d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Long-term tensile bond strength of differently cemented nanocomposite CAD/CAM crowns on dentin abutment Bogna Stawarczyk a,∗ , Nicola Stich a , Marlis Eichberger a , Daniel Edelhoff a , Malgorzata Roos b , Wolfgang Gernet a , Christine Keul a a b
Department of Prosthodontics, Dental School, Ludwig-Maximilians University Munich, Munich, Germany Division of Biostatistics, Institute of Social and Preventive Medicine, University of Zurich, Zurich, Switzerland
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. To test the tensile bond strength of luted composite computer aided
Received 26 April 2013
design/computer aided manufacturing (CAD/CAM) crowns after use of different adhesive
Received in revised form
systems combined with different resin composite cements on dentin abutments.
13 November 2013
Methods. Human molars (n = 200) were embedded in acrylic resin, prepared in a standardized
Accepted 17 December 2013
manner and divided into 20 groups (n = 10). The crowns were treated as follows: (i) Monobond Plus/Heliobond (MH), (ii) Ambarino P60 (AM), (iii) Visio.link (VL), (iv) VP connect (VP), and (v) non-treated as control groups (CG) and luted with Variolink II (VAR) or Clearfil SA Cement
Keyword:
(CSA). Tensile bond strength (TBS) was measured initially (24 h water, 37 ◦ C) and after aging
Nanocomposite
(5000 thermal cycles, 5/55 ◦ C). The failure types were evaluated after debonding. TBS values
CAD/CAM resins
were analyzed using three-way and one-way ANOVA, followed by post hoc Scheffé-test, and
Crowns
two-sample Student’s t-tests.
Adhesive cement
Results. Among VAR and after aging, CG presented significantly higher TBS (p = 0.007) than
Tensile bond strength
groups treated with MH, AM and VP. Other groups showed no impact of pre-treatment. A
Resin composite cement
decrease of TBS values after thermal aging was observed within CSA: CG (p = 0.002), MH (p < 0.001), VL (p < 0.001), AM (p = 0.002), VP (p < 0.001) and within VAR: MH (p = 0.002) and AM (p = 0.014). Groups cemented with VAR showed significantly higher TBS then groups cemented with CSA: non-aged groups: CG (p < 0.001), and after thermal aging: CG (p = 0.003), MH (p < 0.001), VL (p = 0.005), VP (p = 0.010). Significance. According to the study results nano-composite CAD/CAM crowns should be cemented with VAR. Pre-treatment is not necessary if the tested resin composite cements are used. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
In recent years, the demand of non-metallic restorations in dentistry has increased, based on increased esthetic demands
∗
[1]. The combination of these restorations with adequate adhesive techniques allows a more conservative tooth preparation [2]. Hereby insertion, modeling, and light-curing of the direct composite resin restoration takes place directly in the patient’s mouth. Indirect resin composites are used for
Corresponding author. Tel.: +49 89 5160 9573; fax: +49 89 5160 9503. E-mail address: [email protected] (B. Stawarczyk). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.12.012
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
single tooth restorations such as inlays, onlays or crowns [3]. In vitro studies reported good physical properties of direct and indirect resin composite materials, such as flexural strength and hardness [4–7] as well as material wear stability and low abrasiveness to antagonist teeth [8,9]. As an alternative to conventional indirect and direct resin composites, computer aided design/computer aided manufacturing (CAD/CAM) composites, which can be processed more rapidly and at lower cost [10,11], have been recently introduced. Using CAD/CAM blocks, which are prefabricated in controlled conditions by the manufacturer, offers the chance to use materials at their highest obtainable quality. Such blocks are industrially polymerized under standardized parameters at high temperature and pressure. Therefore, the physical and optical (color stability) properties of these blocks are higher compared to conventionally fabricated resin [12–15]. Additionally, composite resins are more fracture resistant than glass–ceramics, especially when the thickness of the restoration is limited [16,17]. Kassem et al. [18] determined the effect of compressive cyclic loading on fatigue resistance and microleakage of CAD/CAM fabricated glass–ceramics and composite crowns. The authors reported that CAD/CAM composite crowns were more fatigue-resistant than glass–ceramic crowns. The microleakage of the composite crowns showed comparable results to glass–ceramic ones. Another in vitro study reported that CAD/CAM composite resins provided better fracture resistance for non-retentive occlusal veneers than for ceramic [19]. Fasbinder et al. [10] investigated the color stability of CAD/CAM milled inlays of composite and glass–ceramic. Composites showed significantly better color-matches after 3 years than glass–ceramic inlays. Several studies have recommended CAD/CAM fabricated composite overlays and crowns for longterm reconstructions [20–22]. Conditioning of the dental hard tissue can take place in a single step before use of etch-and-rinse systems for conventional resin cements (Variolink II). Another possibility that requires no separate conditioning step of the hard dental tissue is the use of self-etch resin composite cement (Clearfil SA Cement). For pre-treatment of the CAD/CAM generated restoration the different methods can be applied according to the material combinations used. For glass–ceramics, hydroflouric-acid etching should be performed before primer application [23], whereas alloys [24] and zirconia can be pretreated by air-abrasion [25]. For pre-treatment of polymeric materials, primarily abrasion with airborne-particles is used, which results in a cleaned and increased surface [26]. For alloys, zirconia and polymeric restorations, bonding liquids can be applied [24,26–30]. According to the material composition it was shown that further chemical pretreatment of polymeric materials with different bonding liquids is required to improve bond strength [26,28–30]. In summary, the recently introduced CAD/CAM composites are considered as alternative materials to glass–ceramics. Resin composite cements are the material of choice for the adhesive cementation of glass–ceramic reconstructions [31,32]. However, limited information is available on the bond strength between industrially polymerized anatomic CAD/CAM composite restorations and resin composite cements. Therefore, the purposes of this study were (a) to
335
test whether the bond strength of resin cements to CAD/CAM composite would improve by surface conditioning methods, and (b) to evaluate the failure types after debonding. The hypothesis tested was that pre-treated, adhesively cemented composite crowns show higher tensile bond strength results than non-treated ones.
