- Email: [email protected]
Effect of cavity preparation instruments (oscillating or rotating) on the composite–dentin interface in primary teeth
dental materials Dental Materials 19 (2003) 259±263
www.elsevier.com/locate/dental
Effect of cavity preparation instruments (oscillating or rotating) on the composite±dentin interface in primary teeth Thomas Pioch a,*, Franklin GarcõÂa-Godoy b, Heinz Duschner c, Martin J. Koch a, Hans J. Staehle a, Christof E. DoÈrfer a a
Department of Restorative Dentistry, University of Heidelberg, Im Neuenheimer Feld 400, D-69120 Heidelberg, Germany b Clinical Research Center, School of Dental Medicine, Tufts University, Boston, USA c Department of Applied Structure and Microanalysis, Medical Faculty, University of Mainz, Mainz, Germany Received 26 June 2001; revised 17 December 2001; accepted 15 January 2002
Abstract Objectives: To evaluate the effect of preparation of instruments on the interfacial integrity between cavity wall and composite restoration. Methods: Two class II slot preparations were done in 10 primary teeth either with SonicSys w or with a conventional bur. The cavities were ®lled using an adhesive system. One layer of a ¯owable composite and one layer of a condensable composite were applied. The specimens were analyzed by confocal laser scanning microscopy. Results: In the SonicSys w group the mean thickness of the hybrid layer was 6.12 (0.60) mm; in the control group it was 6.04 (0.63) mm. The difference was not statistically signi®cant. Two fractures were observed in one tooth of each group. These were located only in the enamel. The cavity margins were beveled in all specimens. Signi®cance: Compared to conventional preparations, cavity preparation with SonicSys w has no deleterious effect on the integrity of the interface. q 2003 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. Keywords: Hybrid layer; Oscillating preparation; Confocal laser scanning microscopy; Adhesion
1. Introduction Oscillating preparation systems claim to enable minimal invasive cavity designs thus avoiding damage to the adjacent tissue [1±3]. In general, two different systems can be distinguished depending on the use of standardized ceramic inserts in combination with preparation of instruments or not [4]. One of the advantages of insert systems matched with oscillating cavity preparation over rotary treatment is the potential to reduce polymerization shrinkage stress and to lower the overall coef®cient of thermal expansion [5]. Additionally, oscillating systems are reported to be easier to handle than conventional burs [6], to be less time consuming [6], and to reduce the risk of damage to the adjacent tooth [1,2]. The SonicSys w preparation system (KaVo, Biberach, Germany) is selectively matched with diamond-coated preparation tips available in three different sizes. SonicSys * Corresponding author. Tel.: 149-6221-566025; fax: 149-6221565074. E-mail address: [email protected] (T. Pioch).
inserts (Vivadent, Schaan, Liechtenstein) are based on leucite-reinforced glass±ceramic (Fig. 1). It was reported that cavity preparation with SonicSys did not increase marginal defects in enamel or dentin compared to conventional preparation techniques [6,7]. Investigations on the integrity between restoration materials and adjacent tissue prepared with oscillating devices were carried out using arti®cial teeth [7] or extracted permanent teeth [5,6,8]. Though there are various reports on cavity preparations with oscillating devices in permanent teeth [1±3,5,6,8], no report exists so far on the interfacial integrity of restorations bonded to primary teeth. Because of the lower mineral density [9,10] and the different chemical composition [11], the resin bonding mechanism to primary teeth may be different from that of permanent teeth. These alterations might explain varying bond strengths between the resin and primary teeth or permanent teeth, respectively, [12± 15]. The smear layer of conventionally prepared dentin of primary teeth was removed using one of the three commercially available bonding systems and a resin-reinforced hybrid layer was readily generated in all specimens [16]. Information on the characteristics of the smear layer
0109-5641/03/$20.00 + 0.00 q 2003 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. PII: S 0109-564 1(02)00037-4
260
T. Pioch et al. / Dental Materials 19 (2003) 259±263
Fig. 1. Oscillating instruments SonicSys for mesial and distal cavity preparation and one ceramic insert.
