Eur J Oral Sci 2011; 119: 316–322 Ó 2011 Eur J Oral Sci

DOI: 10.1111/j.1600-0722.2011.00833.x European Journal of

Printed in Singapore. All rights reserved
Oral Sciences

Frode Staxrud1, Jon E. Dahl1,2

Role of bonding agents in the repair of 1
Department of Cariology and Geriodontology,
Faculty of Dentistry, University of Oslo, Oslo;
composite resin restorations 2
Nordic Institute of Dental Materials, Oslo,
Norway

Staxrud F, Dahl JE. Role of bonding agents in the repair of composite resin restorations.
Eur J Oral Sci 2011; 119: 316–322. Ó 2011 Eur J Oral Sci

Six commonly used composite resin materials and recommended bonding systems
were tested to assess shear bond strength at the interface between aged and new
composites with and without bonding. Test specimens were aged in water for 60 d
before new composite was placed. Shear bond strength was assessed after 22 ± 2 h
(Test 1) and after additional ageing by thermocycling (5–55°C/5,000 cycles) (Test 2).
After an additional 180 d in water, the aged specimens were randomly divided into
three groups to blind the test with respect to the aged composite. New composites were Dr Frode Staxrud, Department of Cariology and
placed on aged specimens (two groups with and one without bonding agent) and Geriodontology, Dental Faculty, University of
thermocycled (Test 3). After 24 h (Test 1), the mean shear bond strength of the test Oslo, PO Box 1109 Blindern, NO-0317 Oslo,
specimens was 21–26 MPa when bonding agents were used, as opposed to 10–15 MPa Norway
without bonding agents. After thermocycling (Test 2), the mean shear bond strength
Telefax: +47–22–852344
was 16–23 MPa with a bonding agent and 17 MPa without a bonding agent. After E-mail: [email protected]
180 d in water and subsequent thermocycling (Test 3), the mean shear bond strength
was 9–13 MPa with bonding agent and 2–3 MPa when no bonding agent was used. Key words: bonding agents; Composite; repair;
shear bond strength; thermocycling
The results of this study therefore indicate that the use of bonding agents significantly
improves the quality of composite repair. Accepted for publication April 2011

The repair or the extension of resin-based composite the composite and the tooth (6). It has been suggested
restorations is an everyday task for dentists. However, that composite-to-composite bonding could be improved
the existing composite restoration may have been in the by the use of sandblasting techniques (7–10), silanes (8,
oral environment for a long, and often unknown, period 10, 11), flowable composites (8, 12, 13), diamond bur
of time, and in many cases the dental clinician is unaware roughening (14, 15), 1 mm-deep grooves (16), hydro-
of which product has been used to create the restoration. fluoric acid pretreatment (14, 15), and dentin-bonding
Regardless, the clinician may find it desirable, or even systems (7–21). In these studies the repair composites
favourable, to keep all or part(s) of the old restoration in were of the same type as the old composite. In a clinical
place. The reasons for this may be financial (in that a situation when a composite is to be repaired, the brand or
small filling represents a lower cost) or practical (in order type of the old composite is not necessarily known. It is
to make the procedure more conservative by sacrificing therefore of interest to see whether the use of a bonding
less adjacent sound tooth tissue and/or to simplify the system will improve or reduce the quality of the interface
replacement restoration). Simpler restorations may be when the old composite resin material is of a different
time saving and may enhance moisture control. Common brand than the new repair composite resin.
situations are cusp fractures adjacent to an existing Thus, the aim of this study was to evaluate the SBS
good-quality restoration, replacement of one side of a between new and old composite restorative materials in
three-surface class II filling, treatment of secondary order to assess the influence of bonding systems on the
caries without removing all of the existing restoration, repair of composite restorations, particularly when the old
adjacent primary caries, and restoration of endodontic composite is of an unknown type or brand. Furthermore,
access. The treatment goal is to achieve a long-lasting we attempted to establish the influence of ageing on the
restoration of good quality, and the repair of resin-based quality of the adhesion between the two composites. Our
composites is reported to have a favourable outcome for null hypothesis was that the use of bonding systems has no
the longevity and quality of the restoration (1–3). effect upon bond strength when repairing an old com-
During repair of a composite restoration the prepared posite restoration with a new composite resin.
cavity usually also exposes enamel and dentin. For that
reason the use of bonding systems is required to obtain
adequate bond strength to the tooth structures (4, 5). Material and methods
Many studies have been carried out to investigate the
bond strength between composite/enamel and composite/ Preparation of specimens
dentin, and current bonding systems give acceptable The materials used for the experiments are given in Table 1.
tensile strength and shear bond strength (SBS) between The selected methacrylate-based composites were the most
 Repair of composite 317

