- Email: [email protected]
In vitro cuspal deflection and microleakage of maxillary premolars restored with novel low-shrink dental composites
Dental Materials (2005) 21, 324–335
www.intl.elsevierhealth.com/journals/dema
In vitro cuspal deflection and microleakage of maxillary premolars restored with novel low-shrink dental composites William M. Palina,*, Garry J.P. Fleminga, Henal Nathwania, F.J. Trevor Burkeb, Ros C. Randallc a
Biomaterials Unit, The University of Birmingham School of Dentistry, St Chad’s Queensway, Birmingham, B4 6NN, UK b Primary Dental Care Research Unit, University of Birmingham, UK c 3M ESPE Dental Products, St Paul, MN, USA Received 6 April 2004; received in revised form 13 May 2004; accepted 20 May 2004
KEYWORDS Cuspal deflection; Oxirane; Silorane; Polymerisation shrinkage; Microleakage
Summary Objectives: The development of novel low-shrink resin-based composites (RBCs) may offer a potential reduction in polymerisation shrinkage stresses generated at the tooth/restoration interface compared with current methacrylate RBCs. In vitro cuspal deflection and microleakage of meiso-occlusal-distal (MOD) cavities restored with experimental oxirane and silorane (EXL596 and H1) and methacrylate (Z100 and Filteke Z250) RBCs were assessed. Methods: Standardised pre-molar MOD cavities were prepared (nZ10) and restored with each material. The flexure of buccal and palatal cusps was recorded 0.1 h following irradiation utilising a differential transformer deflection gauge. Each restored tooth was subjected to a thermocycling regime and microleakage of tooth sections were assessed following 24 h immersion in 0.2% fuschin dye. The degree of conversion (DC) of each RBC material was also assessed using Fourier transform infrared (FTIR) spectroscopy following 0.1, 0.5, 1, 4, 24 and 48 h post-irradiation. Results: A total cuspal deflection was observed for EXL596 (2.5G0.9 mm) and H1 (6.0G1.8 mm) compared with Z100 (20.0G4.7 mm) and Filteke Z250 (16.5G3.3 mm) following 0.1 h. The cavities restored with EXL596 displayed significantly higher microleakage than any other RBC and H1 exhibited non-significant and significantly decreased microleakage compared with Z100 and Filteke Z250, respectively. The DC of EXL596 and H1 was significantly decreased compared with Z100 and Filteke Z250 following 0.1, 0.5 and 1 h. Significance: The ‘living’ polymerisation associated with the novel oxirane and silorane RBCs and the associated decrease in cuspal deflection may suggest a decrease in the magnitude of polymerisation shrinkage stress at the tooth/ restoration interface. The decreased DC following 0.1 h of the oxirane compared
* Corresponding author. Tel.: C44-121-2372915; fax: C44-121-2372932. E-mail address: [email protected] (W.M. Palin). 0109-5641/$ - see front matter Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Cuspal deflection of low-shrink composite filled cavities
325
with the methacrylate RBCs has provided information on the cure rate of cationic and free-radical polymerisation mechanisms, respectively. The inadequate marginal seal of cavities restored with EXL596 would preclude its use as a dental restorative. The reduction in cuspal deflection and decrease in microleakage of cavities restored with H1 compared with Filteke Z250 may be advantageous in terms of marginal integrity following placement. However, the non-significant difference in microleakage between the H1 and Z100 may only present modest decreases in the deleterious effects of shrinkage stress. Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Introduction Many investigators have reported the extent of polymerisation shrinkage of commercially available resin-based composites (RBCs) and the large numbers of publications in the dental literature indicate that the associated deleterious effects remain a major clinical concern.1–6 The majority of RBCs utilised in clinical practice are based on methacrylate chemistry and the volumetric shrinkage of commercially available methacrylate RBC materials has been reported to be in the region of 2–5%.7–10 Following placement and light irradiation, the freeradical polymerisation of methacrylate-based monomers is accompanied by the closer packing of molecules leading to bulk contraction. This can be attributed to the gelation of the resin matrix, whereby stress relaxation of the polymerising material is decreased due to increasing viscous flow of the curing monomer. Consequently, the post-gel contraction of the RBC is constricted by the strength of the adhesive bond at the tooth/ restoration interface and as a result, polymerisation shrinkage of the RBC may be manifested as shrinkage stress.11 The resultant stress may compromise the synergism between the tooth/restoration interface,1 increase the likelihood of mechanical failure,2 permit the ingress of bacteria, which may result in pulpal irritation3 or result in enamel microcrack propagation, enamel fracture4,5 and cuspal deflection.6 Consequently, decreased polymerisation shrinkage and the associated stress may only be realised if novel monomer systems are developed which do not employ methacrylatebased matrices. Previous studies have suggested a decreased polymerisation shrinkage and an associated reduction in the shrinkage stress of novel epoxy or ‘oxirane’-based RBCs compared with conventional methacrylates.12,13 The term ‘epoxy’ refers to an oxygen-containing ring molecule and the polymerisation of epoxy-based monomers proceeds by a cationic ring-opening mechanism. The decrease in resin shrinkage of epoxy-based polymers compared with the free radical polymerisation
of methacrylates is manifested by the ring-opening of epoxide groups which contain three-membered ‘oxirane’ rings.14 The decreased volumetric contraction and associated decrease in post-gel shrinkage stress of oxirane resin formulations reported in the literature has been attributed to the ‘living’ or ‘dark’ polymerisation associated with the cationic mechanism.15,16 The reactive species do not become extinguished as quickly as the free radicals contained within conventional methacrylate-based resins. Although such novel experimental RBC systems have exhibited promising mechanical properties, the slow development of flexural strength17 coupled with cytotoxicity and mutagenic concerns18 of oxirane-based resin blends for dental RBC application has prompted research into the development of an alternative class of ring-opening monomers. Guggenberger and Weinmann19 have proposed an oxirane-based resin formulation that may potentially overcome the disadvantages of previous low-shrink oxirane-based RBCs. The novel resin chemistry has been synthesised from the reaction of oxiranes and siloxane molecules and termed ‘silorane’.20 The inventors have postulated that this novel monomer system may provide RBC systems with decreased polymerisation shrinkage, increased hydrophobicity and improved biocompatibility compared with conventional methacrylatebased restoratives. Previous studies (conducted by the manufacturer) have exhibited a decrease in volumetric shrinkage21 and significantly improved marginal integrity on both enamel and dentin22 of silorane RBCs compared with methacrylate-based RBCs. A recent study has also reported the decreased mutagenic activity of silorane molecules compared with oxiranes.18 The aim of the current investigation was to assess the degree of conversion (DC) and the effect of RBC polymerisation on cuspal deflection and microleakage of standardised cavities restored with conventional methacrylate compared with experimental oxirane and siloranebased RBCs. The tested hypothesis was that, differences in polymerisation reaction of the novel oxirane and silorane RBCs would result in decreased
326
W.M. Palin et al.
cuspal deflection and increased marginal integrity compared with conventional methacrylate RBCs at the tooth/restoration interface.
Materials and methods Materials Two commercially available methacrylate RBCs (Z100 and Filteke Z250) and two experimental oxirane and silorane (EXL596 and H1) RBCs, were provided by 3M ESPE Dental Products (St Paul, MN, US). Z100 consists of a mixture of Bis-GMA and TEGDMA resins and in Filteke Z250, the majority of TEGDMA has been replaced with a blend of UDMA and Bis-EMA(6) resins. The filler constituents of the methacrylate RBCs Z100 and Filteke Z250 are similar and consist of zirconia and silica particles loaded to 84.5% weight with a particle size distribution ranging from 3.5 to 0.01 mm and an average particle size of 0.6 mm. EXL596 and H1 consist of an oxirane and silorane-based resin matrix filled to 82.4 and 74.0% weight, respectively. The specific formulation of the experimental RBCs remains proprietary to the manufacturers.
Figure 1 A schematic diagram of the MOD cavity (C) utilised in the current cuspal deflection experiment. The BPW of the approximal boxes of the cavity were cut to two thirds of the BLW (B) from the buccal to the palatal cusp. The occlusal isthmus (A) was prepared to half the BLW of the tooth and the cavity depth of the isthmus was standardised to 3.5G0.1 mm from the tip of the palatal cusp and to 1.0G0.1 mm above the ACJ at the cervical aspect of the approximal boxes.
Specimen preparation Forty extracted maxillary premolars were selected, which were free from visible hypoplastic defects or fractures. Any calculus deposits were carefully removed from the teeth which were then stored in high purity double distilled water maintained at 23G1 8C prior to cavity preparation. In an attempt to standardise the dimensions of the maxillary premolars, the maximum bucco-palatal widths (BPW) were measured with digital calipers (Dentaguage 1, Erskine Dental, Marina Del Rey, CA, US) accurate to 10 mm. Multiple comparisons of the BPW of each tooth were employed using a one-way analysis of variance (ANOVA) within a statistical computer package (SPSSw for Windowsw, SPSS Inc., Chicago, IL, US) carried out at a 95% significance level. This ensured that the mean BPW of each sample group (nZ10) varied by no more than 5%. A large meiso-occlusal-distal (MOD) cavity was prepared in each maxillary premolar in accordance with Fig. 1. Prior to the restoration of each cavity with the RBC materials under investigation, each tooth was mounted in a stainless steel mould (15 mm width, 15 mm breadth, 16 mm length) with a central cylindrical opening of 12 mm. The teeth were positioned into the centre of the mould, crown uppermost, using a mounting-jig to
maintain parallelism in all directions. The mould was filled with chemically activated orthodontic resin (Meadway Rapid Repair, Mr Dental Supplies Ltd, Surrey, UK) to a depth of 2 mm from the amelocemental junction (ACJ) and allowed to set undisturbed for 20 min. The cavities were washed with distilled water and dried thoroughly using a cold-air dryer for 30 s. Scotchbond bonding system (3M ESPE Dental Materials, St Paul, MN, US) was utilised for the methacrylate RBCs, Z100 and Filteke Z250, to achieve adhesion between the tooth and restoration interface. The acid etchant gel was applied to each cavity surface for 15 s and rinsed with distilled water for a further 10 s. The surface of the cavity was then dried with cold-air for 30 s and two layers of Scotchbond primer and adhesive was painted onto the etched surface and light irradiated from a distance of 2 mm using an XL2500 curing-light (3M ESPE Dental Materials, St Paul, MN, US) for 20 s operating at a light intensity of 689G18 mW/cm2. The MOD cavities restored with the oxirane RBC materials, EXL596 and H1, utilised different bonding agents compared with the methacrylate RBC materials due to the differences in the monomer chemistry. The two-step experimental bonding agent, EXL596b
Cuspal deflection of low-shrink composite filled cavities (3M ESPE Dental Products, St Paul, MN, US) was utilised for the adhesion of the oxirane RBC, EXL596 to the cavity surface. 35% phosphoric acid etchant gel was used to etch the surface of the cavity and was applied for 15 s and washed with distilled water for 10 s. The surface of the cavity was then dried with cold-air for 30 s and two layers of EXL596b were applied to the etched surface and light irradiated from a distance of 2 mm for 40 s with the XL2500 curing-light. A onestep experimental bonding system, H1b was utilised as the bonding agent for the oxirane RBC, H1. H1b was supplied in one applicator with three ‘blister’ type cavities containing the etchant, primer and bonding components. The bonding agent was thoroughly mixed by compressing each ‘blister’ cavity and folding the applicator to produce a transparent liquid for application to the cavity surface whereby etching, priming and bonding took place in one-step. The bonding agent was brushed onto the cavity and manipulated across the surface for 40 s. A cold-air dryer was used to dry the bonding agent for a further 20 s and the adhesive was light irradiated from a distance of 2 mm for 20 s in accordance with the procedure outlined above.