generated by preparing enamel or dentin with SonicSys w so far is not available in the literature. The smear layer is only one aspect, which may affect the bonding mechanisms, as it has to be demineralized by acid-etching or by the etching potential of self-etching primers. The hybrid layer is signi®cantly thicker in primary teeth than in permanent teeth [17]. Therefore, tooth preparations in primary teeth should be studied by oscillating devices vs conventional techniques in particular with respect to the thickness of hybrid layers and the interfacial integrity between tooth and composite restoration. 2. Materials and methods Ten extracted human primary caries-free teeth were used for class II slot preparations. In each tooth two cavities (mesial and distal side) of similar size and location were prepared, either with SonicSys w or with a conventional bur. The oscillating SonicSys w instruments are available in three different sizes and size No. 2 was selected for cavity preparation. The enamel was etched (Table 1) for 30 s, dentin for 15 s and subsequently rinsed with water spray for 15 s. The primer (component of the bonding system, Table 1) was applied using a brush and then left undisturbed for 15 s. The surface was dried by a gentle air stream. The adhesive (component of the bonding system) was also applied with a brush, and air-dried after 10 s. Primer and adhesive components were labeled with a ¯uorescent dye rhodamin-B-isothiocyanate (E. Merck, Darmstadt, Germany). The approximate
concentration was 0.1% by weight. The un®lled resin Heliobond (component of the adhesive system) was also applied with a brush. Excess liquid was blown off with air. The bonding agents were light cured for 10 s. The cavities were ®lled layer by layer with a ¯owable composite and by a packable composite (Table 1), each light cured for 40 s. Each prepared tooth was set in a block of methacrylate resin and vertically sectioned into two halves by a watercooled microtome saw (Leica 1600, Bensheim, Germany). The sections were approximately parallel to the tooth axis. Prior to confocal laser scanning microscopy (CLSM) inspection, the specimens were stored in water for 24 h at 20 8C and kept humid throughout the whole experiment. The measurements were carried out by CLSM (Leica CLSM Diaplan, Bensheim, Germany) equipped with oil immersion objectives (100 £ /NA 1.3; 25 £ /NA 0.75) and an Ar/Kr-mixed gas laser illumination source (adjustable up to 40 mW). For excitation of ¯uorescence at 488 nm, a 510 nm long pass ®lter was used. Using a 530 nm short pass ®lter, ¯uorescent light emitted from the specimens was discriminated from re¯ected and scattered light. CLSM images were recorded in re¯ection mode, in ¯uorescent mode and in an overlay mode of both the techniques. In all experiments, standard and ®xed settings for contrast, brightness and laser power were used. Histo-tomographic images (optical sections parallel to the surfaces) were recorded 10 mm under the sample surfaces. For lateral calibration a silicon testing grid with a certi®ed line width of 8.87 (0.01) mm was used (Plano, Marburg, Germany). In order to quantify the thickness of hybrid layers, each individual specimen was measured at 10 different locations of the dentin/®lling interface. The measured results were the distances perpendicular to the dentin±composite junction relative to a visible boundary (see arrow in Fig. 2). The existence of fractures and typical boundary structures were documented. All data from the measurements were subjected to the statistical software package SPSS (SPSS Inc., Chicago IL, USA). After establishing the distribution of data with the Kolmogoroff±Smirnov/Liljefors test, the paired t-test was used to prove the statistical signi®cance of differences between the mean values. The level of statistical signi®cance was set at p 0:05 [18]. Power calculations were performed with the statistical software SamplePower (SPSS Inc., Chicago IL, USA) [19,20]. The smallest effect, which was de®ned to be relevant for detection, was set to 1.0 mm [21,22].
Table 1 The materials used Restoration materials
Batch number
Manufacturer
Email preparator (etchant) Syntac classic (bonding system) Tetric ¯ow (¯owable composite) Tetric ceram (packable composite)
C09047 532891 546328 546307
Ivoclar-Vivadent, Liechtenstein Ivoclar-Vivadent, Liechtenstein Ivoclar-Vivadent, Liechtenstein Ivoclar-Vivadent, Liechtenstein
T. Pioch et al. / Dental Materials 19 (2003) 259±263
261
Fig. 2. High-magni®cation cross-sectional CLSM image of the resin-dentin interface (combined ¯uorescent- and re¯ection mode). The thickness of the hybrid layer was measured (arrow).
3. Results In all specimens, hybrid layers at the interface between the adhesive resin and the dentin substrates were identi®ed by CLSM (Fig. 2). This zone of resin in®ltration manifested similar micromorphological features in dentin surfaces treated either by SonicSys w or by a rotating instrument. In the control group, the mean thickness of hybrid layer was 6.04 (0.63) mm whereas in the SonicSys w group it was 6.12 (0.60) mm. The difference was not statistically signi®cant (paired t-test, p . 0.05). With a given criterion for the level of statistical signi®cance of p 0:05 and a sample size of 10 per group, the study had a power of 99.5% for detecting a difference of 1 mm thickness of the hybrid layer (present standard deviation 0.6 mm). Independent of the treatment, the cavity margins were beveled in all specimens (Fig. 3). In the SonicSys w group the interface between dentin and composite always appeared as a wave, whereas the interfaces in the control group appeared to be relatively smooth and straight (Fig. 4). Two fractures restricted to enamel only were observed in one tooth of each group (Figs. 5 and 6).