2.
Material and methods
The study used experimental nano-composite CAD/CAM blocks as basis material for milling the crowns. The CAD/CAM material is composed of different nano-adhesives and a high percentage of filler (approx. 80%). The crowns were cemented using a conventional resin composite cement Variolink II (VAR, Ivoclar Vivadent) or a self-adhesive resin composite cement Clearfil SA Cement (CSA, Kuraray) after following pre-treatments: (i) Monobond Plus/Heliobond (MH, Ivoclar Vivadent), (ii) Ambarino P60 (AM, Creamed), (iii) Visio.link (VL, Bredent), (iv) exp. VP connect (VP, Merz Dental), and (v) non-treated as control groups (CG). Table 1 gives detailed information of all tested materials and manufacturer with their application steps. Prior to performing this study, power analysis was calculated with R software (R Development Core Team, The R Foundation for Statistical Computing) using data of a previous study [25]. The aim was to find an impact of composite pretreatment to TBS values. A sample size of 10 in each group will behave 99.9% power to detect the increase by 25% of the mean (difference in means of 0.44 MPa) caused by composite pre-treatment, assuming that the common standard deviation is 0.15 MPa using two group t-test with 0.00082 Bonferroni corrected two-sided significance level. Consequently, 200 specimens were divided into 20 groups of 10 each.
2.1. Preparation of human teeth abutments as well as division in subgroups For this study 200 caries-free human molars were used. The teeth were cleaned from remnant soft tissue and stored in 0.5% Chloramine T (Sigma Aldrich, Seelze, Germany, lot no. 53110) at room temperature during the first 7 days after extraction. Thereafter, the teeth were stored in distilled water at 5 ◦ C for a maximum of 6 months (ISO 11405). The roots of all teeth were subsequently embedded in a cylindrical base metal form as lower holding device with self-polymerized acrylic resin (ScandiQuick, ScanDia, Hagen, Germany). For constant preparation angle of the teeth under constant water-cooling, a parallelometer and a diamond cutting torpedo (Komet Dental, Gebr. Brasseler GmbH & Co. KG, Lemgo, Germany) were used. To obtain a standardized abutment height, all teeth were cut using a grinding machine (Secotom 50, Struers, Ballerup, Denmark). Sharp angles were rounded with a polishing disc (Sof-Lex 1982C/1982M, 3M ESPE, Seefeld, Germany). In summary, prepared teeth had a height of 3 mm, a flat surface, a conicity of 10◦ , rounded edges and a shoulder preparation. After preparation, the tooth abutments were scanned with KaVo Everest Scan (KaVo, Biberach, Germany). For each abutment the STL-datasets were imported into inspection
336
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Table 1 – Brands, manufacturer, lot no., in the study used abbreviations, application and chemical composition of the tested materials. Brands, manufacturer
Abbreviation
Clearfil SA Cement lot no. 058AAA (Kuraray, Tokyo, Japan)
CSA
Variolink II lot no. Base R35481 Catalyst P84939 Syntac Classic lot no. Primer R35489, Adhesive R27600, Heliobond R28391 (Ivoclar Vivadent, Schaan, Liechtenstein)
VAR
Monobond Plus lot no. P20536 Heliobond lot no. P00865 (Ivoclar Vivadent, Schaan, Liechtenstein)
MH
Ambarino P 60 lot no. 2011002057 (Creamed, Marburg, Germany) Visio.link lot no. 114784 (Bredent, Senden, Germany)
AM
VP-Connect lot no 22912 (Merz Dental, Lütjenburg, Germany)
Composition
Resin composite cement Application into the crown. Light curing after placement for 60 s Polymerization unit: Elipar S10 (3M ESPE, Seefeld, Germany)
Pre-treatment of the dentin using Syntac classic: Application 37% Phosphoric acid for 15 s, subsequently rinsing with water and air drying with compressed air Application of Primer for 15 s, subsequently air drying with compressed air Application of Adhesive for 10 s, subsequently air drying with compressed air Application of Bond, subsequently blowing to a thin layer Luting: Application of the cement into the crown. Light curing after placement for 60 s Polymerization unit: Elipar S10 (3M ESPE, Seefeld, Germany) Adhesive systems for pre-treatment Application of Monobond Plus on surface, air dried for 60 s Application of Heliobond, light cured for 10 s Polymerization unit: Elipar S10 (3M ESPE, Seefeld, Germany) Application on surface, air dried for 2 min
VL
Application, light curing for 90 s Polymerization unit: Bre.Lux Power Unit (Bredent, Senden, Germany) Application on surface, air dried for 3 min
VP
TEGDMA: triethylenglycol-dimethacrylate, methacryloyloxydecyl-dihydrogenphosphat.