Table 1
Composites and bonding systems tested

Composite Bonding system

Manufacturer Brand name Batch/Lot Brand name Batch/Lot

3M Filtek Silorane 2007 11 12 7AT Silorane bonding P: 2AP
B: 7AG
3M Filtek Supreme XT 2008 03 30/2008 02 29 Adper Scotchbond 1XT 20080317/8PL
3M Filtek Z 250 2007 11 03 7LU Adper Scotchbond MP 20080417
P: 8BR
B: 8RE
Dentsply CeramX Mono 0803002878 Xeno III 0801002437
Heraeus Kulzer Charisma 010307 2011–10 None
Ivoclar Vivadent Tetric Evo Ceram L08982 AdheSE L13575
1: L06396
2: L07513

All composites were of shade A3 (Lumin Vita).

commonly used products in Norway at the time of the procedure. The aged composite cylinders were placed in a
study, and a newly introduced silorane-based composite was specially designed support and wet ground manually with
included. Test specimen bases were made by packing com- grinding paper (Fepa P #320, silicon carbide, waterproof;
posite in copper rings (8 mm diameter, 10 mm height), Struers, Copenhagen, Denmark). Approximately 0.5 mm
placed on a Mylar strip on a table. The composite cylinders was ground off at the top and 0.2 mm was ground off at the
were light-cured every 2 mm increment from the top, bottom of each cylinder to even the surface before the
according to the manufacturerÕs specifications, using a specimen bases were rinsed with water. The ground surface
Demetron VCL 400 curing lamp (KerrHawe, Orange, CA, of one side of the composite was dried by air blowing for
USA) with an irradiance of 859 mW cm)2, as measured by 10 s and masked with a plastic tape with a circular aperture
the Norwegian Radiation Protection Authorities (Österaas, of 3 mm diameter. Bonding agents were used according to
Norway). After curing, the copper rings were carefully split the manufacturerÕs specifications with respect to drying,
with a carbide disk and removed. Ten test-specimen bases rubbing of the surface, air blowing, time, number of layers,
were produced of each composite-resin product. In total, 60 and light curing. The bonding procedure was performed by
composite cylinders were made, giving 120 available test placing a cylindrical specimen of the new composite
surfaces. One additional cylinder of each composite was (diameter 3 mm, height 2 mm) using the equipment
made, ground as described below, and prepared for surface described in ISO/TS 11405:2003 (22 ) and then light-cured
evaluation using a scanning electron microscope (Phillips (Curing lamp 3M CL XL 3000; 3M, St Paul, MN, USA) at
XL 30, DX4i; EDAX International, Mahwah, NJ, USA). 584 mW cm)2. Bonding agent was not used for one of the
The specimen bases were placed in distilled water at 37°C products (Charisma), and for another product (Filtek Z
immediately after production, and stored for a minimum of 250), test specimens were prepared both with and without
60 d to simulate an ageing process. The water was changed the use of bonding. The test specimens were kept in distilled
every week to prevent bacterial growth. During the entire water at 37°C for 22 ± 2 h after the new composite was
study all test specimens were stored in distilled water at 37°C. attached and testing of the SBS was then performed.
The test procedure for measuring SBS is described in ISO/
TS 11405:2003 (22). The test specimens were fixed in a brass
Test 1: shear bond strength after 24 h (short-term test) cylinder by means of a Teflon ring and a small screw, and
A new composite of the same material was attached to the the cylinder was placed in a test jig. A load parallel to the
aged composite, as shown in Table 2, using the following interphase was applied at a speed of 1 mm min)1 on the test