Fig. 3. Cross-sectional CLSM-image of the bevel-shaped margin (combined ¯uorescent- and re¯ection mode).
Fig. 4. Cross-sectional CLSM images of the composite±dentin interfaces (combined ¯uorescent- and re¯ection mode). (a) SonicSys-preparation: the interface appeared as a wavelike structure. (b) Bur-preparation: the interface appeared to be relatively straight.
4. Discussion Electron microscopic techniques require sample pretreatments including drying and coating. Furthermore, in SEM experiments, only those surfaces can be visualized, the contours of which have been brought into relief by acid-etching, ion-beam-sputtering or by fracturing of the specimen. Undoubtedly, the necessity of sample preparation bears the risk of artifacts. In contrast, CLSM microscopic histo-tomographies of subsurface areas of the specimens can be recorded in their natural wet state. Thus, preparation artifacts practically can be excluded. The interpretation of the present CLSM measurements is a function of the migration and penetration behavior of the dye used for labeling the primer components. The type of rhodamin-B-isothiocyanate used in this study was practically insoluble in water but soluble in the organic solvents of the presently tested dentin bonding systems. In consequence, leaching of the dye from the resin into dentin can be practically excluded and so far has not been observed in own experiments [23]. The dye migrated, however, into the composite material when preparations were carried out as described earlier. This interdiffusion into the resin may occur within the non-polymerized resin, either due to the
262
T. Pioch et al. / Dental Materials 19 (2003) 259±263
Fig. 5. (a) Cross-section of a fracture line (arrow) within the enamel after preparation with a bur (re¯ection mode). (b) Same location as in Fig. 5(a) but in ¯uorescent mode.
Fig. 6. (a) Cross-section of a fracture line (arrow) within the enamel after SonicSys-preparation (re¯ection mode). (b) Same location as in Fig. 6a but in ¯uorescent mode.
air inhibited layer or prior the completion of the polymerization process. The hybrid layer thickness (HLT) of acid-etched dentin of primary teeth, as measured by SEM techniques, was reported to be in a range from 2.9 to 11.7 mm [17,24]. With CLSM, the HLT was found be 6.12 (0.60) mm in the SonicSys-group and 6.04 (0.63) mm in the control group. This difference was not statistically signi®cant. Therefore one of the aims of this study is to test the null hypothesis that both preparation techniques generated hybrid layers of similar width. The criterion for signi®cance (alpha) was set at 0.05 (two-tailed). As there is no published metric data on the hybrid layer thickness of dentin bonding systems in primary teeth, we estimated the minimum effect that would be important to detect (`clinical signi®cance level') at 1 mm, referring to recent publications about the HLT in permanent teeth [21,22]. As a consequence of the two alternate preparation techniques, data obtained under present experimental conditions achieved a statistical power of 99.5% for detecting a difference in HLT on the level of 1 mm [19,20]. Therefore, from this point of view both preparation techniques have to be regarded as equal, so that the SonicSys w-treated dentin surface can be considered not to affect hybridization. Experiments on HLT of permanent teeth with the same material and identical CLSM settings resulted in a value
of 4.1 (3.3) mm [22]. The comparison of data of both studies leads to the conclusion that the HLT in primary teeth is signi®cantly thicker than that of permanent teeth (p , 0.001, Mann±Whitney-Test). Using SEM techniques, similar results were also reported by other authors [17,24]. The CLSM technique, allowing the inspection of naturally wet samples, supports the SEM data. In the SonicSys-group the morphology of the interface between dentin and composite almost always appeared as a wave, whereas that in the control group seemed to be relatively smooth and straight. This con®rms the ®ndings of others [6,7,25]. Cracks or fractures, which might be induced to this individual interface morphology after SonicSys preparation, could not be found in this study. Only one enamel fracture in each group was apparent (Figs. 5 and 6). This low quantity of fractures in enamel is in agreement with that reported in the literature [6,7]. 5. Conclusion Compared to conventional techniques cavity, preparation with SonicSys w in primary teeth does not affect the integrity of the composite±dentin interface.