Application
MMA:
methyl-methacrylate,
software (Quality 12.1.2, Geomagic, Morrisville, NC, US) for the calculation of the cement bond area for TBS calculation (Fig. 1). The 200 prepared teeth were divided into five main groups according the performed pre-treatment listed above (n = 40). In the next step, each of the five pre-treatment groups was additionally divided into two cement groups for the luting procedure: Variolink II (n = 20) and Clearfil SA Cement (n = 20). After cementation, the specimens were stored in distilled water at 37 ◦ C for 24 h. Subsequently, the TBS of half of each cement group (n = 10) was measured. The remaining specimens (n = 10 per subgroup) were thermally aged in a thermocycling machine (SD Mechatronik, Feldkirchen-Westerham, Germany) for 5000 thermal cycles between 5 ◦ C and 55 ◦ C with a dwell time in each water bath of 20 s. Before TBS measurement of the artificially aged groups took place, all specimens were released at room temperature in distilled water for 1 h.
bis-GMA:
MDP, Bis-GMA, TEGDMA, hydrophobic aromatic dimethacrylate, hydrophobic aliphatic dimethacrylate, silanated barium glass filler, silanated colloidal silica, dl-camphorquinone, benzoyl peroxide, initiator, accelerator, pigments, surface treated sodium fluoride Cement: Bis-GMA, TEGDMA, UDMA, benzylperoxide, inorganic fillers, ytterbium trifluoride, Ba–Al–flour silicate glass, spheroid mixed oxide, initiator, stabilizer, pigment Primer: TEGDMA, maleic acid, dimethacrylate, water Adhesive: PEGDMA, maleic acid, gltaraldehyde, water Heliobond: Bis-GMA, triethyleneglycol dimethacrylate
Ethanol, silan, methacrylate phosphoric ester
Dimethacrylate based on phosphor acidesters and phosphon acidesters MMA, pentaerythritltriacrylate, pentaerythritoltetraacrylate, diphenyl (2,4,6,-trimethylbenzoyl)phosphinoxide MMA
bisphenol-A-diglycidylmethacrylate,
MDP:
10-
2.2. Milling, pre-treatment and bonding of the composite CAD/CAM crowns For the milling of the crowns, the prepared tooth abutments were scanned again with a laboratory scanning device (Cerec inEos Blue scanner, Sirona Dental, Bensheim, Germany). The obtained scanning-data of each abutment was used to design a composite crown with a standardized thickness of 2 mm (Cerec Software inLab SW4.0, Sirona Dental) and external retentions for the acrylic resin in the upper holding device. The crowns were milled with a CAD/CAM machine (inLab MC XL milling machine, Sirona), air-abraded in a standardized manner (50 m Al2 O3 , duration: 10 s, angle: 45◦ , pressure: 0.2 MPa, distance: 10 mm) and subsequently cleaned for 5 min using an ultrasonic bath with distilled water. The crowns of the divided specimens were pre-treated as follows:
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
337
Fig. 1 – Calculation of bonding area by inspection software.
Adhesive systems: (i) MH: Monobond Plus was applied as a thin layer on the CAD/CAM composite surface and air-dried for 60 s. Thereafter a thin layer of Heliobond was applied and light cured for 10 s with the polymerizing device Elipar S10 using standard program (3M ESPE, Seefeld, Germany). (ii) AM: applied as a thin coating and air-dried for 120 s. (iii) VL: applied and light cured for 90 s using bre.Lux Power Unit (Bredent, Senden, Germany) using the standard program; intensity according to the manufacturer’s information 190–220 mW/cm2 dependent on wavelength. (iv) VP: applied as thin coating and air-dried for 180 s. (v) CG: non-treated. Within groups cemented with Variolink II, the dentin abutments were pre-treated using Syntac Classic system (Ivoclar Vivadent, Schaan, Liechtenstein) according to the manufacturer’s recommendation (Table 1). For the self-adhesive Clearfil SA Cement no pre-treatment of the abutment teeth was performed. All crowns were cemented according manufacturer’s instructions (Table 1) under 100 g load on dentin abutments. The specimens were light-cured by an LED polymerization light Elipar S10 (3M ESPE) using the standard program with a light intensity of 1200 mW/cm2 . The intensity of the LED lamp was tested using an analyzing device Marc V3 (BlueLight analytics Inc., Halifax, NS, Canada). Both resin cements were light-cured for total polymerization duration of 60 s per crown at two opposed vertical sides and occlusal for 20 s per side. To achieve a constant light polymerization, the output tip was kept in contact with the composite crown. Thereafter, the specimens were stored in distilled water at 37 ◦ C for 24 h. Subsequently, half of all specimens was thermal aged.
2.3.
Tensile bond strength (TBS) measurements
The specimens with the cemented crowns were embedded in the upper holding devices for the TBS tests. To ensure a parallel position of both holding devices with a constant distance of 1.5 mm, C-silicone (Optosil, Heraeus Kulzer, Hanau, Germany) was used to fill the space between the holding devices. The insertion of the acrylic resin (ScandiQuick) into the upper
Fig. 2 – Design of the TBS measurement with upper and lower holding device. Crown after debonding.
holding device was performed through the screw hole at the bottom of the upper cylindrical base metal form. The specimens were positioned in a special device of the testing machine (Zwick 1445, Zwick, Ulm, Germany) and pulled out with a crosshead speed of 5 mm/min until debonding of the crown (Fig. 2). The maximal load was measured before de-bonding occurred. The TBS values were calculated: fracture load/bond area = N/mm2 = MPa. When the crown was separated before the TBS measurement, the TBS result was coded with 0 MPa. The de-bonding surface was examined by two operators under a binocular microscope with 20× magnification (microscope: Stemi 200-C, light source: CL 6000 LED, Zeiss, Oberkochen, Germany) and the failure were classified as follows (Fig. 3): (1) resin cement remains on dentin, (2) resin cement remains on CAD/CAM resin, (3) mixed failure of both types.
2.4.
Statistical analysis
Descriptive statistics such as mean, standard deviation (SD), minimum, median maximum and 95% confidence intervals (95%CI) were calculated. Normality of data distribution was tested using Kolmogorov–Smirnov and Shapiro–Wilk tests. Three- and one-way ANOVA followed by Scheffé post-hoc test was used to determine the significant differences between groups. Unpaired two sample t-test was used for calculation of impact of aging type. Relative frequencies of failure types together with the corresponding 95%CI were provided [33]. The evidence for differences in relative frequencies between treatment groups was obtained by 2 -test. The data were analyzed using SPSS Version 20 (p < 0.05; SPSS INC, Chicago, IL, USA).
