treatingit with 0.27 M EDTA ~pH 7.4! solution for 5 min and0.34 M sodium hypochlorite ~pH 12.3! solution for 3 min,followed by thorough rinsing with water for 5 min ~Schilkeet al., 2000!. The dentin disks were dehydrated in anascending series of ethanol ~30, 50, 70, 80% for 5 min each,and 90, 95, 100% for 10 min each!. After the final ethanolbath, the specimens were dried by immersion in hexameth-yldisilazane ~HMDS, Electron Microscopy Sciences, FortWashington, PA, USA! for 10 min, placed on a filter paperinside a covered glass vial, and kept in vacuum for 24 h.Subsequently, the samples were mounted on aluminumstubs using double-sided carbon tape and sputter-coatedwith gold in a Balzers SDC-050 apparatus ~Bal-Tec AG,Liechtenstein! and analyzed in a SEM LEO 430 ~Leo,Cambridge, UK! operated in the secondary electron modeat 15 kV.

SEM AnalysisElectron micrographs of a representative area from eachtooth at the superficial, medium, and deep depths of thecoronal dentin were taken at 2,000� and 3,000� magnifi-cations. For each tooth, five SEM images were obtainedfrom the central part of the disks at each dentin depth forboth magnifications, resulting in 480 micrographs.

The micrographs at lower magnification were used tocalculate the tubule density by counting the number oftubules in a 2,400 mm2 area of the image measured accord-ing to the scale bar. The tubule density was expressed as theaverage of the number of tubules per mm2.

The average diameter of the tubules was obtained frommeasurements on the higher-magnification micrographs andcalculated by measuring the diameters of five dentinal tu-bules on each micrograph, and the actual dimension wascalculated according to the scale bar ~mm!. Only thosetubules that showed an almost circular lumen were selected.Measurement of the smallest diameter across the tubuleorifice minimized the error caused by tubules cut obliquely.Two trained examiners performed all measurements usingthe Image Tool 3.0 software ~The University of Texas HealthScience Center at San Antonio, TX, USA!.

Statistical AnalysisThe experimental unit in this study was the tooth. Thus, thetubule number and diameter means for every testing groupwas expressed as the average of the eight teeth used pergroup.

Normal distribution of data and equality of varianceswere assumed after Kolmogorov–Smirnov and Bartlett’s tests.Data obtained were subjected to two-way repeated mea-sures analysis of variance ~tooth type—primary and perma-nent versus dentin depth—superficial, middle, and deep!,and a post hoc test ~Tukey’s test at a � 0.05! was used forpairwise comparisons.

RESULTS

Numerical tubule density data are shown as a function ofthe depth and tooth type in Figure 1. The main factors oftooth type and dentin depth, as well as the cross-productinteraction, were statistically significant ~ p , 0.05!.

Tubule density was higher in primary teeth com-pared with permanent teeth, regardless of depth ~primary:124,329 6 43,594 mm2; permanent: 45,972 6 21,098 mm2!.In general, the average numerical tubule density increasedas the dentin depth increased. Only the superficial andmiddle layers of dentin from permanent teeth showed asimilar mean number ~superficial: 28,541 6 3,870 mm2;middle: 39,947 6 2,658 mm2!.

Tubule diameter means for all experimental groups aresummarized in Figure 2. Only the main factor of dentindepth was statistically significant ~ p , 0.05!. Tubule diam-eter was larger in the deep dentin layer nearest to the pulpchamber ~superficial: 2.4 6 0.07 mm; middle: 3.70 60.06 mm; deep: 4.28 6 0.04 mm!. No difference was ob-

2 Tathiane L. Lenzi et al.

served between primary ~3.48 6 0.81 mm! and permanentteeth ~3.47 6 0.73 mm!.

Representative SEM images of the microstructure ofdentin from primary and permanent teeth at different depthsare shown in Figures 3 and 4, respectively. Besides thehigher tubule density at all depths evaluated, the presence ofseveral canaliculae was also observed in dentin of primaryteeth. Primary and permanent teeth did not differ in termsof tubule diameter.

DISCUSSION

Tubule density of coronal dentin was found to increaselinearly from the superficial to the deep layer, except for

permanent teeth, which show statistically similar mean val-ues between the superficial and middle depths. The numberof dentinal tubules is enhanced in a coronoapical direction~Carrigan et al., 1984; Marshall et al., 1997!. This findingcan be explained by the divergence of tubules as theyapproach the pulp.