T. Pioch et al. / Dental Materials 19 (2003) 259±263
Acknowledgements This investigation was supported by the German Research Foundation (Deutsche Forschungsgemeinschaft) Research Grant No.: STA 294/2-1 and Du 106/12-1. References [1] Hugo B. Development and possible applications of oscillating cavity preparation methods (Part I). Dtsch ZahnaÈrztl Z 1997;52:637±49. [2] Hugo B. Development and possible applications of oscillating cavity preparation methods (Part I). Dtsch ZahnaÈrztl Z 1997;52:718±27. [3] Lussi A. Verletzung der NachbarzaÈhne bei der PraÈparation approximaler KavitaÈten. Schweiz Monatsschr Zahnmed 1995;105:1259±64. [4] Federlin M, Thonemann B, Schmalz G. InsertsÐmega®llers in composite restorations: a literature review. Clin Oral Invest 2000;4: 1±8. [5] KarguÈl B, Glockner K, StaÈdtler P. Marginal adaptation of the sonicsys approx system in comparison to other tooth-colored restorations. An in-vitro-study in the SEM. ZahnaÈrztl Welt 2000;109:16±20. [6] Krejci I, Lutz F, Krejci D. Schall-/ultraschallbetriebene diamantierte Instrumente zur KavitaÈtenpraÈparation, zum Konturieren und zum Finieren. ZahnaÈrztl Welt 1995;104:781±6. [7] Boer C, Frentzen M. Histologic evaluation of the cavity margin using the SonicSys-system. ZahnaÈrztl Welt 1999;108:582±9. [8] Schmeiser R, Hannig M. RandqualitaÈt approximaler KompositfuÈllungen in sonoabrasiv bearbeiteten Mikro-KavitaÈten. Dtsch ZahnaÈrztl Z 1999;54:127±32. [9] Weidmann SM, Weatherell JA, Hamm S. Variations of enamel density in sections of human teeth. Arch Oral Biol 1967;12:85±97. [10] Wilson PR, Beynon AD. Mineralization differences between human deciduous and permanent enamel measured by quantitative microradiography. Arch Oral Biol 1989;34:85±8. [11] Naujoks R, Schade H, Zelinka F. Chemical composition of different areas of the enamel of deciduous and permanent teeth. Caries Res 1967;1:137±43.
263
[12] Elkins CJ, McCourt JW. Bond strength of dentinal adhesives in primary teeth. Quintess Int 1993;24:271±3. [13] Jumlongras D, White GE. Bond strengths of composite resin and compomers in primary and permanent teeth. J Clin Pediatr Dent 1997;21:223±9. [14] Salama FS, Tao L. Comparison of Gluma bond strength to primary vs permanent teeth. Pediatr Dent 1991;13:163±6. [15] Walls AW, McCabe JF, Murray JJ. Factors in¯uencing the bond strength between glass polyalkenoate (ionomer) cements and dentine. J Oral Rehabil 1988;15:537±47. [16] Araujo FB, GarcõÂa-Godoy F, Issao M. A comparison of three resin bonding agents to primary teeth. Pediatr Dent 1997;19:253±7. [17] NoÈr JE, Feigal RJ, Dennison JB, Edwards CA. Dentin bonding: SEM comparison of the resin±dentin interface in primary teeth. J Dent Res 1996;75:1396±403. [18] Dawson-Saunders B, Trapp RG. Basic and clinical biostatistics, Norwalk: Appleton & Lange, 1994. p. 125±42. [19] Altman DG. Practical statistics for medical research, London: Chapman & Hall, 1991. p. 455±8. [20] Dupont WD, Plummer WD. Power and sample size calculations: a review and computer program. Controlled Clin Trials 1990;11:116± 28. [21] Hashimoto H, Ohno H, Endo K, Kaga M, Sano H, Oguchi H. The effect of hybrid layer thickness on bond strength: demineralized dentin zone of the hybrid layer. Dent Mater 2000;16:406±11. [22] Pioch T, Stotz S, Buff E, Duschner H, Staehle HJ. In¯uence of different etching times on hybrid layer formation and tensile bond strength. Am J Dent 1998;11:202±6. [23] D'Souza P, Duschner H, Staehle HJ, Pioch T. Dentin bonding systems: a comparative study of SEM and CLSM used to visualize the resin±dentin interface. Acta Med Dent Helv 1999;4:20±6. [24] Fritz U, GarcõÂa-Godoy F, Finger WJ. Enamel and dentin bond strength and bonding mechanism to dentin of Gluma CPS to primary teeth. J Dent Child 1997;64:32±8. [25] Hugo B, Stassinakis A, Hotz P, Klaiber B. Reproducible preparation of standardized class II cavities. Dtsch ZahnaÈrztl Z 1996;51:746± 50.