338
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Table 2 – Mean TBS, standard deviation, 95% confidence intervals, minimum, median and maximum of the resin cements on differently conditioned CAD/CAM resin surfaces (MPa) and one-way ANOVA followed by Scheffé post-hoc. Pre-treatment
24 h H2 O
24 h H2 O + 5000 TC
Mean (SD)
95% CI
Min/median/max
MH VL AM VP CG
2.49 (0.61)a 1.93 (0.48)a 2.07 (0.83)a 2.64 (0.93)a 2.17 (0.72)a
(2.05; 2.93) (1.58; 2.28) (1.46; 2.67) (1.97; 3.32) (1.65; 2.69)
1/2.59/3 1/1.88/3 0/2.16/3 1/2.8/4 1/2.16/3
MH VL AM VP CG
2.94 (0.99)a 2.67 (1.33)a 2.18 (1.11)a 2.11 (0.93)a 3.79 (1.50)a
(2.22; 3.66) (1.71; 3.63) (1.37; 2.98) (1.43; 2.78) (2.71; 4.87)
1/3.12/4 1/2.53/6 0/2.23/4 1/1.86/4 1/3.85/6
Mean (SD)
95% CI
Min/median/max
0.09 (0.18)* a 0.29 (0.34)a 0.78 (0.73)a 0.16 (0.30)* a 0.0 (0)a
(−0.03;0.22) (0.04;0.54) (0.25;1.32) (−0.05;0.39) –
0/0.001/1 0/0.15/1 0/0.57/2 0/0.001/1 –
1.51 (0.71)a 2.46 (2.12)ab 1.07 (0.64)a 1.46 (1.26)a 4.06 (1.40)b
(0.99;2.02) (0.82;4.08) (0.60;1.54) (0.55;2.37) (3.05;5.07)
0/1.46/3 1/1.68/7 0/1.06/2 0/1.48/4 2/3.92/6
CSA
VAR
ab ∗
Different letters show significant differences between the conditioning methods among one resin composite cement. Not normal distributed.
3.
Results
The descriptive statistics of all tested groups are summarized in Table 2. For groups cemented with VAR and initial groups bonded with CSA Kolmogorov–Smirnov and Shapiro–Wilk tests indicated no violation of the assumption of normality. Two of five aged groups bonded with CSA showed a violation of the normality assumption (Table 2). The TBS values are presented non-parametrically by boxplots (Fig. 4). Three-way ANOVA showed an impact of resin composite cement, pretreatment and aging (p < 0.001–0.007) (Table 3). Therefore, the data was spilt according to the type of resin composite cement, pre-treatment and aging level and analyzed using one-way ANOVA and unpaired two samples t-test. Within thermally aged VAR groups, non-treated CG presented significantly higher TBS then groups pre-treated with MH, AM or VP (p = 0.007). Other groups showed no impact of adhesive system on the TBS values.
3.1.
Impact of aging level
Among Clearfil SA Cement, CG (p = 0.002), MH (p < 0.001), VL (p < 0.001), AM (p = 0.002) and VP (p < 0.001) showed a decrease of TBS after thermal aging compared to non-aged groups. A lower TBS was observed after thermal aging for VAR only for two groups, namely MH (p = 0.002) and AM (p = 0.014).
3.2.
Fig. 3 – Systematic drawing of different failure types: (A) resin composite cement remains on dentin, (B) mixed with resin composite cement remains on dentin and nano-composite crown, (C) resin composite cement remains on nano-composite crown.
Impact of cement
Within non-aged groups, higher TBS was observed for nontreated crowns (control group) cemented with VAR compared to crowns cemented with CSA (p = 0.006). After thermal aging, following with VAR cemented groups showed higher TBS CG (p = 0.003), MH (p < 0.001), VL (p = 0.005), VP (p = 0.010) than with CSA cemented ones.
3.3.
Failure types
In Table 4, relative failure type frequencies (%) of all tested groups are presented. In general, significant differences in
339
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Table 3 – Three-way ANOVA results for comparison of Tensile bond strength after different pre-treatment methods and resin composite cements and aging regimens. Sum of squares Constant parameters Resin composite cement Pre-treatment Aging Resin composite cement × pre-treatment Resin composite cement × aging Pre-treatment × aging Resin composite cement × pre-treatment × aging Error Total
564 56 15 71 29 19 6.4 5.7 171 1105
failure types were detected between the different pretreatment methods (p < 0.001–0.003). While resin cement remaining on dentin (80–100%) was observed more frequently when VL was used, control group (80–90%) and pre-treatment with AM (60–100%) demonstrated mixed (resin cement remained on dentin as well as on CAD/CAM block) failures in all resin cement and aging groups. Within CSA, initial groups pre-treated with MH (90%), AM (100%), VP (90%) as well as the non-treated control group (90%) showed predominantly mixed failures. Among aged groups showed AM (90%) and the control group (80%) mixed failures. For the resin cement VAR mainly mixed failures occurred initial for AM (60%) and the control group (90%) and aged for AM (90%), VP (90%) and control group (90%). The remaining groups failed predominantly with resin cement remaining on dentin after tensile
df
Mean squares
F
p-Value
1 1 4 1 4 1 4 4 171
564 56 3.7 71 7.3 19 1.6 1.4 0.999
565 56 3.7 71 7.3 19 1.6 1.4
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.intl.elsevierhealth.com/journals/dema
Long-term tensile bond strength of differently cemented nanocomposite CAD/CAM crowns on dentin abutment Bogna Stawarczyk a,∗ , Nicola Stich a , Marlis Eichberger a , Daniel Edelhoff a , Malgorzata Roos b , Wolfgang Gernet a , Christine Keul a a b
Department of Prosthodontics, Dental School, Ludwig-Maximilians University Munich, Munich, Germany Division of Biostatistics, Institute of Social and Preventive Medicine, University of Zurich, Zurich, Switzerland
a r t i c l e
i n f o
a b s t r a c t
Article history:
Objectives. To test the tensile bond strength of luted composite computer aided
Received 26 April 2013
design/computer aided manufacturing (CAD/CAM) crowns after use of different adhesive
Received in revised form
systems combined with different resin composite cements on dentin abutments.