Deflection measurements Cuspal movement was measured utilising a direct current differential transformer (DCDT) (Mercer 122L Twin Channel Analogue Gauge Unit, Thomas Mercer Ltd, Hertfordshire, UK) following light irradiation. The buccal and palatal cusps of the maxillary premolars were placed contiguously against two receptors of the deflection-measuring gauge. In an attempt to achieve standardised placement of each receptor to the buccal and palatal cusps, adjustments in the horizontal plane of the receptors and vertical plane of the stainless steel mould ensured a fixed position of 2.0G0.5 mm of each receptor from the top midline of each buccal and palatal cusp. The initial RBC increment was placed at the mesial proximal box using a dental compression tool (Dentsply Ltd, Surrey, England). The increment sequence utilised the placement of eight nominally identical triangular increments; three for each proximal box and two for the occlusal surface with each increment in contact exclusively with either the buccal or palatal wall. Each increment of RBC material was light irradiated according to the manufacturers instructions, namely 40 s for Z100, 20 s for Filteke Z250 and 60 s for both EXL596 and H1, at a distance of
327
2 mm above the interproximal boxes. The deflection of each maxillary premolar cusp (mm) for each of the eight increments was measured 0.1 h following the cessation of light irradiation. Statistical analysis of the resultant cuspal deflection data was employed by using a one-way ANOVA and the comparisons of paired groups of data were also assessed using a Tukey’s multiple range test. Prior to statistical analysis, all experimental data was checked for normality and statistical tests were conducted at either 99.9% or 95.0 significance level (P!0.001 and 0.05, respectively).
Thermocycling and cervical gingival microleakage Each sample set was subjected to 500 thermocycles, where one cycle required 70 s to transfer the teeth between two waterbaths maintained at 4G1 and 65G1 8C with a dwell time in each waterbath of 5 s.23,24 Prior to the assessment of the cervical gingival microleakage of each maxillary premolar, the tooth root apices were sealed with sticky wax and the tooth surfaces were coated with nail varnish excluding a 1 mm periphery from the ACJ. In order to assess the effect of cuspal deflection and thermocycling on the cervical gingival microleakage, the restored teeth were washed and dried thoroughly and transferred to 0.2% basic fuschin dye (Hopkin and Williams, Essex, UK) at 23G1 8C for 24 h. Following immersion, the teeth were washed thoroughly with distilled water and sectioned mid-sagitally in the meiso-distal plane using an Isomet Low-Speed Saw with a diamond cutting wheel (Buehler, Lake Bluff, IL, US). The sectioned samples were examined under a stereo-microscope (Wild M3C, Heerburg, Switzerland) at X25 magnification to assess the extent of the cervical gingival microleakage. The scoring criteria outlined in Table 1 was utilised to rank the degree of microleakage and the resultant cervical gingival microleakage scores were statistically analysed using a non-parametric one-way ANOVA (Kruskal– Wallace) test followed by paired group comparisons using Mann–Whitney U-tests at a 95% significance level.
Degree of conversion The DC of each RBC under investigation was assessed following 0.1, 0.5, 1, 4, 24 and 48 h immersion in a lightproof waterbath maintained at
328
W.M. Palin et al.
Table 1 The scoring criteria utilised to rank the degree of cervical gingival microleakage of each MOD restoration following thermocycling and immersion in 0.2% basic fuschin dye for 24 h at 23G1 8C. Microleakage score
Description
0 1 2 3 4 4*
No evidence of dye penetration Superficial dye penetration up to the amelodentinal junction (ADJ) Dye penetration along the gingival floor and up to the axial wall Dye penetration along the axial wall and across the pulpal floor Dye penetration into the pulp chamber from the pulpal floor Dye penetration into the pulp chamber from the axial wall
37G1 8C utilising Fourier transform infra-red (FTIR) spectroscopy in accordance with the procedure outlined previously.17 For the methacrylate RBCs (Z100 and Filteke Z250), the ratio of peak intensities of aliphatic CaC to aromatic CaC (1635 and 1608 cmK1 , respectively) were evaluated before and after irradiation to determine the percentage of unsaturated aliphatic CaC bonds remaining in the test material. Absorption of the aromatic CaC stretching band remains constant during polymerisation (Fig. 2) and serves as
an internal standard where the DC of each methacrylate specimen was calculated according to Eq. (1) ð%C Z CÞ ðaliphatic½C Z C=aromatic½C Z CÞpolymer Z !100 ðaliphatic½C Z C=aromatic½C Z CÞmonomer (1) The monomer chemistry of the oxirane and silorane RBCs do not contain aliphatic CaC groups,
Figure 2 FTIR spectra exhibiting (a) the reduction in peak height at 1635 cmK1 associated with saturation of aliphatic CaC within the methacrylate RBCs, Z100 and Filteke Z250 and (b) following 0.1, 0.5, 1, 4, 24 and 48 h. Absorption of the aromatic CaC stretching band (1608 cmK1) remains constant during polymerisation.
Cuspal deflection of low-shrink composite filled cavities consequently, the DC cannot be calculated from Eq. (1). The mean DC of EXL596 and H1 specimens were identified through regions of the FTIR spectra between 730 and 950 cm-1 which corresponded with the oxirane ring-opening regions (Figs. 3 and 4).17 On irradiation, the oxirane peaks at 807 cmK1 for EXL596 (Fig. 3a) and 882 cm-1 for H1 specimens (Fig. 4a) decreased. A common internal standard was identified where the absorption of aromatic CaC at 1608 cmK1 remained constant during polymerisation (Figs. 3b and 4b), where the DC of each oxirane and silorane specimen was calculated according to Eq. (2) ð%–C–O–C–Þ ðoxirane ½–C–O–C–=aromatic ½C Z CÞpolymer Z !100 ðoxirane ½–C–O–C–=aromatic ½C Z CÞmonomer (2)
For each RBC under investigation, the mean DC was calculated from an average of five samples. Multiple comparisons of group means were made by utilising a one-way ANOVA and a Tukey’s multiple range test was also employed at a significance level of P!0.05.
329
Results Cuspal deflection The standardisation of the dimensions of the maxillary premolars utilised in the current investigation resulted in non-significant differences between each test group (PO0.05) (Table 2). Since there was no significant difference between the overall mean deflection for cusp type, the palatal and buccal cuspal deflection data for each of the eight increments were combined to give an overall mean cuspal deflection for each RBC tested in the current study. The restoration of the MOD cavities utilising the methacrylate RBCs, Z100 and Filteke Z250 resulted in a mean total cuspal deflection and standard deviation of 20.06G4.72 and 16.52G3.27 mm, respectively. For the oxirane and silorane RBCs, EXL596 and H1 a mean total cuspal deflection of 2.52G0.86 and 6.01G1.79 mm was recorded, respectively. A one-way ANOVA revealed non-significant differences between the total cuspal deflection of the methacrylate RBC materials, Z100 and Filteke Z250 (PO0.05). In contrast, a highly significant decrease in total cuspal deflection of the teeth
Figure 3 FTIR spectra exhibiting (a) the reduction in peak height 807 cmK1 associated with the saturation of oxirane rings within the oxirane RBC, EXL596 and (b) following 0.1, 0.5, 1, 4, 24 and 48 h. Absorption of the aromatic CaC stretching band (1608 cmK1) remains constant during polymerisation.
330
W.M. Palin et al.
Figure 4 FTIR spectra exhibiting (a) the reduction in peak height 882 cmK1 associated with the saturation of oxirane rings within the silorane RBC, H1 and (b) following 0.1, 0.5, 1, 4, 24 and 48 h. Absorption of the aromatic CaC stretching band (1608 cmK1) remains constant during polymerisation.
restored with the oxirane RBC material, EXL596 was identified compared with Z100 and Filteke Z250 (P!0.001). A significant increase in cuspal deflection was identified for the MOD cavities restored with the oxirane RBC, H1 compared with EXL596 (P!0.05). However, the utilisation of H1 to restore the teeth resulted in a significant decrease in mean cuspal deflection compared with both the methacrylate RBC materials, Z100 and Filteke Z250 (P!0.05).
Microleakage All of the MOD cavities restored with either the methacrylate or the oxirane and silorane RBC materials exhibited microleakage, whereby no ‘zero’ scores were recorded (Table 1). Following statistical analysis of the microleakage score data by a non-parametric one-way ANOVA (Kruskal– Wallis) and a paired Mann–Whitney test, no significant differences in microleakage were identified
Table 2 The mean BLW widths of large MOD cavities prepared for restoration exhibiting standardisation between each group and the mean palatal and buccal cuspal deflection of each of the eight increments and the total mean cuspal deflection of cavities restored with Z100, Filteke Z250, EXL596 and H1 0.1 h following irradiation. Restorative material
Mean cavity BLW (mm)
Z100 Z250 EXL596 H1
9.56 9.30 9.37 9.49
(0.28)a (0.43)a (0.15)a (0.21)a
Mean palatal deflection (mm) 8.89 8.17 1.33 2.83
(1.53)a (1.60)a (0.51)b (1.26)b
Mean buccal deflection (mm) 11.17 8.35 1.19 3.18
(3.73)a (1.92)a (0.81)b (1.03)b
Total cuspal deflection (mm) 20.06 16.52 2.52 6.01
(4.72)a (3.27)a (0.86)c* (1.79)b
Mean values exhibiting different superscripts (within columns) were significantly different, P!0.05 and *P!0.001.
