Strengths of additions to composite or resin-modified glass-ionomer

Author’s Accepted Manuscript Strengths of additions to composite or resinmodified glass-ionomer Richard H. Sullivan, Robert H. Hatch, Daniel M. Stegall, Crisnicaw Veríssimo, Daranee Tantbirojn, Antheunis Versluis www.elsevier.com/locate/ijadhadh

PII: DOI: Reference:

S0143-7496(16)30065-3 http://dx.doi.org/10.1016/j.ijadhadh.2016.03.017 JAAD1820

To appear in: International Journal of Adhesion and Adhesives Cite this article as: Richard H. Sullivan, Robert H. Hatch, Daniel M. Stegall, Crisnicaw Veríssimo, Daranee Tantbirojn and Antheunis Versluis, Strengths of additions to composite or resin-modified glass-ionomer, International Journal of Adhesion and Adhesives, http://dx.doi.org/10.1016/j.ijadhadh.2016.03.017 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Strengths of additions to composite or resin-modified glass-ionomer

Richard H. Sullivan IV a, Robert H. Hatch b, Daniel M. Stegall a, Crisnicaw Veríssimo c, Daranee Tantbirojn b,*, Antheunis Versluis d

a

Dental student, College of Dentistry, University of Tennessee Health Science Center, Memphis,

TN 38163, USA b

Department of Restorative Dentistry, College of Dentistry, University of Tennessee Health

Science Center, Memphis, TN 38163, USA c

School of Dentistry, University of Uberaba, Uberaba, Minas Gerais, Brazil

d

Department of Bioscience Research, College of Dentistry, University of Tennessee Health

Science Center, Memphis, TN 38163, USA

* Corresponding author: Daranee Tantbirojn, Department of Restorative Dentistry, College of Dentistry, University of Tennessee Health Science Center, Memphis, TN 38163, USA Tel: +1 901 448 6372

E-mail addresses: [email protected] (R. Sullivan), [email protected] (R.H. Hatch), [email protected] (D. Stegall), [email protected] (C. Veríssimo), [email protected] (D. Tantbirojn), [email protected] (A. Versluis)

Abstract Objectives: Adding a new layer of material to a cured resin-based composite (RBC) or resinmodified glass ionomer (RMGI) restorations is necessary in dental practice. This study investigated strengths of additions to the two materials. Material and methods: Bar-shaped specimens were made from monolithic RBC or RMGI, or additions of RBC and RMGI onto RBC or RMGI half-bar substrates. For the additions, the substrates were left undisturbed or were ground with silicon carbide paper followed by the application of a self-etch adhesive. Sample size was ten. Flexural strengths were determined by a 4-point bending test in a universal testing machine. Results were statistically analyzed with one-way ANOVA followed by StudentNewman-Keuls post-hoc test (=0.05). Results: Flexural strength of the monolithic RBC (86.7 ± 21.8 MPa) was significantly higher than RMGI (52.6 ± 13.1 MPa). Addition of RBC to cured RBC significantly reduced flexural strength regardless of the substrate surface conditions (34.1 ± 11.5 to 45.7 ± 21.1 MPa). Addition of RMGI to cured RMGI did not significantly reduce flexural strength (36.2 ± 8.4 to 52.7 ± 25.2 MPa). Flexural strength of RBC added on to cured RMGI that was ground and bonded was the lowest (21.5 ± 10.0 MPa). Most specimens from this group exhibited adhesive failure. Conclusions: RBC/RBC additions reduced flexural strength whereas flexural strength of RMGI/RMGI additions was not significantly lower than its cohesive strength. RBC added onto RMGI in the sandwich restorative configuration had lowest failure strength.