Table 2
Composition of test specimens

ÔAgedÕ composite Bonding Type of bonding ÔNewÕ composite

Filtek Z 250 Adper Scotchbond MP Three-step ER (fourth generation) Filtek Z 250

Filtek Silorane Silorane bonding Two-step SE (fifth generation) Filtek Silorane
Charisma None Charisma
Filtek Supreme XT Adper Scotchbond 1XT Two-step ER (fourth generation) Filtek Supreme XT
CeramX Mono Xeno III One-step SE (sixth generation) CeramX Mono
Tetric Evo Ceram AdheSE Two-step SE (fifth generation) Tetric Evo Ceram
Filtek Z 250* None Filtek Z 250
*
Filtek Z250 was ground again and new composite was bonded without adhesive.
ER, etch and rinse; SE, self-etch.
318 Staxrud & Dahl

specimens at the site of the bonding using a universal testing to 55°C. After thermocycling, the specimens were stored in
machine (Instron 1121, Instron, High Wycombe, Bucks, water for 24 h before SBS testing.
UK). The load at breakage, measured in N, was converted
to MPa after measuring the area of the adherent surface
based on the hole in the masking tape. Statistics
The probability of failure in the test specimens was assessed
Test 2: shear bond strength after ageing by means of a distribution plot, and the significance of the
differences was evaluated using the Kolmogorov–Smirnov
(thermocycling)
test (23). The distribution of adhesive and cohesive fractures
The composite cylinders were stored in water for another was analysed by the hypergeometric distribution (24), and
14 d after completion of the short-term testing. The end of the differences in SBS between Test 1 and Test 2 were
the specimen that was not used in the short-term SBS test analysed using the Wilcoxon rank-sum test/Mann–Whitney
(Test 1) was ground with the same grinding paper (Fepa P U-test (24).
#320) used to prepare the specimens for Test 1. Bonding
agent was applied, and new composite resin was placed and
light-cured as in the short-term test. The test specimens were
thermocycled in water for 5,000 cycles at temperatures Results
ranging from 5 to 55°C (20 s at each temperature, inter- The results are summarized in Table 3. Short-term SBS
mediate time of 2–3 s). Shear bond strength testing was
testing (Test 1) resulted in mean breakage values ranging
performed as described for the short-term SBS test, and all
specimens were tested on the same day, 17 d after the short- from 19.8 to 26.0 MPa when a bonding system was used,
term test. and mean breakage values of 9.9 and 15.2 MPa without
any bonding system. After thermocycling (Test 2) the
mean values varied from 16.0 to 22.7 MPa. Only two of
Test 3: bonding to composite of unknown origin after the six composite-binding systems had a statistically
ageing (thermocycling) significantly higher mean SBS in Test 1 compared with
After another 180 d in water, 59 specimen bases were Test 2. The product without adhesive had a mean value
checked for microfractures using light microscopy and of 16.8 MPa. In the short-term SBS test, 67% of the
randomly divided into three groups of 20, 20, and 19 fractures were apparent interfacial failures (Table 4).
specimens, respectively. One cylinder was completely Sometimes the fracture line followed the new composite/
destroyed in the second test. The different groups consisted bonding agent interface, crossed the bonding agent, and
of both methacrylate- and silorane-based aged specimens. continued at the old composite/bonding agent interface.
The test was blinded with respect to the ÔoldÕ composite. The The remaining 33% of failures were cohesive fractures
aged composite cylinders were ground at one end using the
within the specimen base material. The fracture lines
grinding paper Fepa P #320. The new composite resin and
bonding systems were as follows: Series I, Filtek Supreme started at either the new composite/adhesive interface or
XT/Adper Scotchbond MP; Series II, Tetric Evo Ceram the old composite/adhesive interface, and turned
with no bonding; and Series III, Tetric Evo Ceram/AdheSE approximately 60° into the old composite specimen. The
One Viva-pen. The bonding agents and new composites new composite was still bonded to the fractured part of the
were placed as described for the other tests. The test speci- old composite. After thermocycling, 90% of the failures
mens were stored in distilled water for 14 d followed by were found to be apparent interfacial fractures (Table 4).
thermocycling (5,000 cycles, at temperatures ranging from 5