Cuspal deflection of low-shrink composite filled cavities between the cavities restored with Z100 or Filteke Z250 (PO0.05). In contrast, a significant increase in microleakage (P!0.05) was identified for the cavities restored with EXL596 compared with Z100 (PZ0.003), Filteke Z250 (PZ0.019) and H1 (PZ 0.001) (Fig. 2). On examination of the statistical data for the MOD cavities restored with H1, a nonsignificant decrease in microleakage was identified compared with Z100 (PO0.05; PZ0.315). Conversely, a significant decrease in the microleakage of MOD cavities restored with H1 was identified compared with Filteke Z250 (P!0.05; 0.035) and EXL596 (P!0.05; 0.000), respectively (Fig. 2).
Degree of conversion The DC for each RBC tested in the current investigation are presented in Table 4 and displayed in Fig. 6. A one-way ANOVA revealed non-significant differences between the DC of Z100 and Filteke Z250 following all immersion periods investigated (PO0.05). Multiple statistical comparisons resulted in a significant increase in DC of EXL596 following 4 h compared with 0.1, 0.5 and 1 h immersion, respectively (P!0.05). The DC of EXL596 specimens tested following 24 and 48 h immersion was significantly increased compared with the preceding short-term immersion periods of 0.1, 0.5, 1 and 4 h (P!0.05). The specimens of EXL596 exhibited a significant decrease in DC compared with samples of Z100 and Filteke Z250 following 0.1, 0.5 and 1 h post-irradiation, however, following 24 h, the DC of EXL596 was significantly increased compared with the DC of Z100 and Filteke Z250 (P!0.05). Statistical analysis of H1 specimens resulted in a concomitant significant increase in DC following 0.1, 0.5 to 1 h immersion, respectively (P!0.05). The specimens of H1 exhibited a significant decrease in DC compared with specimens of Z100 and Filteke Z250 following 0.1 and 0.5 h postirradiation (P!0.05), however, following 1, 4, 24 and 48 h, no significant differences were reported (PO0.05).
Discussion The effect of polymerisation shrinkage of RBC materials on the in vitro cuspal flexure of restored teeth has been reported by numerous investigators.6,25–27 The assessment of cuspal movement was quantified by a number of different measuring techniques including; deflection of a stainless steel flexible ribbon,25 strain gauges,26 DCDT6,24 and linear variable differential transformers (LVDT).27
331
It has been suggested that such contact-displacement measuring methods may provide erroneous results since a suitable reference point on the cusps was difficult to identify.27 Jantarat et al.26 investigated the differences in cuspal flexure using a DCDT and strain gauges and concluded that variation in cuspal deflection measurements between the techniques was significant and suggested that strain gauges were less sensitive to the three-dimensional orientation of the teeth compared with DCDTs. Excluding the differences in measurement technique, discrepancies in cuspal movement values reported by the authors should have been expected since the cavity preparations were subject to dimensional variation without an appropriate standardisation technique. In an attempt to provide a suitable reference point for the DCDT probes, small cavities were cut on the outer surface of the cusps which may have resulted in considerable variation in cuspal flexure data due to the possible introduction of microcracks within the enamel. The current investigation employed standardised teeth with differences in the mean BPW being a maximum of 5% between each sample set. The reproducible placement of each DCDT probe against the buccal and palatal cusp associated with the three-dimensional orientation of each tooth in the present study also allowed for a nominally identical reference point for each deflection measurement, thereby decreasing variation in the cuspal deflection data. Tantbirojn et al.4 analysed the cuspal deformation pattern of the occlusal portion of molars for various cavity types and sizes using digital profilometry, where the digitised data of the unrestored and restored tooth were used to calculate the inward movement of each cusp to a precision of 2 mm. For large Class II MOD cavities, similar to the cavities utilised in the current investigation, a mean total cuspal deformation and standard deviation of 22.49G5.67 mm was recorded for cavities restored with Z100. This was in close agreement with the cuspal deflection of the large MOD cavities restored with Z100 in the present investigation (20.06G 4.72 mm; Table 2). The preparation of large MOD cavities from upper premolars in the current study was designed to weaken the remaining tooth structure to favour potential cuspal movement. It might be argued that the weakening of the palatal and buccal cusps through the preparation of large MOD cavities in the current study was not clinically relevant since the MOD cavities may be too extensive for direct composite fillings. However, the number of RBC restorations currently placed in clinical practice has increased since the introduction of improved resin chemistry, filler morphology and the associated
332 adhesive systems of modern RBC materials.28 Moreover, the toxicity and aesthetic concerns of amalgam and the increased chairside procedure time and cost of indirect restorations has justified the increased use of RBC materials for large restorations29 such as the MOD cavities utilised in the current study. The magnitude of cuspal deflection is dependent on several factors namely, the size and configuration of the cavity4 and the mechanical-physical properties of the restorative material and the bonding system.25 Consequently, in addition to the BLW standardisation of the cavities, the incremental packing of the methacrylate and oxirane RBC materials and the use of the associated adhesive systems were carefully administered by one operator to each cavity in the current study. The deflection of the cusps through light irradiation of the restorative RBC material will only occur if there is sufficient resistance to polymerisation shrinkage associated with the adhesive properties at the tooth/restoration interface. In the current investigation, each cavity with each RBC type exhibited cuspal deflection (Table 2) which suggests that adhesion at the tooth/restoration interface was established. The significant increase in cuspal deflection of cavities restored with the methacrylate (Z100 and Filteke Z250) compared with the oxirane and silorane (EXL596 and H1) RBCs (Table 2) may have been expected due to the differences in polymerisation reaction between the free-radical and cationic species, respectively. The utilisation of FTIR spectroscopy to assess DC provided further information for characterising the differences in polymerisation of EXL596 and H1 compared with Z100 and Filteke Z250. The significant decrease in DC of EXL596 following 0.1, 0.5 and 1 h and specimens of H1 following 0.1 and 0.5 h compared with Z100 and Filteke Z250 were attributed to the differences in monomer chemistry (cationic versus free-radical polymerisation) and manifested as a decrease in cuspal deflection following 0.1 h. On irradiation of EXL596 and H1, fragmentation of the photoinitiator arises to generate a ‘super-acid’ catalyst with oxonium ions as the reactive species, which subsequently protonates the functional group of the oxirane molecule.14,30 Following molecular rearrangement, the positively charged species proceeds in three dimensions to form a tightly cross-linked network.14 The cationic reaction has been previously referred to as a ‘living’ polymerisation15,16 in that the reactive species of EXL596 and H1 do not become extinguished as rapidly as the free radicals throughout the polymerisation of Z100 and Filteke Z250. The decreased polymerisation
W.M. Palin et al. kinetics of the oxirane compared with the methacrylate-based monomers generated a temporary excess of free volume that enhanced the mobility of the polymer chains within the system and, as a result, the polymerisation efficiency of the cationic ring-opening monomers compared with the freeradical species was increased. It was suggested that the significant increase in DC of EXL596 compared with Z100 and Filteke Z250 following 24 h immersion was attributed to the increased mobility of the reactive species within EXL596 allowing the system to reach an increased DC compared with the limited mobility of free radicals present in the matrices of the methacrylate RBCs.31 It was also postulated that the significant increase in the DC of H1 compared with EXL596 following 0.1, 0.5 and 1 h immersion was attributed to the increase in the cationic polymerisation rate of H1 associated the incorporation of siloxane groups into the oxiranebased monomer resin of H1.32,33 As the reactivity of the oxirane-based monomer of H1 was significantly increased compared with EXL596, the mobility of the propagating HC ions was significantly reduced, resulting in a significant decrease in the DC of specimens of H1 following 24 and 48 h. The associated ‘living’ nature of the cationic polymerisation identified by FTIR analysis may be manifested as an increased stress relaxation of the polymerising RBC and the associated decrease in cuspal flexure compared with the cavities restored with Z100 and Filteke Z250. However, the effect of polymerisation shrinkage associated with differences in polymerisation mechanism between the free-radical and cationic resins cannot be directly related to the magnitude of shrinkage stress at the tooth/restoration interface since stress is not Table 3 The scores of gingival microleakage for ten large MOD cavities restored with the methacrylate (Z100 and Filteke Z250) and oxirane (EXL596 and H1) RBCs following cuspal deflection and thermocycling. Tooth number 1 2 3 4 5 6 7 8 9 10
Degree of microleakage Z100
Z250
EXL596
H1
1 2 3 2 2 3 3 1 1 2
2 2 3 3 3 1 3 2 3 2
2 4* 3 4* 3 3 3 4* 3 4*
3 1 1 1 1 2 2 1 2 2
The scoring criteria are outlined in Table 1.
Cuspal deflection of low-shrink composite filled cavities
Figure 5 A box and whisker plot of the microleakage scores following cuspal deflection and thermocycling of the MOD cavities restored with each RBC under investigation. The plot illustrates a summary of the microleakage scores based on the median, quartiles, and extreme values. The box represents the interquartile range which contains the 50% of values, the whiskers represent the highest and lowest microleakage values and the bold black line across the box indicates the median microleakage score.
a characteristic property of the material. The stress developed at the tooth/restoration interface remains responsible for the deleterious effects of polymerisation shrinkage in vivo and may only be derived from a combination of material properties, restoration geometry and interfacial adhesive quality of the tooth and filling material.34 Moreover, if the synergism at the tooth/restoration interface withstands the residual shrinkage stress on light irradiation, it may be adversely influenced by thermal stresses induced by fluctuations in the intraoral temperatures. Consequently, to qualify the differences in polymerisation shrinkage of the methacrylate and oxirane RBC materials with respect to the associated polymerisation shrinkage stress, the current investigation endeavoured to assess the effect of a thermocycling regime on the extent of microleakage following cuspal deflection.