Keywords: Self-etch adhesive (A), Resin-based composites (B), Resin-modified glass-ionomer, Strength

2

1. Introduction Adding fresh restorative material to previously set material is often necessary in dental restorative (filling) procedures. An incremental filling technique where the resin-based composite (RBC) material is built up in layers of 2 mm to accommodate curing light penetration for optimal polymerization is a standard dental practice [1]. Even after a restoration is polished a new layer of RBC is often added to the cured material to correct restoration contour or improve esthetic appearance. Monomers of current dental RBC are based on methacrylate chemistry. Their free radical addition polymerization has limited degree of conversion [2]. The unreacted methacrylate carbon double bonds in polymer chains act as sites for bonding when a new increment of uncured RBC is added [3]. However, the free radical polymerization is inhibited by oxygen, evidenced as a tacky layer on the surface that has direct contact with air. Different school-of-thoughts exist about the effect of this oxygen-inhibition layer, where it was shown to interfere with polymerization, improved the bonding, or made no difference in bond strength between RBC increments [4-7]. Regardless of the oxygen-inhibited layer, strengths of fresh RBC additions ranged from 24-91% of the material cohesive strength [3,8-10]. Recently a commercial RBC has been introduced using a proprietary monomer (DX-511) based on urethane dimethacrylate chemistry [11]. DX-511 has twice the molecular weight of traditional RBC monomers in order to decrease polymerization shrinkage by reducing the number of carbon double bonds. Since the bonding between layers of RBC was reduced when the number of unreacted methacrylate groups decreased [8], it would be interesting to know how the new RBC performs. Resin-modified glass-ionomers (RMGI), another type of tooth-color filling materials, have been the restorative material of choice in individuals with high risk of dental caries due to

3

its caries inhibition effect through fluoride release [12,13], and for cases with lesion close to gingival margins where potential for moisture contamination precludes hydrophobic RBCs [14]. RMGIs have a resin component based on methacrylate monomers such as 2hydroyethylmethacrylate or pendent methacryloxy groups on polycarboxylate chains of the conventional glass-ionomer cement [15,16]. The resin components of RMGI undergo a polymerization process that may also have available unreacted methacrylate double bonds similar to cured RBC. RMGI restorations are traditionally not built up in increments. However it would be practical if clinicians have an option to add fresh RMGI to the cured material, for example, to correct insufficient contour. To our knowledge, there is no definitive study regarding the addition of RMGI to RMGI. RMGIs have low mechanical properties and cannot to be used in load-bearing areas such as the occlusal surface of the teeth [15]. In addition, RMGIs lack the surface polish and esthetic quality of RBC. A restorative procedure called a ‘sandwich technique’ was introduced in the 1990s, where the set glass-ionomer in the deeper part of the restoration was laminated with RBC to use the beneficial properties of both materials [17]. It has been suggested that RMGIs bond chemically to RBC through their methacrylate components [15]. Their bond strengths were higher than RBC bonds to conventional glass-ionomers [18,19]. Comparison between the sandwich configuration and RMGI added to RMGI has not been investigated previously. To address the noted gaps in understanding the viability of adding fresh material to previously cured substrates, the objective of this study was to investigate the strengths of additions of 1) RBC to cured RBC, 2) RMGI to cured RMGI, and 3) RBC to cured RMGI.

4

2. Material and methods 2.1. Specimen fabrication Beam-shaped specimens (2 x 2.5 x 25 mm) were made from RBC (GC Kalore, GC Corp, Tokyo, Japan) and/or RMGI (Fuji II LC, GC America, Alsip, IL). Material information is shown in Table 1. Monolithic RBC specimens were made by placing GC Kalore in a stainless steel mold between two glass slides and light-curing them for 40 sec, 4 times on each side. The light curing process was carried out using L.E.Demetron I (Kerr Corp, Danbury, CT) or Rembrandt Allegro (Den-Mat, Santa Maria, CA) light sources with light intensity of 525 mW/cm2 and 560 mW/cm2, respectively, measured with a radiometer (Demetron Model 100, Demetron Research Corp, Danbury, CT). For the monolithic RMGI specimens, 2 capsules of Fuji II LC were triturated for 10 sec using a Rotomix (3M ESPE, Seefeld, Germany) before placing the material in a mold made from vinyl polysiloxane (Express Light Body Impression Material, 3M ESPE, St Paul, MN). The mold was placed between two glass slides and light-cured in the same manner as the composite specimens. Add-on specimens were made by addition of fresh (uncured) material to a cured halflength beam (2 x 2.5 x 12.5 mm) of RBC or RMGI substrate. Figure 1 is a diagram showing the tested groups. Three surface conditions were evaluated for the RBC substrate: 1) undisturbed to represent a surface with oxygen-inhibited layer and to simulate a direct addition of the RBC in an incremental filling technique, 2) cured against a metal insert to represent a surface without oxygen-inhibited layer, and 3) ground with a 320-grit silicon carbide paper (Carbimet 2, Buehler, Lake Bluff, IL) to simulate a clinical condition with bur finishing followed by the application of a self-etching adhesive (G-aenial Bond, GC Corp, Tokyo, Japan). Three surface conditions were