Table 3
Results of shear bond strength testing between specimens made of the same product after 22±2 h of storage in distilled water and after
thermocycling

Test 1
(tested after 22 ± 2 h storage in Test 2
distilled water) (tested after thermocycling)
Test specimen Mean SD Pf10 Pf90 Mean SD Pf10 Pf90 P-value*

Filtek Z250/Adper Scotchb MP, 3M 26.0 10.7 1.5 34.4 22.7 4.3 16.7 28.5 > 0.05
Silorane/Silorane bonding, 3M 19.8 8.9 6.8 33.6 17.7 7.2 4.7 26.4 > 0.05
Charisma, Heraeus Kulzer 15.2 6.4 1.1 21.4 16.8 4.7 10.8 23.7 > 0.05
Filtek Supreme XT/Adper Scotchbond XT, 3M 22.0 7.1 1.5 30.7 19.0 5.6 11.9 27.4 > 0.05
Ceram X Mono/Nano III, Dentsply 22.1 4.8 17.6 32.4 16.6 4.4 8.0 23.2 > 0.01
Tetric Evo Ceram/AdheSE, Vivadent Ivoclar 25.5 7.5 15.0 36.3 16.0 9.8 0.5 33.6 > 0.05
Filtek Z250, 3M, Control 9.9  2.7 6.8 13.8

All values are given in MPa.

 
P < 0.05 compared with other groups in Test 1.
*
Difference between Test 1 and Test 2 for each product.
Pf, probability of failure (at what load a certain percentage of the specimens will fracture); Pf10, probability of failure for 10% of the
specimens; Pf90, probability of failure for 90% of the specimens.
 Repair of composite 319

Table 4
Distribution of type of fracture (adhesive or cohesive), after shear bond strength testing of specimens made from the same product, after
22±2 h of storage in distilled water and after thermocycling

Test 1
(tested after 22 ± 2 h storage Test 2
in distilled water) (tested after thermocycling)
Test specimens Adhesive Cohesive Adhesive Cohesive P-value*

Filtek Z250/Adper Scotchb MP 9 0 10 0 > 0.02

Filtek Silorane/Silorane bonding 10 0 10 0 > 0.02
Charisma/no bonding 9 0 6 4 > 0.02
Filtek Supreme XT/Adper 8 2 10 0 > 0.02
Scotchbond XT
Ceram X Mono/Xeno III 0 10 8 2 < 0.01
Tetric Evo Ceram/AdheSE 3 7 9 0 < 0.01
Sum 39 19 53 6 < 0.01

Values given represent the number of fractures, unless stated otherwise.

*
Differences between Test 1 and Test 2.

Table 5
Results of shear bond strength testing between aged specimens of unknown composite and the new composites Filtek Supreme XT or
Tetric Evo Ceram, after thermocycling

Test 3
(tested after 180 d in water)
Test specimen Mean SD Pf10 Pf90

Unknown composite/Filtek Supreme XT/Adper Scotchbond MP (Series I) 8.9 7.5 0 19.5

Unknown composite/Tetric Evo Ceram/no bonding agent (Series II) 2.6* 4.2 0 11.1
Unknown composite/Tetric Evo Ceram/AdheSE One Vita-pen (Series III) 12.7 5.9 0 19.2

All values are given in MPa.