333
All of the MOD cavities restored with either the methacrylate or the oxirane and silorane RBC materials exhibited microleakage, whereby dye penetration was observed (Table 3). It was therefore proposed that no complete marginal seal between the tooth structure and the RBC material was obtained for any of the cavities restored in the current investigation. The non-significant differences in microleakage of cavities restored with either Z100 and Filteke Z250 may be associated with the similarities in the methacrylate chemistry and the utilisation of the Scotchbond adhesive system for Z100 and Filteke Z250. This may suggest that the replacement of the majority of the TEGDMA with a blend of BisEMA and UDMA in the resin constituents of Filteke Z250 did not significantly reduce the polymerisation shrinkage stress at the tooth/restoration interface compared with its predecessor, Z100. The current investigation also indicated a significant increase in microleakage of the cavities restored with EXL596 compared with Z100, Filteke Z250 and H1 (Fig. 5) even though the associated cuspal deflection was significantly reduced (Table 2). This may be attributed to the significantly decreased polymerisation rate of EXL596 manifested as a decrease in DC up to 24 h following initial irradiation (Table 4; Fig. 6) associated with the inherent cationic polymerisation of the monomer system. The significantly decreased DC of EXL596 following 0.1 h (Fig. 6) may have allowed for an increased uptake of fuschin dye. Alternatively, the thermal stresses encountered following thermocycling resulting in failure of the adhesive system at the interface of the tooth and restoration may have contributed to the increase in microleakage. Since the cavities restored with EXL596 were identified as producing significantly increased gingival microleakage at the cervical dentine cavosurface margin, associated with a microleakage score of 4* (Table 3), failure of the adhesive system at
Table 4 The degree of conversion of each material utilised in the present investigation following immersion in a lightproof waterbath maintained at 37G1 8C for increasing post-irradiation periods. Immersion time (h)
Degree of conversion (%) 0.1
Z100 Z250 EXL596 H1
56.1 48.7 19.1 31.4
0.5 a
(5.2) 1 (5.6)a1 (4.2)c1 (2.2)b1
54.4 51.4 28.4 42.1
1 a
(5.5) 1 (2.1)a1 (5.7)c1,2 (1.9)b2
54.1 54.1 39.4 51.4
4 a
(5.3) 1 (6.1)a1 (5.7)b2,3 (3.6)a3
57.8 55.3 50.7 52.8
24 a
(3.2) 1 (6.0)a1 (5.3)a3 (3.3)a3
61.1 57.8 68.7 50.2
48 b
(3.6) 1 (2.6)b1 (2.2)a4 (2.2)c3
65.0 59.2 62.2 48.4
(7.0)a1 (4.3)a1 (2.4)a4 (3.2)b3
Mean values exhibiting different superscript letters (within columns) and different numbers (within rows) were significantly different (P!0.05).
334
W.M. Palin et al. contrast to total (pre- and post-gel) volumetric contraction. Therefore, polymerisation shrinkage stress is not only dependent upon the volumetric shrinkage of the restorative material but also the nature of the interfacial bond between the RBC restorative and the associated tooth structure.
Conclusion Figure 6 The DC of each material under investigation following 0.1, 0.5, 1, 4, 24 and 48 h post-irradiation exhibiting the differences in monomer conversion of freeradical and cationic polymerisation for the methacrylate and oxirane RBCs, respectively.
the tooth/restoration interface may provide a more reasonable explanation. The significant decrease in microleakage of the cavities restored with H1 compared with Filteke Z250 can be attributed to the significant decrease in cuspal deflection (6.01G 1.79 and 16.52G3.27 mm, respectively; Table 2) manifested as a reduction in polymerisation shrinkage stress at the tooth/restoration interface. Although the cuspal deflection of cavities restored with EXL596 was significantly decreased compared with H1 (2.52G0.86 and 6.01G1.79 mm, respectively; Table 2), it was suggested that partial failure of the adhesive system, EXL596b reduced the restrictive stresses within the polymerising resin matrix of EXL596 compared with the increased synergism of H1b. This postulation was further justified by a significant decrease in the associated microleakage of cavities restored with H1 compared with EXL596 (Fig. 5). From a clinical perspective, it was proposed that the significant reduction in the cuspal deflection of cavities restored with H1 and the apparent increase in synergism of H1b at the tooth/restoration interface compared with Filteke Z250 might be advantageous in terms of clinical longevity. However, the non-significant differences in cervical gingival microleakage between the H1 and Z100 may suggest only a modest decrease in the possible deleterious effects associated with the polymerisation shrinkage stress of the novel silorane RBC, H1. It may be speculated that a decrease in polymerisation shrinkage stress and a reduction in the associated deleterious effects, such as microleakage, was manifested as a significant decrease in the polymerisation shrinkage of H1 compared with Z100. However, the non-significant differences in cervical gingival microleakage between the H1 and Z100 further highlights the clinical relevance of evaluating the post-gel shrinkage of RBC specimens in
The potential for the cationic polymerisation of novel oxirane and silorane RBCs to decrease the magnitude of polymerisation shrinkage stress at the tooth/restoration interface compared with the free-radical polymerisation of conventional methacrylate RBCs may be realised. However, in the current study, failure of the oxirane adhesive system (EXL596b) and the associated inadequate marginal seal at the tooth/restoration interface would preclude its use as a dental restorative. The significant decrease in microleakage of the cavities restored with the experimental silorane (H1) compared with Filteke Z250 was attributed to the significant decrease in cuspal deflection and manifested as a reduction in polymerisation shrinkage stress at the tooth/restoration interface. The nonsignificant differences in the cervical gingival microleakage between H1 and Z100 may only present modest decreases in the possible deleterious effects associated with the shrinkage stress of current RBC formulations.
References [1] Davidson CL, DeGee AJ, Feilzer AJ. The competition between the composite-dentine bond strength and the polymerisation contraction stress. J Dent Res 1984;63: 1396–9. [2] Sakaguchi RL, Peters MC, Nelson SR, Douglas WH, Poort HW. Effects of polymerisation contractions in composite restorations. J Dent 1992;20:178–82. [3] Lutz F, Kreici I, Barbakow F. Quality and durability of marginal adaptation in bonded composite restorations. Dent Mat 1991;7:107–13. [4] Tantbirojn D, Versluis A, Pintado MR, DeLong R, Douglas WH. Tooth deformation patterns in molars after composite restoration. J Dent 2004;20:535–42. [5] Bausch JR, De Lange K, Davidson CR, Peters A, De Gee AJ. Clinical significance of polymerisation shrinkage of composite resins. J Pros Dent 1982;48:59–67. [6] Abbas G, Fleming GJP, Harrington E, Shortall ACC, Burke FJT. Cuspal movement and microleakage in premolar teeth restored with a packable composite cured in bulk or in increments. J Dent 2003;31:437–44. [7] Hofmann N, Denner W, Hugo B, Klaiber B. The influence of plasma arc vs. halogen standard or soft-start irradiation on polymerization shrinkage kinetics of polymer matrix composites. J Dent 2003;31:383–93.
Cuspal deflection of low-shrink composite filled cavities [8] Dennison JB, Yaman P, Seir R, Hamilton JC. Effect of variable light intensity on composite shrinkage. J Pros Dent 2000;84:499–505. [9] Labella R, Lambrechts P, Van Meerbeek B, Vanherle G. Polymerisation shrinkage and elasticity of flowable composites and filled adhesives. Dent Mat 1999;15:128–37. [10] Watts DC, Hindi AA. Intrinsic soft-start polymerisation shrinkage-kinetics in an acrylate-based resin composite. Dent Mat 1999;15:39–45. [11] Davidson CL, Feilzer AJ. Polymerisation shrinkage and polymerisation shrinkage stress in polymer-based restoratives. J Dent 1997;25:435–40. [12] Condon JR, Ferracane JL. The effect of composite formulation on polymerisation stress. J Am Dent Ass 2000; 131:497–503. [13] Kaisaki DA, Mitra SB, Schultz WJ, DeVoe RJ. Dental visiblelight curing epoxy system with enhanced depth of cure. 3M Innovative Properties Company. US Patent: 5, 856, 373; 1999 [14] Odian G. Ring-opening polymerisation-principles of polymerisation. New York: Wiley; 1991. [15] Tilbrook DA, Clarke RL, Howle NE, Braden N. Photocurable epoxy-polyol matrices for use in dental composites. Biomat 2000;21:1743–53. [16] Millich F, Jeang L, Eick JD, Chappelow CC, Pinzino CS. Elements of light cured epoxy-based dental polymer systems. J Dent Res 1998;77:603–8. [17] Palin WM, Fleming GJP, Burke FJT, Marquis PM, Randall RC. Monomer conversion versus flexure strength of a novel dental composite. J Dent 2003;31:547–9. [18] Schweikl H, Schmalz G, Weinmann W. Mutagenic activity of structurally related oxiranes and siloranes in Salmonella typhimurium. Mut Res 2002;521:19–27. [19] Guggenberger R, Weinmann W. Exploring beyond methacrylates. Am J Dent 2000;13:82D–84D. [20] Patent number WO98/22521; 3M ESPE Dental Products, Seefeld, Germany. [21] Weinmann W, Luchterhandt T, Guggenberger R, Stippschild A, Then S, Dede K. Comparative testing of volumetric shrinkage and sealing of silorane and methacrylate filling materials. J Dent Res 2002;81. Abstr. No. 3382 (http://iadr.confex.com/iadr/2002SanDiego/techprogram/abstract_20711.htm).
335
[22] Thalacker C, Heumann A, Weinmann W, Guggenberger R, Luchterhandt T, Syrek A. Marginal integrity of class V silorane and methacrylate composite restorations. J Dent Res 2004. Abstr. No. 1364 (http//iadr.confex.com/iadr/ 2004Hawaii/techprogram/abstract_45046.htm). [23] Spierings TM, Peters MB, Bosman F, Plasschaert AM. Verification of theoretical modelling of heat transmission in teeth by in vivo experiments. J Dent Res 1987;66: 1336–9. [24] Palmer DS, Barco MT, Billy EJ. Temperature extremes produced orally by hot and cold liquids. J Pros Dent 1992; 67:325–7. [25] McCullock AJ, Smith BGN. In vitro studies of cuspal movement produced by adhesive restorative materials. Br Dent J 1986;161:405–9. [26] Jantarat J, Panitvisai P, Palamara JEA, Messer HH. Comparison of methods for measuring cuspal deflection of teeth. J Dent 2001;29:75–82. [27] Meredith N, Setchell DJ. In vitro measurement of cuspal strain and displacement in composite restored teeth. J Dent 1997;25:331–7. [28] Allen EP, Bayne SC, Brodine AH, Cronin RJ, Donovan TE, Kois JC, Summitt JB. Annual review of selected dental literature: report of the committee on scientific investigation of the American academy of restorative dentistry. J Pros Dent 2003;90:50–80. [29] Christensen GJ. Curing restorative resin. a significant controversy. J Am Dent Ass 2000;131:1067–9. [30] Crivello JV. Cationic polymerisation of iodonium and sulfonium salt photoinitiators. Adv Poly Sci 1984;62: 1–48. [31] Andrzejewska E. Photopolymerisation kinetics of multifunctional monomers. Prog Poly Sci 2001;26:605–65. [32] Decker C. Photoinitiated cross-linking polymerisation. Prog Poly Sci 1996;4:593–650. [33] Crivello JV, Lee JL. The synthesis, characterisation, and photoinitiated cationic polymerisation of silicon-containing epoxy resins. J Poly Sci: Poly Chem 1990;28:479–503. [34] Versluis A, Douglas WH, Cross M, Sakaguchi RL. Does an incremental filling technique reduce polymerisation shrinkage stresses? J Dent Res 1996;75:871–8.