5

tested for the RMGI substrate: 1) undisturbed to simulate direct addition of the RMGI, 2) cured against a metal insert to represent a surface without oxygen-inhibition layer, and 3) ground with the silicon carbide paper to simulate a clinical condition where the cured RMGI is finished with a dental bur. The bonding agent was not applied because it is not clinically used in RMGI restorative procedure. The last group of the add-on specimens was made by addition of uncured RBC to a cured RMGI to resemble the configuration of a sandwich restorative technique. RMGI substrate was ground with the silicon carbide paper to simulate a clinical condition where the cured RMGI is finished with a dental bur followed by the application of the self-etching adhesive (G-aenial Bond). After fabrication, specimens were taken out of the mold and flashes were removed using the 320-grit silicon carbide paper. The specimens were stored for 24 hours at 37 °C in 100% humidity. Sample size was ten per group. Experimental groups are summarized in Table 2. 2.2. Flexural test Specimens were subjected to a four-point bending test as shown in Figure 2. The distance between lower supports was 20 mm and between the loading points 10 mm. Prior to the test, length, width (W), and height (H) of each beam were measured using a digital caliper. Width and height values were averaged from 3 different positions along each beam length. The beams were loaded at a crosshead speed of 0.5 mm/min using a universal testing machine (Instron 5567, Instron Corp, Norwood, MA). Load at failure (N) was recorded with Bluehill 2 software (Version 2.6, Instron Corp). Flexural strength (MPa) of each beam was calculated using 3FL0/(4WH2) [20], where F is the load at failure, L0 is the distance between the lower supports (20 mm). Differences in flexural

6

strengths among the experimental groups were statistically analyzed using one-way ANOVA followed by Student-Newman-Keuls post-hoc test at a significance level of 0.05. Type of failure (cohesive, adhesive, or mixed) was recorded for each specimen using a stereomicroscope.

3. Results Flexural strength values with statistical results and fracture modes are shown in Table 2. There were significant differences in flexural strength values among the groups (one-way ANOVA, p=0.0001). Monolithic RBC had significantly higher flexural strength than monolithic RMGI. The addition of fresh RBC to cured RBC, regardless of the surface condition, significantly reduced flexural strength compared to the monolithic composite. However, the addition of fresh RMGI to cured RMGI did not significantly reduce flexural strength compared to the monolithic RMGI. There is a trend that the RBC surface with oxygen-inhibited layer had higher flexural strength than the surface without or ground, but the differences were not statistical significant. Such trend was not observed with the RMGI additions. Adding RBC to cured RMGI as in the sandwich technique resulted in significantly lower flexural strength than either original material and was the lowest among all addition groups. Except one group, the number of specimens with adhesive and cohesive failure was almost equally distributed when RBC was added to RBC or RMGI was added to RMGI. When RBC was added RMGI as in the sandwich technique, adhesive failure was the predominant mode.