www.elsevier.com/locate/dental
Effect of cavity preparation instruments (oscillating or rotating) on the composite±dentin interface in primary teeth Thomas Pioch a,*, Franklin GarcõÂa-Godoy b, Heinz Duschner c, Martin J. Koch a, Hans J. Staehle a, Christof E. DoÈrfer a a
Department of Restorative Dentistry, University of Heidelberg, Im Neuenheimer Feld 400, D-69120 Heidelberg, Germany b Clinical Research Center, School of Dental Medicine, Tufts University, Boston, USA c Department of Applied Structure and Microanalysis, Medical Faculty, University of Mainz, Mainz, Germany Received 26 June 2001; revised 17 December 2001; accepted 15 January 2002
Abstract Objectives: To evaluate the effect of preparation of instruments on the interfacial integrity between cavity wall and composite restoration. Methods: Two class II slot preparations were done in 10 primary teeth either with SonicSys w or with a conventional bur. The cavities were ®lled using an adhesive system. One layer of a ¯owable composite and one layer of a condensable composite were applied. The specimens were analyzed by confocal laser scanning microscopy. Results: In the SonicSys w group the mean thickness of the hybrid layer was 6.12 (0.60) mm; in the control group it was 6.04 (0.63) mm. The difference was not statistically signi®cant. Two fractures were observed in one tooth of each group. These were located only in the enamel. The cavity margins were beveled in all specimens. Signi®cance: Compared to conventional preparations, cavity preparation with SonicSys w has no deleterious effect on the integrity of the interface. q 2003 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. Keywords: Hybrid layer; Oscillating preparation; Confocal laser scanning microscopy; Adhesion
1. Introduction Oscillating preparation systems claim to enable minimal invasive cavity designs thus avoiding damage to the adjacent tissue [1±3]. In general, two different systems can be distinguished depending on the use of standardized ceramic inserts in combination with preparation of instruments or not [4]. One of the advantages of insert systems matched with oscillating cavity preparation over rotary treatment is the potential to reduce polymerization shrinkage stress and to lower the overall coef®cient of thermal expansion [5]. Additionally, oscillating systems are reported to be easier to handle than conventional burs [6], to be less time consuming [6], and to reduce the risk of damage to the adjacent tooth [1,2]. The SonicSys w preparation system (KaVo, Biberach, Germany) is selectively matched with diamond-coated preparation tips available in three different sizes. SonicSys * Corresponding author. Tel.: 149-6221-566025; fax: 149-6221565074. E-mail address: [email protected] (T. Pioch).
inserts (Vivadent, Schaan, Liechtenstein) are based on leucite-reinforced glass±ceramic (Fig. 1). It was reported that cavity preparation with SonicSys did not increase marginal defects in enamel or dentin compared to conventional preparation techniques [6,7]. Investigations on the integrity between restoration materials and adjacent tissue prepared with oscillating devices were carried out using arti®cial teeth [7] or extracted permanent teeth [5,6,8]. Though there are various reports on cavity preparations with oscillating devices in permanent teeth [1±3,5,6,8], no report exists so far on the interfacial integrity of restorations bonded to primary teeth. Because of the lower mineral density [9,10] and the different chemical composition [11], the resin bonding mechanism to primary teeth may be different from that of permanent teeth. These alterations might explain varying bond strengths between the resin and primary teeth or permanent teeth, respectively, [12± 15]. The smear layer of conventionally prepared dentin of primary teeth was removed using one of the three commercially available bonding systems and a resin-reinforced hybrid layer was readily generated in all specimens [16]. Information on the characteristics of the smear layer
0109-5641/03/$20.00 + 0.00 q 2003 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. PII: S 0109-564 1(02)00037-4
260
T. Pioch et al. / Dental Materials 19 (2003) 259±263
Fig. 1. Oscillating instruments SonicSys for mesial and distal cavity preparation and one ceramic insert.