www.intl.elsevierhealth.com/journals/dema
In vitro cuspal deflection and microleakage of maxillary premolars restored with novel low-shrink dental composites William M. Palina,*, Garry J.P. Fleminga, Henal Nathwania, F.J. Trevor Burkeb, Ros C. Randallc a
Biomaterials Unit, The University of Birmingham School of Dentistry, St Chad’s Queensway, Birmingham, B4 6NN, UK b Primary Dental Care Research Unit, University of Birmingham, UK c 3M ESPE Dental Products, St Paul, MN, USA Received 6 April 2004; received in revised form 13 May 2004; accepted 20 May 2004
KEYWORDS Cuspal deflection; Oxirane; Silorane; Polymerisation shrinkage; Microleakage
Summary Objectives: The development of novel low-shrink resin-based composites (RBCs) may offer a potential reduction in polymerisation shrinkage stresses generated at the tooth/restoration interface compared with current methacrylate RBCs. In vitro cuspal deflection and microleakage of meiso-occlusal-distal (MOD) cavities restored with experimental oxirane and silorane (EXL596 and H1) and methacrylate (Z100 and Filteke Z250) RBCs were assessed. Methods: Standardised pre-molar MOD cavities were prepared (nZ10) and restored with each material. The flexure of buccal and palatal cusps was recorded 0.1 h following irradiation utilising a differential transformer deflection gauge. Each restored tooth was subjected to a thermocycling regime and microleakage of tooth sections were assessed following 24 h immersion in 0.2% fuschin dye. The degree of conversion (DC) of each RBC material was also assessed using Fourier transform infrared (FTIR) spectroscopy following 0.1, 0.5, 1, 4, 24 and 48 h post-irradiation. Results: A total cuspal deflection was observed for EXL596 (2.5G0.9 mm) and H1 (6.0G1.8 mm) compared with Z100 (20.0G4.7 mm) and Filteke Z250 (16.5G3.3 mm) following 0.1 h. The cavities restored with EXL596 displayed significantly higher microleakage than any other RBC and H1 exhibited non-significant and significantly decreased microleakage compared with Z100 and Filteke Z250, respectively. The DC of EXL596 and H1 was significantly decreased compared with Z100 and Filteke Z250 following 0.1, 0.5 and 1 h. Significance: The ‘living’ polymerisation associated with the novel oxirane and silorane RBCs and the associated decrease in cuspal deflection may suggest a decrease in the magnitude of polymerisation shrinkage stress at the tooth/ restoration interface. The decreased DC following 0.1 h of the oxirane compared
* Corresponding author. Tel.: C44-121-2372915; fax: C44-121-2372932. E-mail address: [email protected] (W.M. Palin). 0109-5641/$ - see front matter Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Cuspal deflection of low-shrink composite filled cavities
325
with the methacrylate RBCs has provided information on the cure rate of cationic and free-radical polymerisation mechanisms, respectively. The inadequate marginal seal of cavities restored with EXL596 would preclude its use as a dental restorative. The reduction in cuspal deflection and decrease in microleakage of cavities restored with H1 compared with Filteke Z250 may be advantageous in terms of marginal integrity following placement. However, the non-significant difference in microleakage between the H1 and Z100 may only present modest decreases in the deleterious effects of shrinkage stress. Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Introduction Many investigators have reported the extent of polymerisation shrinkage of commercially available resin-based composites (RBCs) and the large numbers of publications in the dental literature indicate that the associated deleterious effects remain a major clinical concern.1–6 The majority of RBCs utilised in clinical practice are based on methacrylate chemistry and the volumetric shrinkage of commercially available methacrylate RBC materials has been reported to be in the region of 2–5%.7–10 Following placement and light irradiation, the freeradical polymerisation of methacrylate-based monomers is accompanied by the closer packing of molecules leading to bulk contraction. This can be attributed to the gelation of the resin matrix, whereby stress relaxation of the polymerising material is decreased due to increasing viscous flow of the curing monomer. Consequently, the post-gel contraction of the RBC is constricted by the strength of the adhesive bond at the tooth/ restoration interface and as a result, polymerisation shrinkage of the RBC may be manifested as shrinkage stress.11 The resultant stress may compromise the synergism between the tooth/restoration interface,1 increase the likelihood of mechanical failure,2 permit the ingress of bacteria, which may result in pulpal irritation3 or result in enamel microcrack propagation, enamel fracture4,5 and cuspal deflection.6 Consequently, decreased polymerisation shrinkage and the associated stress may only be realised if novel monomer systems are developed which do not employ methacrylatebased matrices. Previous studies have suggested a decreased polymerisation shrinkage and an associated reduction in the shrinkage stress of novel epoxy or ‘oxirane’-based RBCs compared with conventional methacrylates.12,13 The term ‘epoxy’ refers to an oxygen-containing ring molecule and the polymerisation of epoxy-based monomers proceeds by a cationic ring-opening mechanism. The decrease in resin shrinkage of epoxy-based polymers compared with the free radical polymerisation
of methacrylates is manifested by the ring-opening of epoxide groups which contain three-membered ‘oxirane’ rings.14 The decreased volumetric contraction and associated decrease in post-gel shrinkage stress of oxirane resin formulations reported in the literature has been attributed to the ‘living’ or ‘dark’ polymerisation associated with the cationic mechanism.15,16 The reactive species do not become extinguished as quickly as the free radicals contained within conventional methacrylate-based resins. Although such novel experimental RBC systems have exhibited promising mechanical properties, the slow development of flexural strength17 coupled with cytotoxicity and mutagenic concerns18 of oxirane-based resin blends for dental RBC application has prompted research into the development of an alternative class of ring-opening monomers. Guggenberger and Weinmann19 have proposed an oxirane-based resin formulation that may potentially overcome the disadvantages of previous low-shrink oxirane-based RBCs. The novel resin chemistry has been synthesised from the reaction of oxiranes and siloxane molecules and termed ‘silorane’.20 The inventors have postulated that this novel monomer system may provide RBC systems with decreased polymerisation shrinkage, increased hydrophobicity and improved biocompatibility compared with conventional methacrylatebased restoratives. Previous studies (conducted by the manufacturer) have exhibited a decrease in volumetric shrinkage21 and significantly improved marginal integrity on both enamel and dentin22 of silorane RBCs compared with methacrylate-based RBCs. A recent study has also reported the decreased mutagenic activity of silorane molecules compared with oxiranes.18 The aim of the current investigation was to assess the degree of conversion (DC) and the effect of RBC polymerisation on cuspal deflection and microleakage of standardised cavities restored with conventional methacrylate compared with experimental oxirane and siloranebased RBCs. The tested hypothesis was that, differences in polymerisation reaction of the novel oxirane and silorane RBCs would result in decreased
326
W.M. Palin et al.
cuspal deflection and increased marginal integrity compared with conventional methacrylate RBCs at the tooth/restoration interface.
Materials and methods Materials Two commercially available methacrylate RBCs (Z100 and Filteke Z250) and two experimental oxirane and silorane (EXL596 and H1) RBCs, were provided by 3M ESPE Dental Products (St Paul, MN, US). Z100 consists of a mixture of Bis-GMA and TEGDMA resins and in Filteke Z250, the majority of TEGDMA has been replaced with a blend of UDMA and Bis-EMA(6) resins. The filler constituents of the methacrylate RBCs Z100 and Filteke Z250 are similar and consist of zirconia and silica particles loaded to 84.5% weight with a particle size distribution ranging from 3.5 to 0.01 mm and an average particle size of 0.6 mm. EXL596 and H1 consist of an oxirane and silorane-based resin matrix filled to 82.4 and 74.0% weight, respectively. The specific formulation of the experimental RBCs remains proprietary to the manufacturers.
Figure 1 A schematic diagram of the MOD cavity (C) utilised in the current cuspal deflection experiment. The BPW of the approximal boxes of the cavity were cut to two thirds of the BLW (B) from the buccal to the palatal cusp. The occlusal isthmus (A) was prepared to half the BLW of the tooth and the cavity depth of the isthmus was standardised to 3.5G0.1 mm from the tip of the palatal cusp and to 1.0G0.1 mm above the ACJ at the cervical aspect of the approximal boxes.
Specimen preparation Forty extracted maxillary premolars were selected, which were free from visible hypoplastic defects or fractures. Any calculus deposits were carefully removed from the teeth which were then stored in high purity double distilled water maintained at 23G1 8C prior to cavity preparation. In an attempt to standardise the dimensions of the maxillary premolars, the maximum bucco-palatal widths (BPW) were measured with digital calipers (Dentaguage 1, Erskine Dental, Marina Del Rey, CA, US) accurate to 10 mm. Multiple comparisons of the BPW of each tooth were employed using a one-way analysis of variance (ANOVA) within a statistical computer package (SPSSw for Windowsw, SPSS Inc., Chicago, IL, US) carried out at a 95% significance level. This ensured that the mean BPW of each sample group (nZ10) varied by no more than 5%. A large meiso-occlusal-distal (MOD) cavity was prepared in each maxillary premolar in accordance with Fig. 1. Prior to the restoration of each cavity with the RBC materials under investigation, each tooth was mounted in a stainless steel mould (15 mm width, 15 mm breadth, 16 mm length) with a central cylindrical opening of 12 mm. The teeth were positioned into the centre of the mould, crown uppermost, using a mounting-jig to
maintain parallelism in all directions. The mould was filled with chemically activated orthodontic resin (Meadway Rapid Repair, Mr Dental Supplies Ltd, Surrey, UK) to a depth of 2 mm from the amelocemental junction (ACJ) and allowed to set undisturbed for 20 min. The cavities were washed with distilled water and dried thoroughly using a cold-air dryer for 30 s. Scotchbond bonding system (3M ESPE Dental Materials, St Paul, MN, US) was utilised for the methacrylate RBCs, Z100 and Filteke Z250, to achieve adhesion between the tooth and restoration interface. The acid etchant gel was applied to each cavity surface for 15 s and rinsed with distilled water for a further 10 s. The surface of the cavity was then dried with cold-air for 30 s and two layers of Scotchbond primer and adhesive was painted onto the etched surface and light irradiated from a distance of 2 mm using an XL2500 curing-light (3M ESPE Dental Materials, St Paul, MN, US) for 20 s operating at a light intensity of 689G18 mW/cm2. The MOD cavities restored with the oxirane RBC materials, EXL596 and H1, utilised different bonding agents compared with the methacrylate RBC materials due to the differences in the monomer chemistry. The two-step experimental bonding agent, EXL596b
Cuspal deflection of low-shrink composite filled cavities (3M ESPE Dental Products, St Paul, MN, US) was utilised for the adhesion of the oxirane RBC, EXL596 to the cavity surface. 35% phosphoric acid etchant gel was used to etch the surface of the cavity and was applied for 15 s and washed with distilled water for 10 s. The surface of the cavity was then dried with cold-air for 30 s and two layers of EXL596b were applied to the etched surface and light irradiated from a distance of 2 mm for 40 s with the XL2500 curing-light. A onestep experimental bonding system, H1b was utilised as the bonding agent for the oxirane RBC, H1. H1b was supplied in one applicator with three ‘blister’ type cavities containing the etchant, primer and bonding components. The bonding agent was thoroughly mixed by compressing each ‘blister’ cavity and folding the applicator to produce a transparent liquid for application to the cavity surface whereby etching, priming and bonding took place in one-step. The bonding agent was brushed onto the cavity and manipulated across the surface for 40 s. A cold-air dryer was used to dry the bonding agent for a further 20 s and the adhesive was light irradiated from a distance of 2 mm for 20 s in accordance with the procedure outlined above.