13 November 2013
Methods. Human molars (n = 200) were embedded in acrylic resin, prepared in a standardized
Accepted 17 December 2013
manner and divided into 20 groups (n = 10). The crowns were treated as follows: (i) Monobond Plus/Heliobond (MH), (ii) Ambarino P60 (AM), (iii) Visio.link (VL), (iv) VP connect (VP), and (v) non-treated as control groups (CG) and luted with Variolink II (VAR) or Clearfil SA Cement
Keyword:
(CSA). Tensile bond strength (TBS) was measured initially (24 h water, 37 ◦ C) and after aging
Nanocomposite
(5000 thermal cycles, 5/55 ◦ C). The failure types were evaluated after debonding. TBS values
CAD/CAM resins
were analyzed using three-way and one-way ANOVA, followed by post hoc Scheffé-test, and
Crowns
two-sample Student’s t-tests.
Adhesive cement
Results. Among VAR and after aging, CG presented significantly higher TBS (p = 0.007) than
Tensile bond strength
groups treated with MH, AM and VP. Other groups showed no impact of pre-treatment. A
Resin composite cement
decrease of TBS values after thermal aging was observed within CSA: CG (p = 0.002), MH (p < 0.001), VL (p < 0.001), AM (p = 0.002), VP (p < 0.001) and within VAR: MH (p = 0.002) and AM (p = 0.014). Groups cemented with VAR showed significantly higher TBS then groups cemented with CSA: non-aged groups: CG (p < 0.001), and after thermal aging: CG (p = 0.003), MH (p < 0.001), VL (p = 0.005), VP (p = 0.010). Significance. According to the study results nano-composite CAD/CAM crowns should be cemented with VAR. Pre-treatment is not necessary if the tested resin composite cements are used. © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
In recent years, the demand of non-metallic restorations in dentistry has increased, based on increased esthetic demands
∗
[1]. The combination of these restorations with adequate adhesive techniques allows a more conservative tooth preparation [2]. Hereby insertion, modeling, and light-curing of the direct composite resin restoration takes place directly in the patient’s mouth. Indirect resin composites are used for
Corresponding author. Tel.: +49 89 5160 9573; fax: +49 89 5160 9503. E-mail address: [email protected] (B. Stawarczyk). 0109-5641/$ – see front matter © 2013 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dental.2013.12.012
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
single tooth restorations such as inlays, onlays or crowns [3]. In vitro studies reported good physical properties of direct and indirect resin composite materials, such as flexural strength and hardness [4–7] as well as material wear stability and low abrasiveness to antagonist teeth [8,9]. As an alternative to conventional indirect and direct resin composites, computer aided design/computer aided manufacturing (CAD/CAM) composites, which can be processed more rapidly and at lower cost [10,11], have been recently introduced. Using CAD/CAM blocks, which are prefabricated in controlled conditions by the manufacturer, offers the chance to use materials at their highest obtainable quality. Such blocks are industrially polymerized under standardized parameters at high temperature and pressure. Therefore, the physical and optical (color stability) properties of these blocks are higher compared to conventionally fabricated resin [12–15]. Additionally, composite resins are more fracture resistant than glass–ceramics, especially when the thickness of the restoration is limited [16,17]. Kassem et al. [18] determined the effect of compressive cyclic loading on fatigue resistance and microleakage of CAD/CAM fabricated glass–ceramics and composite crowns. The authors reported that CAD/CAM composite crowns were more fatigue-resistant than glass–ceramic crowns. The microleakage of the composite crowns showed comparable results to glass–ceramic ones. Another in vitro study reported that CAD/CAM composite resins provided better fracture resistance for non-retentive occlusal veneers than for ceramic [19]. Fasbinder et al. [10] investigated the color stability of CAD/CAM milled inlays of composite and glass–ceramic. Composites showed significantly better color-matches after 3 years than glass–ceramic inlays. Several studies have recommended CAD/CAM fabricated composite overlays and crowns for longterm reconstructions [20–22]. Conditioning of the dental hard tissue can take place in a single step before use of etch-and-rinse systems for conventional resin cements (Variolink II). Another possibility that requires no separate conditioning step of the hard dental tissue is the use of self-etch resin composite cement (Clearfil SA Cement). For pre-treatment of the CAD/CAM generated restoration the different methods can be applied according to the material combinations used. For glass–ceramics, hydroflouric-acid etching should be performed before primer application [23], whereas alloys [24] and zirconia can be pretreated by air-abrasion [25]. For pre-treatment of polymeric materials, primarily abrasion with airborne-particles is used, which results in a cleaned and increased surface [26]. For alloys, zirconia and polymeric restorations, bonding liquids can be applied [24,26–30]. According to the material composition it was shown that further chemical pretreatment of polymeric materials with different bonding liquids is required to improve bond strength [26,28–30]. In summary, the recently introduced CAD/CAM composites are considered as alternative materials to glass–ceramics. Resin composite cements are the material of choice for the adhesive cementation of glass–ceramic reconstructions [31,32]. However, limited information is available on the bond strength between industrially polymerized anatomic CAD/CAM composite restorations and resin composite cements. Therefore, the purposes of this study were (a) to
335
test whether the bond strength of resin cements to CAD/CAM composite would improve by surface conditioning methods, and (b) to evaluate the failure types after debonding. The hypothesis tested was that pre-treated, adhesively cemented composite crowns show higher tensile bond strength results than non-treated ones.