Most studies have evaluated the tubule density only inthe middle and deep layers ~Garberoglio & Brannstrom,1976; Fosse et al., 1992; Dourda et al., 1994; Schilke et al.,2000!. The similarity between the superficial and middledepths of dentin from permanent teeth might be related toa higher homogeneity in this substrate, which might explainthe superior bonding efficacy of adhesive systems relative todentin of primary teeth ~Senawongse et al., 2004; Uekusaet al., 2006!. Moreover, dentin depth has an important rolein the bond strength of adhesive systems only when super-ficial and deep dentin are compared, whereas no significantdifference was found between superficial and middle dentinlayers ~Hebling et al., 2007!.

Numerical tubule density was greater in primary teeth,independent of the dentin depth. Sumikawa et al. ~1999!evaluated variations in dentin microstructure from primaryanterior teeth at different areas and depths in relation to thedentinoenamel junction ~DEJ! and verified that the tubuledensity in primary teeth was higher compared with perma-nent ones. Conversely, Koutsi et al. ~1994! and Ruschel andChevitarese ~2002! found a lower density of dentinal tu-bules in primary molars. However, the authors comparedthe results obtained with the findings that were alreadyreported in the literature for permanent teeth, which aredifferent from the present study.

There is little information regarding studies that di-rectly compared the density of dentine tubules from pri-mary and permanent teeth. Schilke et al. ~2000! demonstratedno difference in the number of tubules per mm2 betweencoronal dentin layers of human primary and permanentmolars. In the present study, the tubule density means fromprimary teeth were higher than the values reported inliterature ~Koutsi et al., 1994; Sumikawa et al., 1999; Schilkeet al., 2000; Ruschel & Chevitarese, 2002!. This could ex-plain the difference between primary and permanent teeth,which were not observed in a previous study ~Schilke et al.,2000!.

The reported number of dentinal tubules per mm2 forboth permanent and primary teeth varies considerably be-tween previous studies. These differences may be due tofactors such as using only small numbers of teeth ~samplesize! or lack of standardization of distance from the DEJ,including several types and/or age of teeth or dental surfaces.

The average tubule diameter increased as the dentindepth increased, regardless of the tooth type. Garberoglioand Brannstrom ~1976! found that in sound dentin the sizeof the tubule lumen depends on the distance from the pulp.Regarding the comparison of tubule diameter between pri-mary and permanent teeth, the findings reported in theliterature are conflicting. Although some studies have shownlarger tubule diameter for primary teeth in comparison

Figure 1. Mean number of dentinal tubules per mm2 and stan-dard deviation as a function of dentin depth and tooth type.Different capital letters indicate a statistically significant differencefor dentin depth ~ p , 0.05!. Different lower letters indicate astatistically significant difference between tooth type ~ p , 0.05!.

Figure 2. Tubule diameter means ~mm! and standard deviationsfor all experimental groups. Values are in mm. Different capitalletters indicate a statistically significant difference ~ p , 0.05!.

Primary and Permanent Dentin Microstructure 3

with permanent teeth ~Sumikawa et al., 1999!, others havefound similar values ~Schilke et al., 2000! or even smallertubule diameter in primary teeth ~Hirayama, 1990; Koutsiet al., 1994; Ruschel & Chevitarese, 2002!. Some differencesregarding sample selection and methods applied in previousstudies hamper comparisons among them and the presentinvestigation. Tooth age is a factor directly related to tubulediameter. Old teeth that were submitted to external stimuli~attrition, occlusal contacts! induce the deposit of second-ary and tertiary dentin, reducing the dentinal tubule diam-eter. However, in our current study, no differences in tubulediameter were observed between primary and permanentmolars, which corroborates the study of Schilke et al. ~2000!who evaluated human teeth of the same age as this sample.

Results for tubule diameters were slightly higher whencompared with the means described in the literature. In ourcurrent study, the smear layer was removed with a combina-tion of EDTA and NaOCl. Higher concentrations or longerapplication times could lead to dilatation of the tubulelumina ~Schilke et al., 2000!. It is possible that the chosenprocedures could have caused some increase in tubule diam-eter. Koutsi et al. ~1994! also removed the smear layer withEDTA solutions ~0.5 M, pH 7.4, for 2 min and 1 M, pH 7.0,for 1 min, respectively!. Their mean tubule diameters were

smaller when compared with those of the current investiga-tion. However, they used a small sample size, and differentdental surfaces were evaluated relative to this study.