Deflection measurements Cuspal movement was measured utilising a direct current differential transformer (DCDT) (Mercer 122L Twin Channel Analogue Gauge Unit, Thomas Mercer Ltd, Hertfordshire, UK) following light irradiation. The buccal and palatal cusps of the maxillary premolars were placed contiguously against two receptors of the deflection-measuring gauge. In an attempt to achieve standardised placement of each receptor to the buccal and palatal cusps, adjustments in the horizontal plane of the receptors and vertical plane of the stainless steel mould ensured a fixed position of 2.0G0.5 mm of each receptor from the top midline of each buccal and palatal cusp. The initial RBC increment was placed at the mesial proximal box using a dental compression tool (Dentsply Ltd, Surrey, England). The increment sequence utilised the placement of eight nominally identical triangular increments; three for each proximal box and two for the occlusal surface with each increment in contact exclusively with either the buccal or palatal wall. Each increment of RBC material was light irradiated according to the manufacturers instructions, namely 40 s for Z100, 20 s for Filteke Z250 and 60 s for both EXL596 and H1, at a distance of
327
2 mm above the interproximal boxes. The deflection of each maxillary premolar cusp (mm) for each of the eight increments was measured 0.1 h following the cessation of light irradiation. Statistical analysis of the resultant cuspal deflection data was employed by using a one-way ANOVA and the comparisons of paired groups of data were also assessed using a Tukey’s multiple range test. Prior to statistical analysis, all experimental data was checked for normality and statistical tests were conducted at either 99.9% or 95.0 significance level (P!0.001 and 0.05, respectively).
Thermocycling and cervical gingival microleakage Each sample set was subjected to 500 thermocycles, where one cycle required 70 s to transfer the teeth between two waterbaths maintained at 4G1 and 65G1 8C with a dwell time in each waterbath of 5 s.23,24 Prior to the assessment of the cervical gingival microleakage of each maxillary premolar, the tooth root apices were sealed with sticky wax and the tooth surfaces were coated with nail varnish excluding a 1 mm periphery from the ACJ. In order to assess the effect of cuspal deflection and thermocycling on the cervical gingival microleakage, the restored teeth were washed and dried thoroughly and transferred to 0.2% basic fuschin dye (Hopkin and Williams, Essex, UK) at 23G1 8C for 24 h. Following immersion, the teeth were washed thoroughly with distilled water and sectioned mid-sagitally in the meiso-distal plane using an Isomet Low-Speed Saw with a diamond cutting wheel (Buehler, Lake Bluff, IL, US). The sectioned samples were examined under a stereo-microscope (Wild M3C, Heerburg, Switzerland) at X25 magnification to assess the extent of the cervical gingival microleakage. The scoring criteria outlined in Table 1 was utilised to rank the degree of microleakage and the resultant cervical gingival microleakage scores were statistically analysed using a non-parametric one-way ANOVA (Kruskal– Wallace) test followed by paired group comparisons using Mann–Whitney U-tests at a 95% significance level.
Degree of conversion The DC of each RBC under investigation was assessed following 0.1, 0.5, 1, 4, 24 and 48 h immersion in a lightproof waterbath maintained at
328
W.M. Palin et al.
Table 1 The scoring criteria utilised to rank the degree of cervical gingival microleakage of each MOD restoration following thermocycling and immersion in 0.2% basic fuschin dye for 24 h at 23G1 8C. Microleakage score
Description
0 1 2 3 4 4*
No evidence of dye penetration Superficial dye penetration up to the amelodentinal junction (ADJ) Dye penetration along the gingival floor and up to the axial wall Dye penetration along the axial wall and across the pulpal floor Dye penetration into the pulp chamber from the pulpal floor Dye penetration into the pulp chamber from the axial wall
37G1 8C utilising Fourier transform infra-red (FTIR) spectroscopy in accordance with the procedure outlined previously.17 For the methacrylate RBCs (Z100 and Filteke Z250), the ratio of peak intensities of aliphatic CaC to aromatic CaC (1635 and 1608 cmK1 , respectively) were evaluated before and after irradiation to determine the percentage of unsaturated aliphatic CaC bonds remaining in the test material. Absorption of the aromatic CaC stretching band remains constant during polymerisation (Fig. 2) and serves as
an internal standard where the DC of each methacrylate specimen was calculated according to Eq. (1) ð%C Z CÞ ðaliphatic½C Z C=aromatic½C Z CÞpolymer Z !100 ðaliphatic½C Z C=aromatic½C Z CÞmonomer (1) The monomer chemistry of the oxirane and silorane RBCs do not contain aliphatic CaC groups,
Figure 2 FTIR spectra exhibiting (a) the reduction in peak height at 1635 cmK1 associated with saturation of aliphatic CaC within the methacrylate RBCs, Z100 and Filteke Z250 and (b) following 0.1, 0.5, 1, 4, 24 and 48 h. Absorption of the aromatic CaC stretching band (1608 cmK1) remains constant during polymerisation.
Cuspal deflection of low-shrink composite filled cavities consequently, the DC cannot be calculated from Eq. (1). The mean DC of EXL596 and H1 specimens were identified through regions of the FTIR spectra between 730 and 950 cm-1 which corresponded with the oxirane ring-opening regions (Figs. 3 and 4).17 On irradiation, the oxirane peaks at 807 cmK1 for EXL596 (Fig. 3a) and 882 cm-1 for H1 specimens (Fig. 4a) decreased. A common internal standard was identified where the absorption of aromatic CaC at 1608 cmK1 remained constant during polymerisation (Figs. 3b and 4b), where the DC of each oxirane and silorane specimen was calculated according to Eq. (2) ð%–C–O–C–Þ ðoxirane ½–C–O–C–=aromatic ½C Z CÞpolymer Z !100 ðoxirane ½–C–O–C–=aromatic ½C Z CÞmonomer (2)
For each RBC under investigation, the mean DC was calculated from an average of five samples. Multiple comparisons of group means were made by utilising a one-way ANOVA and a Tukey’s multiple range test was also employed at a significance level of P!0.05.
329
Results Cuspal deflection The standardisation of the dimensions of the maxillary premolars utilised in the current investigation resulted in non-significant differences between each test group (PO0.05) (Table 2). Since there was no significant difference between the overall mean deflection for cusp type, the palatal and buccal cuspal deflection data for each of the eight increments were combined to give an overall mean cuspal deflection for each RBC tested in the current study. The restoration of the MOD cavities utilising the methacrylate RBCs, Z100 and Filteke Z250 resulted in a mean total cuspal deflection and standard deviation of 20.06G4.72 and 16.52G3.27 mm, respectively. For the oxirane and silorane RBCs, EXL596 and H1 a mean total cuspal deflection of 2.52G0.86 and 6.01G1.79 mm was recorded, respectively. A one-way ANOVA revealed non-significant differences between the total cuspal deflection of the methacrylate RBC materials, Z100 and Filteke Z250 (PO0.05). In contrast, a highly significant decrease in total cuspal deflection of the teeth
Figure 3 FTIR spectra exhibiting (a) the reduction in peak height 807 cmK1 associated with the saturation of oxirane rings within the oxirane RBC, EXL596 and (b) following 0.1, 0.5, 1, 4, 24 and 48 h. Absorption of the aromatic CaC stretching band (1608 cmK1) remains constant during polymerisation.
330
W.M. Palin et al.
Figure 4 FTIR spectra exhibiting (a) the reduction in peak height 882 cmK1 associated with the saturation of oxirane rings within the silorane RBC, H1 and (b) following 0.1, 0.5, 1, 4, 24 and 48 h. Absorption of the aromatic CaC stretching band (1608 cmK1) remains constant during polymerisation.
restored with the oxirane RBC material, EXL596 was identified compared with Z100 and Filteke Z250 (P!0.001). A significant increase in cuspal deflection was identified for the MOD cavities restored with the oxirane RBC, H1 compared with EXL596 (P!0.05). However, the utilisation of H1 to restore the teeth resulted in a significant decrease in mean cuspal deflection compared with both the methacrylate RBC materials, Z100 and Filteke Z250 (P!0.05).
Microleakage All of the MOD cavities restored with either the methacrylate or the oxirane and silorane RBC materials exhibited microleakage, whereby no ‘zero’ scores were recorded (Table 1). Following statistical analysis of the microleakage score data by a non-parametric one-way ANOVA (Kruskal– Wallis) and a paired Mann–Whitney test, no significant differences in microleakage were identified
Table 2 The mean BLW widths of large MOD cavities prepared for restoration exhibiting standardisation between each group and the mean palatal and buccal cuspal deflection of each of the eight increments and the total mean cuspal deflection of cavities restored with Z100, Filteke Z250, EXL596 and H1 0.1 h following irradiation. Restorative material
Mean cavity BLW (mm)
Z100 Z250 EXL596 H1
9.56 9.30 9.37 9.49
(0.28)a (0.43)a (0.15)a (0.21)a
Mean palatal deflection (mm) 8.89 8.17 1.33 2.83
(1.53)a (1.60)a (0.51)b (1.26)b
Mean buccal deflection (mm) 11.17 8.35 1.19 3.18
(3.73)a (1.92)a (0.81)b (1.03)b
Total cuspal deflection (mm) 20.06 16.52 2.52 6.01
(4.72)a (3.27)a (0.86)c* (1.79)b
Mean values exhibiting different superscripts (within columns) were significantly different, P!0.05 and *P!0.001.