7

4. Discussion Polymerization of dental RBC is not completed after the irradiation process ends, up to half of the methacrylate groups are left unreacted [2]. This behavior allows clinicians to build up dental restorations in increments and accomplish thorough curing in the light-activated materials with limited depth of cure. The opportunity of bonding between fresh and cured RBC decreases when the number of available unreacted methacrylate groups diminishes [8]. In the present study, flexural strengths of the RBC-RBC additions were approximately 40-50% of monolithic specimens. Although it was not an objective of this study to compare between different RBC materials, the strengths of additions were at the lower end of the range reported in literature [3,8,9]. It can be speculated that the high molecular weight monomer may have contributed to less unreacted carbon double bond sites for chemical adhesion to the added fresh material. In addition, the composite used in this study contained prepolymerized fillers to reduce polymerization shrinkage [11]. Prepolymerized fillers have been suggested to lower mechanical properties of some composites [21]. This aspect needs further study. The results of this study showed that flexural strengths significantly decreased compared to the monolithic specimens when the RBC substrates were cured with the interface exposed to air to create the oxygen-inhibited layer, and further decreased when the RBC substrates were cured against a metal insert, representing an interface without oxygen-inhibited layer. However, the difference was not statistically significant. The role of an oxygen-inhibited layer is still controversial. Truffier-Boutry et al [6] showed that the presence of the oxygen-inhibited layer increased shear bond strength, whereas Eliades and Caputo [4] reported that removing such layer increased the interfacial strength. Shawkat et al [7] concluded that bond strength between RBC increments was not entirely dependent on oxygen inhibition at the surface since no significant

8

difference was found whether the substrates were cured in air or nitrogen. Apart from the oxygen-inhibited layer, bonding to ground surfaces (as in polished restorations) was weaker than to uncut RBC. Grinding RBC substrate exposes inorganic fillers, which further reduce the presence of unreacted methacrylate groups at the bonding surface [3,8]. The additions of fresh RMGI to cured RMGI in the present study did not significantly decrease flexural strength compared to the monolithic specimens. We experienced several fractures of the RMGI specimens during the fabrication process, likely due to low initial strength. Natural selection might have excluded weak specimens from the bending tests, although there are many other factors during specimen production that could cause premature failure of good specimens. The results also showed that surface conditions of the RMGI substrates did not affect the flexural strengths of the RMGI additions. The lack of surface effect implies other bonding mechanisms than polymerization between the fresh RMGI to the unreacted methacrylate groups of its resin component. RMGIs possess an acid-base reaction of its glass-ionomer heritage, which continues for a prolonged period after the light-polymerization has stopped [15]. It is possible that fresh RMGI bonds to the RMGI substrate through the continued acid-base reaction or simply adheres by mechanical interlocking. The result of this study, although based on in vitro tests, suggests that the addition of RMGI to freshly-cured RMGI did not necessarily reduce the failure strength. If this finding is confirmed, clinicians will have an option to correct insufficient contour of RMGI restorations. On the other hand, the addition of RBC onto cured RMGI as in a sandwich restoration had significantly lower flexural strength than addition of RBC onto RBC or addition of RMGI onto RMGI. Most specimens failed adhesively at the interface. RMGI/RBC structure is not only used in the sandwich configuration [17], but also used in a general practice as a cavity lining

9

material under composite restorations. As shown in this study, clinicians should be aware that the bonding between RMGI and RBC is a weak link of this laminated structure. Therefore, RMGI liner should be placed only in small areas when it is necessary. The weak link between RMGI and RBC could be a limiting factor for the longevity of restorations as shown in a clinical study where posterior RBC placed with an RMGI liner had more frequent fracture than without RMGI liner [22]. Any variation in surface treatment, type and age of substrates, and bonding agent may affect the failure strength of additions [23]. It was not the objective of this study to cover all potential variations, but to obtain a general insight in how some common clinical addition scenarios could affect the strength in an RBC or RMGI restoration. This study used four-point bending tests to determine the strengths. Bending tests subject materials to both tensile and compressive stresses, with the highest values occurring at the external surfaces (lower and upper) of the specimens. In the four-point bending configuration, these maximum stress values are constant between the two upper supports. This means any weakest cross-section located between the upper supports will fail regardless of the exact centering of the addition interface. Other investigators may choose shear or tensile tests to determine bond strengths. Theoretically, if stress values are determined correctly, strength values should be comparable among these tests. For example, Shawkat et al [7] reported shear bond strength values of additions ranging between 15 and 30 MPa for two RBCs when bonded with or without an oxygen-inhibited layer, and Boyer et al [8] found transverse (three-point bending) strength values of 29-67 MPa at 10 minutes after addition for five RBC products. And RBC addition onto RMGI at 10 minutes yielded 20 MPa in micro-shear bond strength [19]. Although the RBC and RMGI materials tested were not the same, the range of reported strength values appears comparable to this study.