generated by preparing enamel or dentin with SonicSys w so far is not available in the literature. The smear layer is only one aspect, which may affect the bonding mechanisms, as it has to be demineralized by acid-etching or by the etching potential of self-etching primers. The hybrid layer is signi®cantly thicker in primary teeth than in permanent teeth [17]. Therefore, tooth preparations in primary teeth should be studied by oscillating devices vs conventional techniques in particular with respect to the thickness of hybrid layers and the interfacial integrity between tooth and composite restoration. 2. Materials and methods Ten extracted human primary caries-free teeth were used for class II slot preparations. In each tooth two cavities (mesial and distal side) of similar size and location were prepared, either with SonicSys w or with a conventional bur. The oscillating SonicSys w instruments are available in three different sizes and size No. 2 was selected for cavity preparation. The enamel was etched (Table 1) for 30 s, dentin for 15 s and subsequently rinsed with water spray for 15 s. The primer (component of the bonding system, Table 1) was applied using a brush and then left undisturbed for 15 s. The surface was dried by a gentle air stream. The adhesive (component of the bonding system) was also applied with a brush, and air-dried after 10 s. Primer and adhesive components were labeled with a ¯uorescent dye rhodamin-B-isothiocyanate (E. Merck, Darmstadt, Germany). The approximate
concentration was 0.1% by weight. The un®lled resin Heliobond (component of the adhesive system) was also applied with a brush. Excess liquid was blown off with air. The bonding agents were light cured for 10 s. The cavities were ®lled layer by layer with a ¯owable composite and by a packable composite (Table 1), each light cured for 40 s. Each prepared tooth was set in a block of methacrylate resin and vertically sectioned into two halves by a watercooled microtome saw (Leica 1600, Bensheim, Germany). The sections were approximately parallel to the tooth axis. Prior to confocal laser scanning microscopy (CLSM) inspection, the specimens were stored in water for 24 h at 20 8C and kept humid throughout the whole experiment. The measurements were carried out by CLSM (Leica CLSM Diaplan, Bensheim, Germany) equipped with oil immersion objectives (100 £ /NA 1.3; 25 £ /NA 0.75) and an Ar/Kr-mixed gas laser illumination source (adjustable up to 40 mW). For excitation of ¯uorescence at 488 nm, a 510 nm long pass ®lter was used. Using a 530 nm short pass ®lter, ¯uorescent light emitted from the specimens was discriminated from re¯ected and scattered light. CLSM images were recorded in re¯ection mode, in ¯uorescent mode and in an overlay mode of both the techniques. In all experiments, standard and ®xed settings for contrast, brightness and laser power were used. Histo-tomographic images (optical sections parallel to the surfaces) were recorded 10 mm under the sample surfaces. For lateral calibration a silicon testing grid with a certi®ed line width of 8.87 (0.01) mm was used (Plano, Marburg, Germany). In order to quantify the thickness of hybrid layers, each individual specimen was measured at 10 different locations of the dentin/®lling interface. The measured results were the distances perpendicular to the dentin±composite junction relative to a visible boundary (see arrow in Fig. 2). The existence of fractures and typical boundary structures were documented. All data from the measurements were subjected to the statistical software package SPSS (SPSS Inc., Chicago IL, USA). After establishing the distribution of data with the Kolmogoroff±Smirnov/Liljefors test, the paired t-test was used to prove the statistical signi®cance of differences between the mean values. The level of statistical signi®cance was set at p 0:05 [18]. Power calculations were performed with the statistical software SamplePower (SPSS Inc., Chicago IL, USA) [19,20]. The smallest effect, which was de®ned to be relevant for detection, was set to 1.0 mm [21,22].
Table 1 The materials used Restoration materials
Batch number
Manufacturer
Email preparator (etchant) Syntac classic (bonding system) Tetric ¯ow (¯owable composite) Tetric ceram (packable composite)
C09047 532891 546328 546307
Ivoclar-Vivadent, Liechtenstein Ivoclar-Vivadent, Liechtenstein Ivoclar-Vivadent, Liechtenstein Ivoclar-Vivadent, Liechtenstein
T. Pioch et al. / Dental Materials 19 (2003) 259±263
261
Fig. 2. High-magni®cation cross-sectional CLSM image of the resin-dentin interface (combined ¯uorescent- and re¯ection mode). The thickness of the hybrid layer was measured (arrow).
3. Results In all specimens, hybrid layers at the interface between the adhesive resin and the dentin substrates were identi®ed by CLSM (Fig. 2). This zone of resin in®ltration manifested similar micromorphological features in dentin surfaces treated either by SonicSys w or by a rotating instrument. In the control group, the mean thickness of hybrid layer was 6.04 (0.63) mm whereas in the SonicSys w group it was 6.12 (0.60) mm. The difference was not statistically signi®cant (paired t-test, p . 0.05). With a given criterion for the level of statistical signi®cance of p 0:05 and a sample size of 10 per group, the study had a power of 99.5% for detecting a difference of 1 mm thickness of the hybrid layer (present standard deviation 0.6 mm). Independent of the treatment, the cavity margins were beveled in all specimens (Fig. 3). In the SonicSys w group the interface between dentin and composite always appeared as a wave, whereas the interfaces in the control group appeared to be relatively smooth and straight (Fig. 4). Two fractures restricted to enamel only were observed in one tooth of each group (Figs. 5 and 6).