Cuspal deflection of low-shrink composite filled cavities between the cavities restored with Z100 or Filteke Z250 (PO0.05). In contrast, a significant increase in microleakage (P!0.05) was identified for the cavities restored with EXL596 compared with Z100 (PZ0.003), Filteke Z250 (PZ0.019) and H1 (PZ 0.001) (Fig. 2). On examination of the statistical data for the MOD cavities restored with H1, a nonsignificant decrease in microleakage was identified compared with Z100 (PO0.05; PZ0.315). Conversely, a significant decrease in the microleakage of MOD cavities restored with H1 was identified compared with Filteke Z250 (P!0.05; 0.035) and EXL596 (P!0.05; 0.000), respectively (Fig. 2).
Degree of conversion The DC for each RBC tested in the current investigation are presented in Table 4 and displayed in Fig. 6. A one-way ANOVA revealed non-significant differences between the DC of Z100 and Filteke Z250 following all immersion periods investigated (PO0.05). Multiple statistical comparisons resulted in a significant increase in DC of EXL596 following 4 h compared with 0.1, 0.5 and 1 h immersion, respectively (P!0.05). The DC of EXL596 specimens tested following 24 and 48 h immersion was significantly increased compared with the preceding short-term immersion periods of 0.1, 0.5, 1 and 4 h (P!0.05). The specimens of EXL596 exhibited a significant decrease in DC compared with samples of Z100 and Filteke Z250 following 0.1, 0.5 and 1 h post-irradiation, however, following 24 h, the DC of EXL596 was significantly increased compared with the DC of Z100 and Filteke Z250 (P!0.05). Statistical analysis of H1 specimens resulted in a concomitant significant increase in DC following 0.1, 0.5 to 1 h immersion, respectively (P!0.05). The specimens of H1 exhibited a significant decrease in DC compared with specimens of Z100 and Filteke Z250 following 0.1 and 0.5 h postirradiation (P!0.05), however, following 1, 4, 24 and 48 h, no significant differences were reported (PO0.05).
Discussion The effect of polymerisation shrinkage of RBC materials on the in vitro cuspal flexure of restored teeth has been reported by numerous investigators.6,25–27 The assessment of cuspal movement was quantified by a number of different measuring techniques including; deflection of a stainless steel flexible ribbon,25 strain gauges,26 DCDT6,24 and linear variable differential transformers (LVDT).27
331
It has been suggested that such contact-displacement measuring methods may provide erroneous results since a suitable reference point on the cusps was difficult to identify.27 Jantarat et al.26 investigated the differences in cuspal flexure using a DCDT and strain gauges and concluded that variation in cuspal deflection measurements between the techniques was significant and suggested that strain gauges were less sensitive to the three-dimensional orientation of the teeth compared with DCDTs. Excluding the differences in measurement technique, discrepancies in cuspal movement values reported by the authors should have been expected since the cavity preparations were subject to dimensional variation without an appropriate standardisation technique. In an attempt to provide a suitable reference point for the DCDT probes, small cavities were cut on the outer surface of the cusps which may have resulted in considerable variation in cuspal flexure data due to the possible introduction of microcracks within the enamel. The current investigation employed standardised teeth with differences in the mean BPW being a maximum of 5% between each sample set. The reproducible placement of each DCDT probe against the buccal and palatal cusp associated with the three-dimensional orientation of each tooth in the present study also allowed for a nominally identical reference point for each deflection measurement, thereby decreasing variation in the cuspal deflection data. Tantbirojn et al.4 analysed the cuspal deformation pattern of the occlusal portion of molars for various cavity types and sizes using digital profilometry, where the digitised data of the unrestored and restored tooth were used to calculate the inward movement of each cusp to a precision of 2 mm. For large Class II MOD cavities, similar to the cavities utilised in the current investigation, a mean total cuspal deformation and standard deviation of 22.49G5.67 mm was recorded for cavities restored with Z100. This was in close agreement with the cuspal deflection of the large MOD cavities restored with Z100 in the present investigation (20.06G 4.72 mm; Table 2). The preparation of large MOD cavities from upper premolars in the current study was designed to weaken the remaining tooth structure to favour potential cuspal movement. It might be argued that the weakening of the palatal and buccal cusps through the preparation of large MOD cavities in the current study was not clinically relevant since the MOD cavities may be too extensive for direct composite fillings. However, the number of RBC restorations currently placed in clinical practice has increased since the introduction of improved resin chemistry, filler morphology and the associated
332 adhesive systems of modern RBC materials.28 Moreover, the toxicity and aesthetic concerns of amalgam and the increased chairside procedure time and cost of indirect restorations has justified the increased use of RBC materials for large restorations29 such as the MOD cavities utilised in the current study. The magnitude of cuspal deflection is dependent on several factors namely, the size and configuration of the cavity4 and the mechanical-physical properties of the restorative material and the bonding system.25 Consequently, in addition to the BLW standardisation of the cavities, the incremental packing of the methacrylate and oxirane RBC materials and the use of the associated adhesive systems were carefully administered by one operator to each cavity in the current study. The deflection of the cusps through light irradiation of the restorative RBC material will only occur if there is sufficient resistance to polymerisation shrinkage associated with the adhesive properties at the tooth/restoration interface. In the current investigation, each cavity with each RBC type exhibited cuspal deflection (Table 2) which suggests that adhesion at the tooth/restoration interface was established. The significant increase in cuspal deflection of cavities restored with the methacrylate (Z100 and Filteke Z250) compared with the oxirane and silorane (EXL596 and H1) RBCs (Table 2) may have been expected due to the differences in polymerisation reaction between the free-radical and cationic species, respectively. The utilisation of FTIR spectroscopy to assess DC provided further information for characterising the differences in polymerisation of EXL596 and H1 compared with Z100 and Filteke Z250. The significant decrease in DC of EXL596 following 0.1, 0.5 and 1 h and specimens of H1 following 0.1 and 0.5 h compared with Z100 and Filteke Z250 were attributed to the differences in monomer chemistry (cationic versus free-radical polymerisation) and manifested as a decrease in cuspal deflection following 0.1 h. On irradiation of EXL596 and H1, fragmentation of the photoinitiator arises to generate a ‘super-acid’ catalyst with oxonium ions as the reactive species, which subsequently protonates the functional group of the oxirane molecule.14,30 Following molecular rearrangement, the positively charged species proceeds in three dimensions to form a tightly cross-linked network.14 The cationic reaction has been previously referred to as a ‘living’ polymerisation15,16 in that the reactive species of EXL596 and H1 do not become extinguished as rapidly as the free radicals throughout the polymerisation of Z100 and Filteke Z250. The decreased polymerisation
W.M. Palin et al. kinetics of the oxirane compared with the methacrylate-based monomers generated a temporary excess of free volume that enhanced the mobility of the polymer chains within the system and, as a result, the polymerisation efficiency of the cationic ring-opening monomers compared with the freeradical species was increased. It was suggested that the significant increase in DC of EXL596 compared with Z100 and Filteke Z250 following 24 h immersion was attributed to the increased mobility of the reactive species within EXL596 allowing the system to reach an increased DC compared with the limited mobility of free radicals present in the matrices of the methacrylate RBCs.31 It was also postulated that the significant increase in the DC of H1 compared with EXL596 following 0.1, 0.5 and 1 h immersion was attributed to the increase in the cationic polymerisation rate of H1 associated the incorporation of siloxane groups into the oxiranebased monomer resin of H1.32,33 As the reactivity of the oxirane-based monomer of H1 was significantly increased compared with EXL596, the mobility of the propagating HC ions was significantly reduced, resulting in a significant decrease in the DC of specimens of H1 following 24 and 48 h. The associated ‘living’ nature of the cationic polymerisation identified by FTIR analysis may be manifested as an increased stress relaxation of the polymerising RBC and the associated decrease in cuspal flexure compared with the cavities restored with Z100 and Filteke Z250. However, the effect of polymerisation shrinkage associated with differences in polymerisation mechanism between the free-radical and cationic resins cannot be directly related to the magnitude of shrinkage stress at the tooth/restoration interface since stress is not Table 3 The scores of gingival microleakage for ten large MOD cavities restored with the methacrylate (Z100 and Filteke Z250) and oxirane (EXL596 and H1) RBCs following cuspal deflection and thermocycling. Tooth number 1 2 3 4 5 6 7 8 9 10
Degree of microleakage Z100
Z250
EXL596
H1
1 2 3 2 2 3 3 1 1 2
2 2 3 3 3 1 3 2 3 2
2 4* 3 4* 3 3 3 4* 3 4*
3 1 1 1 1 2 2 1 2 2
The scoring criteria are outlined in Table 1.
Cuspal deflection of low-shrink composite filled cavities
Figure 5 A box and whisker plot of the microleakage scores following cuspal deflection and thermocycling of the MOD cavities restored with each RBC under investigation. The plot illustrates a summary of the microleakage scores based on the median, quartiles, and extreme values. The box represents the interquartile range which contains the 50% of values, the whiskers represent the highest and lowest microleakage values and the bold black line across the box indicates the median microleakage score.
a characteristic property of the material. The stress developed at the tooth/restoration interface remains responsible for the deleterious effects of polymerisation shrinkage in vivo and may only be derived from a combination of material properties, restoration geometry and interfacial adhesive quality of the tooth and filling material.34 Moreover, if the synergism at the tooth/restoration interface withstands the residual shrinkage stress on light irradiation, it may be adversely influenced by thermal stresses induced by fluctuations in the intraoral temperatures. Consequently, to qualify the differences in polymerisation shrinkage of the methacrylate and oxirane RBC materials with respect to the associated polymerisation shrinkage stress, the current investigation endeavoured to assess the effect of a thermocycling regime on the extent of microleakage following cuspal deflection.