2.
Material and methods
The study used experimental nano-composite CAD/CAM blocks as basis material for milling the crowns. The CAD/CAM material is composed of different nano-adhesives and a high percentage of filler (approx. 80%). The crowns were cemented using a conventional resin composite cement Variolink II (VAR, Ivoclar Vivadent) or a self-adhesive resin composite cement Clearfil SA Cement (CSA, Kuraray) after following pre-treatments: (i) Monobond Plus/Heliobond (MH, Ivoclar Vivadent), (ii) Ambarino P60 (AM, Creamed), (iii) Visio.link (VL, Bredent), (iv) exp. VP connect (VP, Merz Dental), and (v) non-treated as control groups (CG). Table 1 gives detailed information of all tested materials and manufacturer with their application steps. Prior to performing this study, power analysis was calculated with R software (R Development Core Team, The R Foundation for Statistical Computing) using data of a previous study [25]. The aim was to find an impact of composite pretreatment to TBS values. A sample size of 10 in each group will behave 99.9% power to detect the increase by 25% of the mean (difference in means of 0.44 MPa) caused by composite pre-treatment, assuming that the common standard deviation is 0.15 MPa using two group t-test with 0.00082 Bonferroni corrected two-sided significance level. Consequently, 200 specimens were divided into 20 groups of 10 each.
2.1. Preparation of human teeth abutments as well as division in subgroups For this study 200 caries-free human molars were used. The teeth were cleaned from remnant soft tissue and stored in 0.5% Chloramine T (Sigma Aldrich, Seelze, Germany, lot no. 53110) at room temperature during the first 7 days after extraction. Thereafter, the teeth were stored in distilled water at 5 ◦ C for a maximum of 6 months (ISO 11405). The roots of all teeth were subsequently embedded in a cylindrical base metal form as lower holding device with self-polymerized acrylic resin (ScandiQuick, ScanDia, Hagen, Germany). For constant preparation angle of the teeth under constant water-cooling, a parallelometer and a diamond cutting torpedo (Komet Dental, Gebr. Brasseler GmbH & Co. KG, Lemgo, Germany) were used. To obtain a standardized abutment height, all teeth were cut using a grinding machine (Secotom 50, Struers, Ballerup, Denmark). Sharp angles were rounded with a polishing disc (Sof-Lex 1982C/1982M, 3M ESPE, Seefeld, Germany). In summary, prepared teeth had a height of 3 mm, a flat surface, a conicity of 10◦ , rounded edges and a shoulder preparation. After preparation, the tooth abutments were scanned with KaVo Everest Scan (KaVo, Biberach, Germany). For each abutment the STL-datasets were imported into inspection
336
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Table 1 – Brands, manufacturer, lot no., in the study used abbreviations, application and chemical composition of the tested materials. Brands, manufacturer
Abbreviation
Clearfil SA Cement lot no. 058AAA (Kuraray, Tokyo, Japan)
CSA
Variolink II lot no. Base R35481 Catalyst P84939 Syntac Classic lot no. Primer R35489, Adhesive R27600, Heliobond R28391 (Ivoclar Vivadent, Schaan, Liechtenstein)
VAR
Monobond Plus lot no. P20536 Heliobond lot no. P00865 (Ivoclar Vivadent, Schaan, Liechtenstein)
MH
Ambarino P 60 lot no. 2011002057 (Creamed, Marburg, Germany) Visio.link lot no. 114784 (Bredent, Senden, Germany)
AM
VP-Connect lot no 22912 (Merz Dental, Lütjenburg, Germany)
Composition
Resin composite cement Application into the crown. Light curing after placement for 60 s Polymerization unit: Elipar S10 (3M ESPE, Seefeld, Germany)
Pre-treatment of the dentin using Syntac classic: Application 37% Phosphoric acid for 15 s, subsequently rinsing with water and air drying with compressed air Application of Primer for 15 s, subsequently air drying with compressed air Application of Adhesive for 10 s, subsequently air drying with compressed air Application of Bond, subsequently blowing to a thin layer Luting: Application of the cement into the crown. Light curing after placement for 60 s Polymerization unit: Elipar S10 (3M ESPE, Seefeld, Germany) Adhesive systems for pre-treatment Application of Monobond Plus on surface, air dried for 60 s Application of Heliobond, light cured for 10 s Polymerization unit: Elipar S10 (3M ESPE, Seefeld, Germany) Application on surface, air dried for 2 min
VL
Application, light curing for 90 s Polymerization unit: Bre.Lux Power Unit (Bredent, Senden, Germany) Application on surface, air dried for 3 min
VP
TEGDMA: triethylenglycol-dimethacrylate, methacryloyloxydecyl-dihydrogenphosphat.