*Statistically significantly different (P<0.05) from Series I and Series III.
Pf, probability of failure (at what load a certain percentage of the specimens will fracture); Pf10, probability of failure for 10% of the
specimens; Pf90, probability of failure for 90% of the specimens.

Fig. 1. Distribution of the test results with probability of failure

Fig. 2. Scanning electron micrograph of Tetric Evo Ceram,
as a function of load. The value 0.9 means that 90% of the test
showing an uneven surface with pits, grooves, grinding marks,
specimens have fractured and 0.5 represents the median value.
porosities, and filler particles that provide opportunities for
The tests were blinded with respect to the old composite. The new
mechanical interlocking.
composites were as follows: Series I, Filtek Supreme XT/
Scotchbond MP bonding; Series II, Tetric Evo Ceram/control,
no bonding; and Series III, Tetric Evo Ceram/AdheSE One
Vivapen bonding. Series I and Series III were statistically signifi- bonding was used (Table 5 and Fig. 1). When no
cantly different from the control group (Series II) (Kolmogorov– bonding was used, the mean value was as low as
Smirnov test, P > 0.05). SBS, shear bond strength. 2.6 MPa. All fractures were of the adhesive type in this
test. Test specimens where the attachment of the new
Testing new known composite bonded to old unknown composite did not withstand the thermocycling proce-
composite (180 d in water) (Test 3) revealed that the dure alone, or detached before SBS testing, were given
mean values of SBS were 8.9 and 12.7 MPa when the value 0 N.
320 Staxrud & Dahl