333
All of the MOD cavities restored with either the methacrylate or the oxirane and silorane RBC materials exhibited microleakage, whereby dye penetration was observed (Table 3). It was therefore proposed that no complete marginal seal between the tooth structure and the RBC material was obtained for any of the cavities restored in the current investigation. The non-significant differences in microleakage of cavities restored with either Z100 and Filteke Z250 may be associated with the similarities in the methacrylate chemistry and the utilisation of the Scotchbond adhesive system for Z100 and Filteke Z250. This may suggest that the replacement of the majority of the TEGDMA with a blend of BisEMA and UDMA in the resin constituents of Filteke Z250 did not significantly reduce the polymerisation shrinkage stress at the tooth/restoration interface compared with its predecessor, Z100. The current investigation also indicated a significant increase in microleakage of the cavities restored with EXL596 compared with Z100, Filteke Z250 and H1 (Fig. 5) even though the associated cuspal deflection was significantly reduced (Table 2). This may be attributed to the significantly decreased polymerisation rate of EXL596 manifested as a decrease in DC up to 24 h following initial irradiation (Table 4; Fig. 6) associated with the inherent cationic polymerisation of the monomer system. The significantly decreased DC of EXL596 following 0.1 h (Fig. 6) may have allowed for an increased uptake of fuschin dye. Alternatively, the thermal stresses encountered following thermocycling resulting in failure of the adhesive system at the interface of the tooth and restoration may have contributed to the increase in microleakage. Since the cavities restored with EXL596 were identified as producing significantly increased gingival microleakage at the cervical dentine cavosurface margin, associated with a microleakage score of 4* (Table 3), failure of the adhesive system at
Table 4 The degree of conversion of each material utilised in the present investigation following immersion in a lightproof waterbath maintained at 37G1 8C for increasing post-irradiation periods. Immersion time (h)
Degree of conversion (%) 0.1
Z100 Z250 EXL596 H1
56.1 48.7 19.1 31.4
0.5 a
(5.2) 1 (5.6)a1 (4.2)c1 (2.2)b1
54.4 51.4 28.4 42.1
1 a
(5.5) 1 (2.1)a1 (5.7)c1,2 (1.9)b2
54.1 54.1 39.4 51.4
4 a
(5.3) 1 (6.1)a1 (5.7)b2,3 (3.6)a3
57.8 55.3 50.7 52.8
24 a
(3.2) 1 (6.0)a1 (5.3)a3 (3.3)a3
61.1 57.8 68.7 50.2
48 b
(3.6) 1 (2.6)b1 (2.2)a4 (2.2)c3
65.0 59.2 62.2 48.4
(7.0)a1 (4.3)a1 (2.4)a4 (3.2)b3
Mean values exhibiting different superscript letters (within columns) and different numbers (within rows) were significantly different (P!0.05).
334
W.M. Palin et al. contrast to total (pre- and post-gel) volumetric contraction. Therefore, polymerisation shrinkage stress is not only dependent upon the volumetric shrinkage of the restorative material but also the nature of the interfacial bond between the RBC restorative and the associated tooth structure.
Conclusion Figure 6 The DC of each material under investigation following 0.1, 0.5, 1, 4, 24 and 48 h post-irradiation exhibiting the differences in monomer conversion of freeradical and cationic polymerisation for the methacrylate and oxirane RBCs, respectively.
the tooth/restoration interface may provide a more reasonable explanation. The significant decrease in microleakage of the cavities restored with H1 compared with Filteke Z250 can be attributed to the significant decrease in cuspal deflection (6.01G 1.79 and 16.52G3.27 mm, respectively; Table 2) manifested as a reduction in polymerisation shrinkage stress at the tooth/restoration interface. Although the cuspal deflection of cavities restored with EXL596 was significantly decreased compared with H1 (2.52G0.86 and 6.01G1.79 mm, respectively; Table 2), it was suggested that partial failure of the adhesive system, EXL596b reduced the restrictive stresses within the polymerising resin matrix of EXL596 compared with the increased synergism of H1b. This postulation was further justified by a significant decrease in the associated microleakage of cavities restored with H1 compared with EXL596 (Fig. 5). From a clinical perspective, it was proposed that the significant reduction in the cuspal deflection of cavities restored with H1 and the apparent increase in synergism of H1b at the tooth/restoration interface compared with Filteke Z250 might be advantageous in terms of clinical longevity. However, the non-significant differences in cervical gingival microleakage between the H1 and Z100 may suggest only a modest decrease in the possible deleterious effects associated with the polymerisation shrinkage stress of the novel silorane RBC, H1. It may be speculated that a decrease in polymerisation shrinkage stress and a reduction in the associated deleterious effects, such as microleakage, was manifested as a significant decrease in the polymerisation shrinkage of H1 compared with Z100. However, the non-significant differences in cervical gingival microleakage between the H1 and Z100 further highlights the clinical relevance of evaluating the post-gel shrinkage of RBC specimens in
The potential for the cationic polymerisation of novel oxirane and silorane RBCs to decrease the magnitude of polymerisation shrinkage stress at the tooth/restoration interface compared with the free-radical polymerisation of conventional methacrylate RBCs may be realised. However, in the current study, failure of the oxirane adhesive system (EXL596b) and the associated inadequate marginal seal at the tooth/restoration interface would preclude its use as a dental restorative. The significant decrease in microleakage of the cavities restored with the experimental silorane (H1) compared with Filteke Z250 was attributed to the significant decrease in cuspal deflection and manifested as a reduction in polymerisation shrinkage stress at the tooth/restoration interface. The nonsignificant differences in the cervical gingival microleakage between H1 and Z100 may only present modest decreases in the possible deleterious effects associated with the shrinkage stress of current RBC formulations.
References [1] Davidson CL, DeGee AJ, Feilzer AJ. The competition between the composite-dentine bond strength and the polymerisation contraction stress. J Dent Res 1984;63: 1396–9. [2] Sakaguchi RL, Peters MC, Nelson SR, Douglas WH, Poort HW. Effects of polymerisation contractions in composite restorations. J Dent 1992;20:178–82. [3] Lutz F, Kreici I, Barbakow F. Quality and durability of marginal adaptation in bonded composite restorations. Dent Mat 1991;7:107–13. [4] Tantbirojn D, Versluis A, Pintado MR, DeLong R, Douglas WH. Tooth deformation patterns in molars after composite restoration. J Dent 2004;20:535–42. [5] Bausch JR, De Lange K, Davidson CR, Peters A, De Gee AJ. Clinical significance of polymerisation shrinkage of composite resins. J Pros Dent 1982;48:59–67. [6] Abbas G, Fleming GJP, Harrington E, Shortall ACC, Burke FJT. Cuspal movement and microleakage in premolar teeth restored with a packable composite cured in bulk or in increments. J Dent 2003;31:437–44. [7] Hofmann N, Denner W, Hugo B, Klaiber B. The influence of plasma arc vs. halogen standard or soft-start irradiation on polymerization shrinkage kinetics of polymer matrix composites. J Dent 2003;31:383–93.
Cuspal deflection of low-shrink composite filled cavities [8] Dennison JB, Yaman P, Seir R, Hamilton JC. Effect of variable light intensity on composite shrinkage. J Pros Dent 2000;84:499–505. [9] Labella R, Lambrechts P, Van Meerbeek B, Vanherle G. Polymerisation shrinkage and elasticity of flowable composites and filled adhesives. Dent Mat 1999;15:128–37. [10] Watts DC, Hindi AA. Intrinsic soft-start polymerisation shrinkage-kinetics in an acrylate-based resin composite. Dent Mat 1999;15:39–45. [11] Davidson CL, Feilzer AJ. Polymerisation shrinkage and polymerisation shrinkage stress in polymer-based restoratives. J Dent 1997;25:435–40. [12] Condon JR, Ferracane JL. The effect of composite formulation on polymerisation stress. J Am Dent Ass 2000; 131:497–503. [13] Kaisaki DA, Mitra SB, Schultz WJ, DeVoe RJ. Dental visiblelight curing epoxy system with enhanced depth of cure. 3M Innovative Properties Company. US Patent: 5, 856, 373; 1999 [14] Odian G. Ring-opening polymerisation-principles of polymerisation. New York: Wiley; 1991. [15] Tilbrook DA, Clarke RL, Howle NE, Braden N. Photocurable epoxy-polyol matrices for use in dental composites. Biomat 2000;21:1743–53. [16] Millich F, Jeang L, Eick JD, Chappelow CC, Pinzino CS. Elements of light cured epoxy-based dental polymer systems. J Dent Res 1998;77:603–8. [17] Palin WM, Fleming GJP, Burke FJT, Marquis PM, Randall RC. Monomer conversion versus flexure strength of a novel dental composite. J Dent 2003;31:547–9. [18] Schweikl H, Schmalz G, Weinmann W. Mutagenic activity of structurally related oxiranes and siloranes in Salmonella typhimurium. Mut Res 2002;521:19–27. [19] Guggenberger R, Weinmann W. Exploring beyond methacrylates. Am J Dent 2000;13:82D–84D. [20] Patent number WO98/22521; 3M ESPE Dental Products, Seefeld, Germany. [21] Weinmann W, Luchterhandt T, Guggenberger R, Stippschild A, Then S, Dede K. Comparative testing of volumetric shrinkage and sealing of silorane and methacrylate filling materials. J Dent Res 2002;81. Abstr. No. 3382 (http://iadr.confex.com/iadr/2002SanDiego/techprogram/abstract_20711.htm).
335
[22] Thalacker C, Heumann A, Weinmann W, Guggenberger R, Luchterhandt T, Syrek A. Marginal integrity of class V silorane and methacrylate composite restorations. J Dent Res 2004. Abstr. No. 1364 (http//iadr.confex.com/iadr/ 2004Hawaii/techprogram/abstract_45046.htm). [23] Spierings TM, Peters MB, Bosman F, Plasschaert AM. Verification of theoretical modelling of heat transmission in teeth by in vivo experiments. J Dent Res 1987;66: 1336–9. [24] Palmer DS, Barco MT, Billy EJ. Temperature extremes produced orally by hot and cold liquids. J Pros Dent 1992; 67:325–7. [25] McCullock AJ, Smith BGN. In vitro studies of cuspal movement produced by adhesive restorative materials. Br Dent J 1986;161:405–9. [26] Jantarat J, Panitvisai P, Palamara JEA, Messer HH. Comparison of methods for measuring cuspal deflection of teeth. J Dent 2001;29:75–82. [27] Meredith N, Setchell DJ. In vitro measurement of cuspal strain and displacement in composite restored teeth. J Dent 1997;25:331–7. [28] Allen EP, Bayne SC, Brodine AH, Cronin RJ, Donovan TE, Kois JC, Summitt JB. Annual review of selected dental literature: report of the committee on scientific investigation of the American academy of restorative dentistry. J Pros Dent 2003;90:50–80. [29] Christensen GJ. Curing restorative resin. a significant controversy. J Am Dent Ass 2000;131:1067–9. [30] Crivello JV. Cationic polymerisation of iodonium and sulfonium salt photoinitiators. Adv Poly Sci 1984;62: 1–48. [31] Andrzejewska E. Photopolymerisation kinetics of multifunctional monomers. Prog Poly Sci 2001;26:605–65. [32] Decker C. Photoinitiated cross-linking polymerisation. Prog Poly Sci 1996;4:593–650. [33] Crivello JV, Lee JL. The synthesis, characterisation, and photoinitiated cationic polymerisation of silicon-containing epoxy resins. J Poly Sci: Poly Chem 1990;28:479–503. [34] Versluis A, Douglas WH, Cross M, Sakaguchi RL. Does an incremental filling technique reduce polymerisation shrinkage stresses? J Dent Res 1996;75:871–8.