Second primary molars and third permanent molarswere selected and the microstructure of dentin in differentcoronal depths was evaluated, as these teeth are commonlyused in adhesion and other laboratory tests. Thus, theresults obtained could be directly correlated to bond strengthmeasurements.

As no difference was observed in terms of tubule diam-eter between primary and permanent teeth, the tubuledensity seems to be the most important micromorphologi-cal aspect related to bond strength results. The greatertubule density for primary teeth in comparison with perma-nent ones results in a reduced area of intertubular dentinavailable for bonding, especially nearest to the pulp. Thischaracteristic might jeopardize the bonding performance ofadhesive systems, as evidenced in research studies investigat-ing the influence of dentin depth on the bond strength ofthese materials ~Giannini et al., 2001; Hebling et al., 2007!.

Numerous canaliculae were observed in primary teeth,especially at the middle dentin depth. The presence ofcanaliculae may also decrease the area of solid dentin,resulting in higher moisture and lower bond strengths.

Figure 3. Representative scanning electron microscopy images of superficial, middle, and deep dentin from primaryteeth, respectively. Tubule diameter was larger in the deep dentin layer nearest to the pulp chamber ~S: 2.39 6 0.60 mm;M: 3.75 6 0.59 mm; D: 4.31 6 0.91 mm!. Similarly, the average numerical tubule density increased as the dentin depthincreased ~S: 85,541 6 14,116 mm2; M: 115,937 6 23,275 mm2; P: 171,510 6 34,326 mm2!. The presence of numerouscanaliculae, mainly in the middle dentin depth ~pointers!. S, superficial; M, middle; D, deep.

Figure 4. Representative scanning electron microscopy images of superficial, middle, and deep dentin from permanentteeth, respectively. Tubule diameter increased as a function of depth ~S: 2.49 6 0.71 mm; M: 3.66 6 0.45 mm; D: 4.25 60.45 mm!. Tubule density was similar between superficial ~28,541 6 3,870 mm2! and middle ~39,947 6 2,658 mm2!layers, but highest in deep dentin ~69,427 6 10,397 mm2!. The presence of a few canaliculae in the middle dentin depth~pointer!. S: superficial; M: middle; D: deep.

4 Tathiane L. Lenzi et al.

Thus, this study shows that there are important structuraldifferences in coronal dentin of primary teeth comparedwith permanent teeth that may have significant implica-tions for bonding behavior.

CONCLUSION

Tubule diameter increases as the dentin depth increases forprimary and permanent teeth; however, the tubule densityis higher in primary teeth.

ACKNOWLEDGMENTS

The authors thank FAPESP ~Grant #2009/16579/0! andCNPq ~Brazil! for financial support.

REFERENCESArana-Chavez, V.E. & Massa, L.F. ~2004!. Odontoblasts: The

cells forming and maintaining dentine. Int J Biochem Cell Biol36~8!, 1367–1373.

Burrow, M.F., Nopnakeepong, U. & Phrukkanon, S. ~2002!. Acomparison of microtensile bond strengths of several dentinbonding systems to primary and permanent dentin. Dent Mater18~3!, 239–245.

Carrigan, P.J., Morse, D.R., Furst, M.L. & Sinai, I.H. ~1984!. Ascanning electron microscopic evaluation of human dentinaltubules according to age and location. J Endod 10~8!, 359–363.

Dourda, A.O., Moule, A.J. & Young, W.G. ~1994!. A morphomet-ric analysis of the cross-sectional area of dentine occupied bydentinal tubules in human third molar teeth. Int Endod J 27~4!,184–189.

Erhardt, M.C., Toledano, M., Osorio, R. & Pimenta, L.A.~2008!. Histomorphologic characterization and bond strengthevaluation of caries-affected dentin/resin interfaces: Effects oflong-term water exposure. Dent Mater 24~6!, 786–798.

Fosse, G., Saele, P.K. & Eide, R. ~1992!. Numerical density anddistributional pattern of dentin tubules. Acta Odontol Scand50~4!, 201–210.