The scanning electron microscopic evaluation of the composite is of a mechanical or micromechanical nature.
ground test specimens revealed uneven surfaces with pits, Roughness of the surface and the size and the shape of
grooves, grinding marks, porosities, and extending filler the filler particles provide possibilities for the resin to
particles (Fig. 2). flow into undercuts and are important factors for
micromechanical retention. The scanning electron
microscope studies of freshly ground composite surfaces
before the bonding procedure demonstrated an uneven
Discussion
surface with pits, grooves, grinding marks, porosities,
Joining a new composite resin material with the same and protruding filler particles (Fig. 2) that give oppor-
resin material that was aged in water for 60 d resulted in tunities for mechanical interlocking. Sandblasting with
SBS values comparable to values obtained for the aluminium oxide or silica-treated aluminium oxide has
bonding of composite to dentin specimens (Test 1) (25, been introduced and is used by many practitioners. This
26). However, the SBS values found in the present study has been shown to have a potentially better effect on the
are lower than those observed by other researchers when micromechanical bond strength (7–10, 12, 14). However,
joining an old and a new composite of the same brand these procedures require use of a fully rubber dam-
(7–19, 27). After 60 d of water uptake, the composite is controlled working field in order to avoid damage to the
considered to be saturated with respect to water (28, 29). periodontium and to protect the airway of the patient.
Increasing the age of the base specimens lowered the Roughening of the surface will markedly improve the
bond strength to a second, new material, as revealed in retention (14, 15) and will, of course, be a natural part of
test specimens that were aged by thermocycling before cavity preparation. In this study, the use of grinding
the bonding procedure (Test 3). paper mimics the roughening produced by a diamond
We observed a reduced SBS after ageing by thermo- bur. The present study was limited to focussing on the
cycling (Table 3) although the reduction was statistically effect of the bonding agent alone.
significant for only two of the six composite brands. It is The same bond strength could not be seen in the series
suggested that bond-strength data gathered after labo- where a bonding agent was not used (Charisma and
ratory ageing to some extent predict the clinical perfor- Filtek Z250 in Test 1, and Tetric Evo Ceram without
mance in the longer term (30), indicating that the repair bonding in Test 3). Possible explanations for this might
strength might be reduced over time. be that the new hydrophobic composite adhered poorly
In most situations when repairing or extending an old to the water-saturated ÔoldÕ composite. Modern adhesives
composite restoration, fresh dentin and enamel are use hydrophilic primers with a rather high content of the
exposed. For optimal bonding of the restoration to tooth low-molecular-weight monomer 2-hydroxyethylmethyl-
structures the use of an adhesive system is required (31). acrylate (2-HEMA). This molecule has the ability to wet
One question is whether the surface of the remaining the old composite by altering the surface tension and
composite should be included or not in the bonding thus allows deep penetration into pits, grooves, and
procedure. The present results show that the use of porosities. As well as superior wetting, presumably some
bonding systems, including acid-etching for the etch-and- of the adhesives have Ôwater chasingÕ or dehydrating
rinse products, improved the bond strength between old capabilities if they contain acetone or alcohol. The very
and new composites, compared to joining the two small size and amphiphilic nature of 2-HEMA allows it
materials without a bonding system. This finding was to penetrate into parts of the composite containing water
most striking in cases where the ÔagedÕ material was (a hydrophilic property of 2-HEMA) (33, 34) and to
exposed to thermocycling before the bonding procedure chemically bond to the monomers when light cured
was carried out (Test 3) (Table 5). In these cases, the new (a hydrophobic property). Hydrophobic monomers in the
material was probably a different product from the ÔagedÕ restorative material bind easily to the monomers in the
material. The results indicate that in order to obtain bonding resin and this is the same mechanism that takes
optimal bonding, the remaining resin-based restoration place between bonding agents and dentine and enamel.
should be treated in the same way as the tooth structure Ageing the test specimens revealed that the interface
regarding the bonding procedure. This facilitates a sim- between the old and the new materials was dependent on
pler clinical procedure in cases of repair or extension of the bonding layer. Shear bond strength testing of speci-
an existing composite restoration because the entire mens after 24 h resulted in both adhesive and cohesive
cavity will then be treated in the same manner. fractures (Table 4). The cohesive fractures were in the
It is unlikely that the bonding between new and old aged specimens, indicating that the bond strength was
composite restorative material is of a chemical nature. superior to the inherent strength of the aged composite.
Probably only small amounts of monomers remain in the By increased ageing, most or all fractures were of the
old composite and these may have lost any ability to adhesive type, indicating a weakening of the bonding
make a chemical interaction with the new resin, as over interface (Test 2 and Test 3). The hypothesis that there is
time there is both a passive polymerization of resins and no difference in fracture modes between short-term and
an elution of residual monomers. It has been shown that thermocycled specimens (Test 1 and Test 2) was tested by
the half-life of free radicals in composites is 50 h at 37°C, comparison with the hypergeometric distribution (24).
making it unlikely that any free radicals are left after a The hypothesis failed for three materials (Ceram X,
few weeks of storage (32). It is therefore most likely that Charisma, and Tetric Evo Ceram), but could not be
the connection between the new composite and the old rejected for the other materials. For Filtek Z250 and
 Repair of composite 321