10

Within the experimental limitations, this study found that monolithic RBC had the highest structural strength. If, for practical or polymerization reasons an incremental buildup of RBCs is necessary, the presence of oxygen-inhibited layer was found to slightly improve bonding between layers, but not significant. Failure strength of RMGI, on the other hand, was independent of being monolithic or applied as addition. However, the interface between cured RMGI and added RBC as occurs in sandwich configurations may create a weak link in such restorations.

5. Conclusions Adding fresh composite to cured composite significantly reduced failure strength, but adding RMGI to cured RMGI did not. Adding composite to cured RMGI as in the sandwich restorative technique significantly reduced failure strength.

11

Acknowledgements Supported by the UTHSC College of Dentistry Alumni Endowment Fund and the Tennessee Dental Association Foundation. The materials were donated by GC America Inc.

12

References [1] Bayne SC, Thompson JY, Taylor DF. Dental materials. In: Robertson TM, Heymann HO, Swift EJ jr, editors. Sturdevant’s art & science of operative dentistry. 4th ed. St Louis: Mosby Inc; 2002, Chapter 4. [2] Ferracane JL. Current trends in dental composites. Crit Rev Oral Biol Med 1995;6:302-18. [3] Vankerckhoven H, Lambrechts P, van Beylen M, Davidson CL, Vanherle G. Unreacted methacrylate groups on the surfaces of composite resins. J Dent Res 1982;61:791-5. [4] Eliades GC, Caputo AA. The strength of layering technique in visible light-cured composites. J Prosthet Dent 1989;61:31-8. [5] Rueggeberg FA. From vulcanite to vinyl, a history of resins in restorative dentistry. J Prosthet Dent 2002;87:364-79. [6] Truffier-Boutry D, Place E, Devaux J, Leloup G. Interfacial layer characterization in dental composite. J Oral Rehabil 2003;30:74-7. [7] Shawkat ES, Shortall AC, Addison O, Palin WM. Oxygen inhibition and incremental layer bond strengths of resin composites Dent Mater 2009;25:1338-46. [8] Boyer DB, Chan KC, Reinhardt JW. Build-up and repair of light-cured composites: bond strength. J Dent Res 1984;63:1241-4. [9] El-Askary FS, El-Banna AH, van Noort R. Immediate vs delayed repair bond strength of a nanohybrid resin composite. J Adhes Dent 2012;14:265-74. [10] Tantbirojn D, Fernando C, Versluis A. Failure strengths of composite additions and repairs. Oper Dent 2015;40:364-71.

13

[11] GC Corporation. Kalore Technical Manual Version 4.0 March 2012. http://www.gcamerica.com/products/operatory/KALORE/KALORE_Technical_Manual.pdf Accessed on 25 Sept, 2015. [12] Tantbirojn D, Rusin RP, Bui HT, Mitra SB. Inhibition of dentin demineralization adjacent to a glass-ionomer/composite sandwich restoration. Quintessence Int 2009;40:287-94. [13] Flores-Mir C, John M, Matthews D. Limited evidence exists that glass ionomer restorations in permanent teeth offer a lower risk of developing carious lesions at margins compared with amalgam restorations. J Am Dent Assoc 2010;141:193-4. [14] Dietrich T, Kraemer M, Lösche GM, Wernecke KD, Roulet JF. Influence of dentin conditioning and contamination on the marginal integrity of sandwich Class II restorations. Oper Dent 2000;25:401-10. [15] Hewlett ER, Mount GJ. Glass ionomers in contemporary restorative dentistry – a clinical update. J Calif Dent Assoc 2003;31:483-92. [16] Mitra SB. Adhesion to dentin and physical properties of a light-cured glass-ionomer liner/base. J Dent Res 1991;70:72-4. [17] Mount GJ. Esthetics with glass-ionomer cements and the "sandwich" technique. Quintessence Int 1990;21:93-101. [18] Knight GM, McIntyre JM, Mulyani. Bond strengths between composite resin and auto cure glass ionomer cement using the co-cure technique. Aust Dent J 2006;51:175-9. [19] Costa SB, De Oliveira RVD, Montenegro RV, Fonseca RB, De Carvalho FG, De Barros S, et al. Bond strength evaluation of composite resin bonded to glass ionomer cements after different periods of setting. Int J Adhes Adhes 2013;47:146-50.