Fig. 3. Cross-sectional CLSM-image of the bevel-shaped margin (combined ¯uorescent- and re¯ection mode).
Fig. 4. Cross-sectional CLSM images of the composite±dentin interfaces (combined ¯uorescent- and re¯ection mode). (a) SonicSys-preparation: the interface appeared as a wavelike structure. (b) Bur-preparation: the interface appeared to be relatively straight.
4. Discussion Electron microscopic techniques require sample pretreatments including drying and coating. Furthermore, in SEM experiments, only those surfaces can be visualized, the contours of which have been brought into relief by acid-etching, ion-beam-sputtering or by fracturing of the specimen. Undoubtedly, the necessity of sample preparation bears the risk of artifacts. In contrast, CLSM microscopic histo-tomographies of subsurface areas of the specimens can be recorded in their natural wet state. Thus, preparation artifacts practically can be excluded. The interpretation of the present CLSM measurements is a function of the migration and penetration behavior of the dye used for labeling the primer components. The type of rhodamin-B-isothiocyanate used in this study was practically insoluble in water but soluble in the organic solvents of the presently tested dentin bonding systems. In consequence, leaching of the dye from the resin into dentin can be practically excluded and so far has not been observed in own experiments [23]. The dye migrated, however, into the composite material when preparations were carried out as described earlier. This interdiffusion into the resin may occur within the non-polymerized resin, either due to the
262
T. Pioch et al. / Dental Materials 19 (2003) 259±263
Fig. 5. (a) Cross-section of a fracture line (arrow) within the enamel after preparation with a bur (re¯ection mode). (b) Same location as in Fig. 5(a) but in ¯uorescent mode.
Fig. 6. (a) Cross-section of a fracture line (arrow) within the enamel after SonicSys-preparation (re¯ection mode). (b) Same location as in Fig. 6a but in ¯uorescent mode.
air inhibited layer or prior the completion of the polymerization process. The hybrid layer thickness (HLT) of acid-etched dentin of primary teeth, as measured by SEM techniques, was reported to be in a range from 2.9 to 11.7 mm [17,24]. With CLSM, the HLT was found be 6.12 (0.60) mm in the SonicSys-group and 6.04 (0.63) mm in the control group. This difference was not statistically signi®cant. Therefore one of the aims of this study is to test the null hypothesis that both preparation techniques generated hybrid layers of similar width. The criterion for signi®cance (alpha) was set at 0.05 (two-tailed). As there is no published metric data on the hybrid layer thickness of dentin bonding systems in primary teeth, we estimated the minimum effect that would be important to detect (`clinical signi®cance level') at 1 mm, referring to recent publications about the HLT in permanent teeth [21,22]. As a consequence of the two alternate preparation techniques, data obtained under present experimental conditions achieved a statistical power of 99.5% for detecting a difference in HLT on the level of 1 mm [19,20]. Therefore, from this point of view both preparation techniques have to be regarded as equal, so that the SonicSys w-treated dentin surface can be considered not to affect hybridization. Experiments on HLT of permanent teeth with the same material and identical CLSM settings resulted in a value
of 4.1 (3.3) mm [22]. The comparison of data of both studies leads to the conclusion that the HLT in primary teeth is signi®cantly thicker than that of permanent teeth (p , 0.001, Mann±Whitney-Test). Using SEM techniques, similar results were also reported by other authors [17,24]. The CLSM technique, allowing the inspection of naturally wet samples, supports the SEM data. In the SonicSys-group the morphology of the interface between dentin and composite almost always appeared as a wave, whereas that in the control group seemed to be relatively smooth and straight. This con®rms the ®ndings of others [6,7,25]. Cracks or fractures, which might be induced to this individual interface morphology after SonicSys preparation, could not be found in this study. Only one enamel fracture in each group was apparent (Figs. 5 and 6). This low quantity of fractures in enamel is in agreement with that reported in the literature [6,7]. 5. Conclusion Compared to conventional techniques cavity, preparation with SonicSys w in primary teeth does not affect the integrity of the composite±dentin interface.