Filtek Silorane, all specimens failed by adhesive fracture. systematic review of current clinical trials. Biomaterials 2005;
Both the bonding resin and the new composite resin 21: 864–881.
6. De MJ, Van LK, Peumans M, Poitevin A, Lambrechts P,
materials are inclined to take up water during storage in Braem M, Van MB. A critical review of the durability of
water and during thermocycling. Water will break the adhesion to tooth tissue: methods and results. J Dent Res 2005;
chemical structures, such as carboxyl and hydroxyl 84: 118–132.
groups, that are found in the composite. The major 7. O¨ ztas N, Alacam A, Bardakcy Y. The effect of air abrasion
factor affecting durability of the bonding agent is thus with two new bonding agents on composite repair. Oper Dent
2003; 28: 149–154.
hydrolysis of the resin interface and subsequent elution 8. Frankenberger R, Kramer N, Ebert J, Lohbauer U,
of the breakdown products (6, 12). Therefore, the Kappel S, ten WS, Petschelt A. Fatigue behavior of the
hydrolytic stability of the bonding system is of major resin-resin bond of partially replaced resin-based composite
importance for the success of a composite repair resto- restorations. Am J Dent 2003; 16: 17–22.
9. Papacchini F, DallÕOca S, Chieffi N, Goracci C, Sadek FT,
ration. Some investigators suggest that improved Suh BI, Tay FR, Ferrari M. Composite-to-composite
hydrolytic stability and bond strength can be achieved by microtensile bond strength in the repair of a microfilled hybrid
the use of flowable resin (8, 12, 13). This is probably resin: effect of surface treatment and oxygen inhibition. J Adhes
because of the hydrophobic nature of such resins, and Dent 2007; 9: 25–31.
10. Rathke A, Tymina Y, Haller B. Effect of different surface
excellent strength values are shown when the repair is
treatments on the composite–composite repair bond strength.
totally in composite, especially in combination with air- Clin Oral Invest 2009; 13: 317–323.
abrasion techniques (7–10, 12). 11. Brendeke J, Ozcan M. Effect of physicochemical aging
It has been suggested that the three-step, ethanol/ conditions on the composite–composite repair bond strength.
water-based etch-and-rinse bonding procedure remains J Adhes Dent 2007; 9: 399–406.
12. Papacchini F, Toledano M, Monticelli F, Osorio R,
the Ôgold standardÕ in terms of adhesion durability (6, Radovic I, Polimeni A, Garcı́a-Godoy F, Ferrari M.
32). This is a result of the hydrolytic stability of this Hydrolytic stability of composite repair bond. Eur J Oral Sci
bonding system, where most of the initial priming layer 2007; 115: 417–424.
containing hydrophilic resin is evaporated by air blowing 13. Papacchini F, Magni E, Radovic I, Mazzitelli C,
Monticellia F, Goracci C, Polimeni A, Ferrari M.
before applying the hydrophobic resin. In the present
Effect of intermediate agents and pre-heating of repairing
study, the average bonding values for specimens pre- resin on composite-repair bonds. Oper Dent 2007; 32: 363–
pared using the three-step etch-and-rinse procedure were 371.
the highest (Filtek Z250/Adper Scotchbond MP) when 14. Rodrigues JS, Ferracane JL, Bona AD. Influence of
compared with other bonding strategies, both after 24 h surface treatments on the bond strength of repaired com-
posite restorative materials. Dent Mater 2009; 25: 442–
(Test 1) and after ageing by thermocycling (Test 2). The 451.
differences were, however, not statistically significant. 15. Celik EU, Ergücü Z, Türcün LS, Ercan UK. Tensile bond
In summary, increased SBS was observed between the strength of an aged resin composite repaired with different
aged and the new composite materials when using Protocols. J Adhes Dent 2010; Epub ahead of print.
16. Vivas J, Yaman P, Taylor G. Effect of different surface
bonding agents compared to specimens without such
treatments on the shear and flexural re-bond strengths of a
treatment. The longer a resin-based composite restora- micro-hybrid composite. J Contemp Dent Pract. 2009; 10:
tion had been exposed to water or to a humid environ- E001–008.
ment before repair, the weaker the SBS and the greater 17. Yesilyurt C, Kusgoz A, Bayram M, Ulker M. Initial repair
the difference between repair with and repair without bond strength of a nano-filled hybrid resin: effect of surface
treatments and bonding agents. J Esthet Restor Dent 2009; 21:
bonding systems. Thus, the null hypothesis has to be 251–260.
rejected. We recommend the use of bonding agents when 18. Tezvergil A, Lassila LV, Vallittu PK. Composite–com-
repairing resin-based composite restorations. posite repair bond strength: effect of different adhesion primers.
J Dent 2003; 31: 521–525.
Acknowledgements – The expert assistance of Ms Ellen 19. Padipatvuthikul P, Mair LH. Bonding of composite to water
Austrheim, NIOM, Mr Ketil Kvam, NIOM, Dr John Tibballs, aged composite with surface treatments. Dent Mater 2007; 23:
NIOM, and Dr Alix R. Young Vik, University of Oslo, is 519–525.
20. Maneenut C, Sakoolnamarka R, Tyas MJ. The repair
greatly acknowledged.
potential of resin composite materials. Dent Mater 2011; 27:
e20–e27.
21. Cavalcanti AN, deLima AF, Peris AR, Mitsui FHO,
Marchi GM. Effect of surface treatments and bonding agents
References on the bond strength of repaired composites. J Esthet Restor
1. Gordan VV, Garvan CW, Blaser PK, Mondragon E, Mjör Dent 2007; 19: 90–99.
IA. A long-term evaluation of alternative treatments to 22. International Organisation of Standardization. ISO/TS
replacement of resin-based composite restorations: results of a 11405:2003. Dental materials – testing of adhesion to tooth
seven-year study. J Am Dent Assoc 2009; 140: 1476–1484. structure. Geneva: International Organisation of Standardiza-
2. Moncada G, Martin J, Fernandez E, Hempel MC, Mjör IA, tion2003.
Gordan VV. Sealing, refurbishment and repair of class I and 23. Press WP, Flannery BP, Teukolsky SA, Vetterling WT.
class II defective restorations: a three-year clinical trial. J Am Numerical recipes; the art of scientific computing. Cambridge:
Dent Assoc 2009; 140: 425–432. Cambridge University Press, 1986; 472–475.
3. Mjör IA, Gordan VV. Failure, repair, refurbishing and lon- 24. Rice JA. Mathematical statistics and data analysis. 2nd edn.
gevity of restorations. Oper Dent 2002; 27: 528–534. Wadsworth: Duxbury Press, 1995.
4. Perdigão J. New developments in dental adhesion. Dental 25. Bradna P, Vrbova R, Dudek M, Roubickova A, Housova D.
Clinics North Am 2007; 51: 333–357. Comparison of bonding performance of self-etching and etch-
5. Peumans M, Kanumilli P, De MJ, Van LK, Lambrechts P, and-rinse adhesives on human dentin using reliability analysis.
Van MB. Clinical effectiveness of contemporary adhesives: a J Adhes Dent 2008; 10: 423–429.
322 Staxrud & Dahl