14

[20] Gere JM, Timoshenko SP. Mechanics of materials. 3rd ed. Boston: PWS Publishing Company; 1990. [21] Blackham JT, Vandewalle KS, Lien W. Properties of hybrid resin composite systems containing prepolymerized filler particles. Oper Dent 2009;34:697-702. [22] Opdam NJ, Bronkhorst EM, Roeters JM, Loomans BA. Longevity and reasons for failure of sandwich and total-etch posterior composite resin restorations. J Adhes Dent 2007;9:469-75. [23] Özcan M, Kojima AN, Pekkan G, Mesquita AMM, Bottino MA. Adhesion of substrateadherent combinations for early composite repairs: Effect of intermediate adhesive resin application. Int J Adhes Adhes 2014;49:97-102.

15

Table 1 Material information. Material

Product name

Composition

& lot number Light cured

GC Kalore

Matrix (18 wt%): urethane dimethacrylate,

universal composite

Shade A2

dimethacrylate comonomers, proprietary DX-511

Lot 1310042

monomer Fillers (82 wt%): Fluoroaluminosilicate glass, strontium glass, pre-polymerized fillers (modified strontium glass, and lanthanoid fluoride), silicon dioxide Others: photoinitiator <1%, pigment <1%

Light cured

GC Fuji II LC

Powder: fluoro alumino silicate glass 100%

reinforced glass

capsule

Liquid: distilled water 20 - 30%, polyacrylic acid 20 -

ionomer restorative

Shade A2

30%, 2-hydroxyethylmethacrylate 30 - 35%,

Lot 1312188

urethanedimethacrylate <10%, camphorqunone <1% Capsule: 0.33g powder / 0.085ml liquid

One component

G-aniel Bond

4-Methacryloxyethyltrimellitate anhydride 5-10%,

self-etching light

Lot 1401231

phosphoric acid ester monomer 5-10 %, dimethacrylate

cured adhesive

15-20%, distilled water 15 -20%, acetone 35-40%, silicon dioxide 1-5%, trace of photo initiator

16

Table 2 Flexural strength (mean ± standard deviation) and fracture mode classification. Different superscript letters indicate significant differences inflexural strength values among groups (oneway ANOVA followed by Student-Newman-Keuls post-hoc test; significance level 0.05). Flexural Groups

Failure mode (counts)

strength (MPa)

Adhesive

Cohesive

Mixed

RBC Monolithic

86.7 ± 21.8 a

-

-

-

RMGI Monolithic

52.6 ± 13.1 b

-

-

-

Substrate

Add-on

RBC undisturbed

RBC

45.7 ± 21.1 b

8

2

0

RBC cured against metal

RBC

38.7 ± 16.3 b,c

5

5

0

RBC ground & adhesive

RBC

34.1 ± 11.5 b,c

6

4

0

RMGI undisturbed

RMGI

36.2 ± 8.4 b,c

5

5

0

RMGI cured against metal

RMGI

52.7 ± 25.2 b

6

4

0

RMGI ground

RMGI

47.6 ± 20.4 b

4

6

0

RMGI ground & adhesive

RBC

21.5 ± 10.0 c

7

1

2

insert

insert

17

Fig. 1. Flow diagram showing schematic specimen fabrication of the tested groups.

18

Fig. 2. Four-point bending test configuration.

19