Garberoglio, R. & Brannstrom, M. ~1976!. Scanning electronmicroscopic investigation of human dentinal tubules. Arch OralBiol 21~6!, 355–362.

Giannini, M., Carvalho, R.M., Martins, L.R., Dias, C.T. &Pashley, D.H. ~2001!. The influence of tubule density and areaof solid dentin on bond strength of two adhesive systems todentin. J Adhes Dent 3~4!, 315–324.

Hebling, J., Castro, F.L. & Costa, C.A. ~2007!. Adhesive perfor-mance of dentin bonding agents applied in vivo and in vitro.Effect of intrapulpal pressure and dentin depth. J Biomed MaterRes B Appl Biomater 83~2!, 295–303.

Hirayama, A. ~1990!. Experimental analytical electron micro-scopic studies on the quantitative analysis of elemental concen-

trations in biological thin specimens and its application todental science. Shikwa Gakuho 90~8!, 1019–1036.

Hosoya, Y., Nishiguchi, M., Kashiwabara, Y., Horiuchi, A. &Goto, G. ~1997!. Comparison of two dentin adhesives toprimary vs. permanent bovine dentin. J Clin Pediatr Dent22~1!, 69–76.

Koutsi, V., Noonan, R.G., Horner, J.A., Simpson, M.D., Mat-thews, W.G. & Pashley, D.H. ~1994!. The effect of dentindepth on the permeability and ultrastructure of primary mo-lars. Pediatr Dent 16~1!, 29–35.

Lopes, M.B., Sinhoreti, M.A., Gonini Junior, A., Consani, S. &McCabe, J.F. ~2009!. Comparative study of tubular diameterand quantity for human and bovine dentin at different depths.Braz Dent J 20~4!, 279–283.

Marshall, G.W., Jr., Marshall, S.J., Kinney, J.H. & Balooch,M. ~1997!. The dentin substrate: Structure and properties re-lated to bonding. J Dent 25~6!, 441–458.

Pashley, D.H. ~1989!. Dentin: A dynamic substrate—A review.Scanning Microsc 3~1!, 161–174; discussion 174–176.

Pashley, D.H. & Carvalho, R.M. ~1997!. Dentine permeabilityand dentine adhesion. J Dent 25~5!, 355–372.

Reis, C., De-Deus, G., Marins, J., Fidel, S., Fidel, R. & Pa-ciornik, S. ~2012!. Mapping large extensions of flat dentinthrough digital microscopy: Introduction to the method andpossible applications. J Adhes Dent 14~4!, 349–354.

Ricci, H.A., Sanabe, M.E., Costa, C.A. & Hebling, J. ~2010!.Bond strength of two-step etch-and-rinse adhesive systems tothe dentin of primary and permanent teeth. J Clin Pediatr Dent35~2!, 163–168.

Ruschel, H.C. & Chevitarese, O. ~2002!. Density and diameterof dentinal tubules of first and second primary human molars—Comparative scanning electron microscopy study. J Clin Pedi-atr Dent 26~3!, 297–304.

Schilke, R., Lisson, J.A., Bauss, O. & Geurtsen, W. ~2000!.Comparison of the number and diameter of dentinal tubules inhuman and bovine dentine by scanning electron microscopicinvestigation. Arch Oral Biol 45~5!, 355–361.

Senawongse, P., Harnirattisai, C., Shimada, Y. & Tagami, J.~2004!. Effective bond strength of current adhesive systems ondeciduous and permanent dentin. Oper Dent 29~2!, 196–202.

Soares, F.Z., Rocha Rde, O., Raggio, D.P., Sadek, F.T. & Car-doso, P.E. ~2005!. Microtensile bond strength of different adhe-sive systems to primary and permanent dentin. Pediatr Dent27~6!, 457–462.

Sumikawa, D.A., Marshall, G.W., Gee, L. & Marshall, S.J.~1999!. Microstructure of primary tooth dentin. Pediatr Dent21~7!, 439–444.

Uekusa, S., Yamaguchi, K., Miyazaki, M., Tsubota, K., Kuro-kawa, H. & Hosoya, Y. ~2006!. Bonding efficacy of single-stepself-etch systems to sound primary and permanent tooth den-tin. Oper Dent 31~5!, 569–576.

Primary and Permanent Dentin Microstructure 5