26. Sunico MC, Shinkai K, Medina VO III, Shirono M, Tanaka 30. Van MB, Peumans M, Poitevin A, Mine A, Van EA, Neves
N, Katoh Y. Effect of surface conditioning and restorative A, De MJ. Relationship between bond-strength tests and
material on the shear bond strength and resin-dentin interface clinical outcomes. Dent Mater 2010; 26: e100–e121.
of a new one bottle nanofilled adhesive. Dent Mater 2002; 18: 31. Leinfelder KF. Current status of dentin adhesive systems.
535–542. Alpha Omegan 1998; 91: 17–22.
27. Erdemir A, Eldeniz AU, Belli S. Effect of temporary filling 32. Burtscher P. Stability of radicals in cured composite materi-
materials on repair bond strengths of composite resins. J Bio- als. Dent Mater 1993; 9: 218–221.
med Mater Res B Appl Biomater 2008; 86B: 303–309. 33. Breschi L, Mazzoni A, Ruggeri A, Cadenaro M, Di
28. Lagouvardos PE, Pissis P, Kyritsis A, Daoukaki D. Water Lenarda R, De Stefano E. Dental adhesion review: aging
sorption and water-induced molecular mobility in dental com- and stability of the bonded interface. Dent Mater 2008; 24:
posite resin. J Mater Sci Med 2003; 14: 753–759. 90–101.
29. Sideridou I, Achilias DS, Chrysa SC, Karabela M. Water 34. Van LK, Snauwaert J, De Munck J, Peumans M, Yoshida Y,
sorption characteristics of light-cured dental resins and Poitevin A, Coutinho E, Suzuki K, Lambrechts P, Van
composites based on Bis-EMA/PCDMA. Biomaterials 2004; Meerbeek B. Systematic review of the composition of contem-
25: 367–376. porary dental adhesives. Biomaterials 2007; 28: 3757–3785.