Application
MMA:
methyl-methacrylate,
software (Quality 12.1.2, Geomagic, Morrisville, NC, US) for the calculation of the cement bond area for TBS calculation (Fig. 1). The 200 prepared teeth were divided into five main groups according the performed pre-treatment listed above (n = 40). In the next step, each of the five pre-treatment groups was additionally divided into two cement groups for the luting procedure: Variolink II (n = 20) and Clearfil SA Cement (n = 20). After cementation, the specimens were stored in distilled water at 37 ◦ C for 24 h. Subsequently, the TBS of half of each cement group (n = 10) was measured. The remaining specimens (n = 10 per subgroup) were thermally aged in a thermocycling machine (SD Mechatronik, Feldkirchen-Westerham, Germany) for 5000 thermal cycles between 5 ◦ C and 55 ◦ C with a dwell time in each water bath of 20 s. Before TBS measurement of the artificially aged groups took place, all specimens were released at room temperature in distilled water for 1 h.
bis-GMA:
MDP, Bis-GMA, TEGDMA, hydrophobic aromatic dimethacrylate, hydrophobic aliphatic dimethacrylate, silanated barium glass filler, silanated colloidal silica, dl-camphorquinone, benzoyl peroxide, initiator, accelerator, pigments, surface treated sodium fluoride Cement: Bis-GMA, TEGDMA, UDMA, benzylperoxide, inorganic fillers, ytterbium trifluoride, Ba–Al–flour silicate glass, spheroid mixed oxide, initiator, stabilizer, pigment Primer: TEGDMA, maleic acid, dimethacrylate, water Adhesive: PEGDMA, maleic acid, gltaraldehyde, water Heliobond: Bis-GMA, triethyleneglycol dimethacrylate
Ethanol, silan, methacrylate phosphoric ester
Dimethacrylate based on phosphor acidesters and phosphon acidesters MMA, pentaerythritltriacrylate, pentaerythritoltetraacrylate, diphenyl (2,4,6,-trimethylbenzoyl)phosphinoxide MMA
bisphenol-A-diglycidylmethacrylate,
MDP:
10-
2.2. Milling, pre-treatment and bonding of the composite CAD/CAM crowns For the milling of the crowns, the prepared tooth abutments were scanned again with a laboratory scanning device (Cerec inEos Blue scanner, Sirona Dental, Bensheim, Germany). The obtained scanning-data of each abutment was used to design a composite crown with a standardized thickness of 2 mm (Cerec Software inLab SW4.0, Sirona Dental) and external retentions for the acrylic resin in the upper holding device. The crowns were milled with a CAD/CAM machine (inLab MC XL milling machine, Sirona), air-abraded in a standardized manner (50 m Al2 O3 , duration: 10 s, angle: 45◦ , pressure: 0.2 MPa, distance: 10 mm) and subsequently cleaned for 5 min using an ultrasonic bath with distilled water. The crowns of the divided specimens were pre-treated as follows:
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
337
Fig. 1 – Calculation of bonding area by inspection software.
Adhesive systems: (i) MH: Monobond Plus was applied as a thin layer on the CAD/CAM composite surface and air-dried for 60 s. Thereafter a thin layer of Heliobond was applied and light cured for 10 s with the polymerizing device Elipar S10 using standard program (3M ESPE, Seefeld, Germany). (ii) AM: applied as a thin coating and air-dried for 120 s. (iii) VL: applied and light cured for 90 s using bre.Lux Power Unit (Bredent, Senden, Germany) using the standard program; intensity according to the manufacturer’s information 190–220 mW/cm2 dependent on wavelength. (iv) VP: applied as thin coating and air-dried for 180 s. (v) CG: non-treated. Within groups cemented with Variolink II, the dentin abutments were pre-treated using Syntac Classic system (Ivoclar Vivadent, Schaan, Liechtenstein) according to the manufacturer’s recommendation (Table 1). For the self-adhesive Clearfil SA Cement no pre-treatment of the abutment teeth was performed. All crowns were cemented according manufacturer’s instructions (Table 1) under 100 g load on dentin abutments. The specimens were light-cured by an LED polymerization light Elipar S10 (3M ESPE) using the standard program with a light intensity of 1200 mW/cm2 . The intensity of the LED lamp was tested using an analyzing device Marc V3 (BlueLight analytics Inc., Halifax, NS, Canada). Both resin cements were light-cured for total polymerization duration of 60 s per crown at two opposed vertical sides and occlusal for 20 s per side. To achieve a constant light polymerization, the output tip was kept in contact with the composite crown. Thereafter, the specimens were stored in distilled water at 37 ◦ C for 24 h. Subsequently, half of all specimens was thermal aged.
2.3.
Tensile bond strength (TBS) measurements
The specimens with the cemented crowns were embedded in the upper holding devices for the TBS tests. To ensure a parallel position of both holding devices with a constant distance of 1.5 mm, C-silicone (Optosil, Heraeus Kulzer, Hanau, Germany) was used to fill the space between the holding devices. The insertion of the acrylic resin (ScandiQuick) into the upper
Fig. 2 – Design of the TBS measurement with upper and lower holding device. Crown after debonding.
holding device was performed through the screw hole at the bottom of the upper cylindrical base metal form. The specimens were positioned in a special device of the testing machine (Zwick 1445, Zwick, Ulm, Germany) and pulled out with a crosshead speed of 5 mm/min until debonding of the crown (Fig. 2). The maximal load was measured before de-bonding occurred. The TBS values were calculated: fracture load/bond area = N/mm2 = MPa. When the crown was separated before the TBS measurement, the TBS result was coded with 0 MPa. The de-bonding surface was examined by two operators under a binocular microscope with 20× magnification (microscope: Stemi 200-C, light source: CL 6000 LED, Zeiss, Oberkochen, Germany) and the failure were classified as follows (Fig. 3): (1) resin cement remains on dentin, (2) resin cement remains on CAD/CAM resin, (3) mixed failure of both types.
2.4.
Statistical analysis
Descriptive statistics such as mean, standard deviation (SD), minimum, median maximum and 95% confidence intervals (95%CI) were calculated. Normality of data distribution was tested using Kolmogorov–Smirnov and Shapiro–Wilk tests. Three- and one-way ANOVA followed by Scheffé post-hoc test was used to determine the significant differences between groups. Unpaired two sample t-test was used for calculation of impact of aging type. Relative frequencies of failure types together with the corresponding 95%CI were provided [33]. The evidence for differences in relative frequencies between treatment groups was obtained by 2 -test. The data were analyzed using SPSS Version 20 (p < 0.05; SPSS INC, Chicago, IL, USA).