T. Pioch et al. / Dental Materials 19 (2003) 259±263
Acknowledgements This investigation was supported by the German Research Foundation (Deutsche Forschungsgemeinschaft) Research Grant No.: STA 294/2-1 and Du 106/12-1. References [1] Hugo B. Development and possible applications of oscillating cavity preparation methods (Part I). Dtsch ZahnaÈrztl Z 1997;52:637±49. [2] Hugo B. Development and possible applications of oscillating cavity preparation methods (Part I). Dtsch ZahnaÈrztl Z 1997;52:718±27. [3] Lussi A. Verletzung der NachbarzaÈhne bei der PraÈparation approximaler KavitaÈten. Schweiz Monatsschr Zahnmed 1995;105:1259±64. [4] Federlin M, Thonemann B, Schmalz G. InsertsÐmega®llers in composite restorations: a literature review. Clin Oral Invest 2000;4: 1±8. [5] KarguÈl B, Glockner K, StaÈdtler P. Marginal adaptation of the sonicsys approx system in comparison to other tooth-colored restorations. An in-vitro-study in the SEM. ZahnaÈrztl Welt 2000;109:16±20. [6] Krejci I, Lutz F, Krejci D. Schall-/ultraschallbetriebene diamantierte Instrumente zur KavitaÈtenpraÈparation, zum Konturieren und zum Finieren. ZahnaÈrztl Welt 1995;104:781±6. [7] Boer C, Frentzen M. Histologic evaluation of the cavity margin using the SonicSys-system. ZahnaÈrztl Welt 1999;108:582±9. [8] Schmeiser R, Hannig M. RandqualitaÈt approximaler KompositfuÈllungen in sonoabrasiv bearbeiteten Mikro-KavitaÈten. Dtsch ZahnaÈrztl Z 1999;54:127±32. [9] Weidmann SM, Weatherell JA, Hamm S. Variations of enamel density in sections of human teeth. Arch Oral Biol 1967;12:85±97. [10] Wilson PR, Beynon AD. Mineralization differences between human deciduous and permanent enamel measured by quantitative microradiography. Arch Oral Biol 1989;34:85±8. [11] Naujoks R, Schade H, Zelinka F. Chemical composition of different areas of the enamel of deciduous and permanent teeth. Caries Res 1967;1:137±43.
263
[12] Elkins CJ, McCourt JW. Bond strength of dentinal adhesives in primary teeth. Quintess Int 1993;24:271±3. [13] Jumlongras D, White GE. Bond strengths of composite resin and compomers in primary and permanent teeth. J Clin Pediatr Dent 1997;21:223±9. [14] Salama FS, Tao L. Comparison of Gluma bond strength to primary vs permanent teeth. Pediatr Dent 1991;13:163±6. [15] Walls AW, McCabe JF, Murray JJ. Factors in¯uencing the bond strength between glass polyalkenoate (ionomer) cements and dentine. J Oral Rehabil 1988;15:537±47. [16] Araujo FB, GarcõÂa-Godoy F, Issao M. A comparison of three resin bonding agents to primary teeth. Pediatr Dent 1997;19:253±7. [17] NoÈr JE, Feigal RJ, Dennison JB, Edwards CA. Dentin bonding: SEM comparison of the resin±dentin interface in primary teeth. J Dent Res 1996;75:1396±403. [18] Dawson-Saunders B, Trapp RG. Basic and clinical biostatistics, Norwalk: Appleton & Lange, 1994. p. 125±42. [19] Altman DG. Practical statistics for medical research, London: Chapman & Hall, 1991. p. 455±8. [20] Dupont WD, Plummer WD. Power and sample size calculations: a review and computer program. Controlled Clin Trials 1990;11:116± 28. [21] Hashimoto H, Ohno H, Endo K, Kaga M, Sano H, Oguchi H. The effect of hybrid layer thickness on bond strength: demineralized dentin zone of the hybrid layer. Dent Mater 2000;16:406±11. [22] Pioch T, Stotz S, Buff E, Duschner H, Staehle HJ. In¯uence of different etching times on hybrid layer formation and tensile bond strength. Am J Dent 1998;11:202±6. [23] D'Souza P, Duschner H, Staehle HJ, Pioch T. Dentin bonding systems: a comparative study of SEM and CLSM used to visualize the resin±dentin interface. Acta Med Dent Helv 1999;4:20±6. [24] Fritz U, GarcõÂa-Godoy F, Finger WJ. Enamel and dentin bond strength and bonding mechanism to dentin of Gluma CPS to primary teeth. J Dent Child 1997;64:32±8. [25] Hugo B, Stassinakis A, Hotz P, Klaiber B. Reproducible preparation of standardized class II cavities. Dtsch ZahnaÈrztl Z 1996;51:746± 50.