338
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Table 2 – Mean TBS, standard deviation, 95% confidence intervals, minimum, median and maximum of the resin cements on differently conditioned CAD/CAM resin surfaces (MPa) and one-way ANOVA followed by Scheffé post-hoc. Pre-treatment
24 h H2 O
24 h H2 O + 5000 TC
Mean (SD)
95% CI
Min/median/max
MH VL AM VP CG
2.49 (0.61)a 1.93 (0.48)a 2.07 (0.83)a 2.64 (0.93)a 2.17 (0.72)a
(2.05; 2.93) (1.58; 2.28) (1.46; 2.67) (1.97; 3.32) (1.65; 2.69)
1/2.59/3 1/1.88/3 0/2.16/3 1/2.8/4 1/2.16/3
MH VL AM VP CG
2.94 (0.99)a 2.67 (1.33)a 2.18 (1.11)a 2.11 (0.93)a 3.79 (1.50)a
(2.22; 3.66) (1.71; 3.63) (1.37; 2.98) (1.43; 2.78) (2.71; 4.87)
1/3.12/4 1/2.53/6 0/2.23/4 1/1.86/4 1/3.85/6
Mean (SD)
95% CI
Min/median/max
0.09 (0.18)* a 0.29 (0.34)a 0.78 (0.73)a 0.16 (0.30)* a 0.0 (0)a
(−0.03;0.22) (0.04;0.54) (0.25;1.32) (−0.05;0.39) –
0/0.001/1 0/0.15/1 0/0.57/2 0/0.001/1 –
1.51 (0.71)a 2.46 (2.12)ab 1.07 (0.64)a 1.46 (1.26)a 4.06 (1.40)b
(0.99;2.02) (0.82;4.08) (0.60;1.54) (0.55;2.37) (3.05;5.07)
0/1.46/3 1/1.68/7 0/1.06/2 0/1.48/4 2/3.92/6
CSA
VAR
ab ∗
Different letters show significant differences between the conditioning methods among one resin composite cement. Not normal distributed.
3.
Results
The descriptive statistics of all tested groups are summarized in Table 2. For groups cemented with VAR and initial groups bonded with CSA Kolmogorov–Smirnov and Shapiro–Wilk tests indicated no violation of the assumption of normality. Two of five aged groups bonded with CSA showed a violation of the normality assumption (Table 2). The TBS values are presented non-parametrically by boxplots (Fig. 4). Three-way ANOVA showed an impact of resin composite cement, pretreatment and aging (p < 0.001–0.007) (Table 3). Therefore, the data was spilt according to the type of resin composite cement, pre-treatment and aging level and analyzed using one-way ANOVA and unpaired two samples t-test. Within thermally aged VAR groups, non-treated CG presented significantly higher TBS then groups pre-treated with MH, AM or VP (p = 0.007). Other groups showed no impact of adhesive system on the TBS values.
3.1.
Impact of aging level
Among Clearfil SA Cement, CG (p = 0.002), MH (p < 0.001), VL (p < 0.001), AM (p = 0.002) and VP (p < 0.001) showed a decrease of TBS after thermal aging compared to non-aged groups. A lower TBS was observed after thermal aging for VAR only for two groups, namely MH (p = 0.002) and AM (p = 0.014).
3.2.
Fig. 3 – Systematic drawing of different failure types: (A) resin composite cement remains on dentin, (B) mixed with resin composite cement remains on dentin and nano-composite crown, (C) resin composite cement remains on nano-composite crown.
Impact of cement
Within non-aged groups, higher TBS was observed for nontreated crowns (control group) cemented with VAR compared to crowns cemented with CSA (p = 0.006). After thermal aging, following with VAR cemented groups showed higher TBS CG (p = 0.003), MH (p < 0.001), VL (p = 0.005), VP (p = 0.010) than with CSA cemented ones.
3.3.
Failure types
In Table 4, relative failure type frequencies (%) of all tested groups are presented. In general, significant differences in
339
d e n t a l m a t e r i a l s 3 0 ( 2 0 1 4 ) 334–342
Table 3 – Three-way ANOVA results for comparison of Tensile bond strength after different pre-treatment methods and resin composite cements and aging regimens. Sum of squares Constant parameters Resin composite cement Pre-treatment Aging Resin composite cement × pre-treatment Resin composite cement × aging Pre-treatment × aging Resin composite cement × pre-treatment × aging Error Total
564 56 15 71 29 19 6.4 5.7 171 1105
failure types were detected between the different pretreatment methods (p < 0.001–0.003). While resin cement remaining on dentin (80–100%) was observed more frequently when VL was used, control group (80–90%) and pre-treatment with AM (60–100%) demonstrated mixed (resin cement remained on dentin as well as on CAD/CAM block) failures in all resin cement and aging groups. Within CSA, initial groups pre-treated with MH (90%), AM (100%), VP (90%) as well as the non-treated control group (90%) showed predominantly mixed failures. Among aged groups showed AM (90%) and the control group (80%) mixed failures. For the resin cement VAR mainly mixed failures occurred initial for AM (60%) and the control group (90%) and aged for AM (90%), VP (90%) and control group (90%). The remaining groups failed predominantly with resin cement remaining on dentin after tensile
df
Mean squares
F
p-Value
1 1 4 1 4 1 4 4 171
564 56 3.7 71 7.3 19 1.6 1.4 0.999
565 56 3.7 71 7.3 19 1.6 1.4
