A n in vitro study of the bond strength of two light-cured composites used in the direct bonding of orthodontic brackets to molars G. Bradburn, BDS, FDS, DDO, M.Orth., and N. Pender, BDS, FDS, DDO, MScD, PhD
Liverpool, England This study examines methods of improving the bond strengths of the light-activated composites, Heliosit-orthodontic, and Transbond. Begg brackets were bonded onto each of the four surfaces of 50 molar teeth that had been previously extracted. The bonding systems used were Right-on (R), Heliosit-orthodontic (H), Heliosit with a precured composite resin on the mesh (HPC), Heliosit with a precured intermediate bonding resin on the mesh (HPR), Transbond (T), and Transbond with a precured layer of resin on the mesh (TPC). The shear bond strengths were tested on a M5K tensile tester, crosshead speed 0.5 mm/min. After bracket removal, the enamel surface of the bonding site was examined and assessed with the adhesive remnant index (ARI). The shear bond strength in mean mPa __ 1 SD was found to be significantly greater for HPC, HPR, and TPC than for H (P < 0.001). The ARI scores suggest that bond failure is associated more with the micromechanical bond with enamel for HPC and HPR. The results indicate that the chemical properties of the two light-activated adhesives were improved by precuring on the mesh base of the bracket before bonding. (AM J ORTHOD DENTOFACORTHOP 1992;102:418-26.)
T h e development of adhesives that will satisfactorily bond orthodontic attachments directly to enamel has been greatly influenced by the research work directed toward improving adhesive properties of materials used in conservative dentistry.~ Light-cured materials are an example of such product development. While a strong and durable bond is required, the problem of removing the bracket without damaging the labial enamel must not be overlooked. The ideal bonding material sandwich should fail during debonding by the clean separation of the resin from the etched enamel. 2 The in vitro use of light-cured materials for orthodontic bonding was first described in 1979, 3 and laboratory trials were completed 5 years later. 4 In the direct bonding technique, the material is cured under metalbased brackets by transillumination as thetooth structure transmits visible light very well. 3's In vitro investigations have suggested that light-cured materials compare favorably with chemically cured adhesives. 4 In a clinical investigation, 6 the survivorship of the chemically activated bonding agent Right-on (R) (TP From the Department of Clinical Dental Sciences, School of Dentistry, University of Liverpool. 8 / 1/32325 ,,
418
~ o .
Product, La Porte, Ind.) was compared with the lightactivated material Heliosit-orthodontic (H) (Vivadent, Schaan-Lichtenstein). The study compared the failure rate of 1366 attachments bonded by graduate students. This was 16% for the chemically cured material and 23% for the light-cured material, with both significantly greater on posterior teeth compared with anterior teeth. After this clinical trial, 6 an in vitro study 7 investigated failure of bonds made with Concise (3M HealthCare, St. Paul, Minn.) and Right-on, two chemically activated systems, and the light-activated system Heliosit-orthodontic. The light-activated composite provided the weakest bond to enamel, and bonds on molar teeth were weaker than bonds on incisors, canines, and premolars. The clinical trial 6 suggested that the increased failure rates in the posterior segments were due to difficult access or salivary contamination. This hypothesis appears unjustified. A clinical investigation s using an experimental hybrid type of light-cured material with a filler loading of 62.8% demonstrated a failure rate of 4.7% compared with a chemically cured material at 6%. Before applying the filled resin phase to the bracket, a low viscosity resin phase was precured onto the bracket base in a procedure first prescribed by Tavas. 9
Volume102
Number 5
Bond strength of light-cured composites 419
3
Fig. 1. Diagrammatic representation of method of allocating identification number to bracket site.
OBJECTIVES The objective of this study was to investigate further the properties of Heliosit-orthodontic, and then to compare these with the light-activated material Transbond (T) (Unitek, 3M, St. Paul, Minn.). The molar tooth was chosen as the in vitro model system since the previously lower bond strengths found in shear 7 would permit a more critical assay. The aim was to assess the behavior of light-cured materials and their use with metal brackets and, in particular, to investigate the effects of pretreatment of the bracket base on bond strength of these materials. In addition, the site of bond failure under shear was to be assessed in the context of bond strength with these different modalities and thus further describe bonding to both bracket mesh and etched enamel.
MATERIALS AND METHODS Extracted human molar teeth with sound enamel were stored in distilled water at room temperature. Thirty teeth were mounted in dental stone within hardwood blocks .7 Each of the four comers of the molar teeth was numbered for the purposes of identification. For each specimen, the identification number was advanced in a downward and forward direction throughout this series. Thus on the first tooth the mesiobuccal bonding site was allocated the number 1, while the distobuccal bonding site received this identification number on the second in the series (Fig. 1). The bonding sites on each tooth were then polished with a rubber cup and a pumice and water paste. The surface was then rinsed with water and dried with warm air. Small curved Begg brackets (TP Products, La Porte, Ind.) were cleaned in 70% alcohol, followed by a rinsing in distilled water, and then allowed to dry. The brackets (nominal surface area 9.92 mm'-) were then bonded to the "appropriate bracket site" by one of the following four methods. The bonding procedures were carried out by the same operator using a standard technique. To a~,oid deficiencies around the bracket margins, an excess of material
Fig. 2. Molar tooth mounted in stone in hardwood block with Begg bracket bonded to each of four bonding sites.
was used and this was extruded around the entire periphery of the base on seating. A probe was then used to remove all excess material before polymerization (Fig. 2). Three adhesives were used. Right-on is a chemically activated system comprising a filled diacrylatc resin and an unfilled low viscosity resin containing the activator that is applied to the bracket base and enamel. Polymerization is initiated when the two resins come into contact. The other two materials were light-activated. Heliosit-orthodontic is a diacrylate micro-filled composite resin. It consists ofa methone dimcthyacrylate resin with a 14% filler of pyrolytic sillicium dioxide. Transbond is a microfine composite, which uses a hybrid filler of submicron silica. The average particle size is 3 microns, and the filler loading is approximately 82%. For systems using white light, a diketone is used in conjunction with an organic amine. The dikctonc absorbs blue light in the 420 to 450 nm range, and an excited triplet state is produced, which, together with the amine, results in ion radicals to initiate polymerization.'~
Bracket site 1 (R), No-mix composite resin (Right-on) The bonding site was etched for 60 seconds with Righton etching fluid containing phosphoric acid, rinsed with copious amounts of water and dried with warm air. A bracket was bonded with the composite resin according to the manufacturer's instructions.
Bracket site 2 (H), Heliosit-orthodontic light-cured resin The bracket site was etched for 90 seconds with the proprietary phosphoric acid etching fluid provided. After rinsing and drying, a bracket was bonded with the composite resin. This was cured for two periods of 20 seconds according to the manufacturer's instructions.
420
Am. J. Orthod. Oentofac. Orthop.
Bradburn and Pender
November 1992
/
1
The specimen block was set in the lower jaw of the MK tensile tester (JJ Instruments, United Kingdom) (Fig. 3). With a crosshead speed of 0.5 mm/min, the attachments were shear tested to failure. Shear testing was along an axis parallel to the long axis of the tooth. The point at which failure occurred was noted in Newtons and bond strength calculated in mega Pascals by dividing the force in Newtons by the area of the mesh in milllimeters squared. In a second series of experiments 20 molars were prepared for bonding in a similar manner to that already described. These were used to test the bond strength of the light-cured material Transbond. The following four methods were used. Bracket site 1 (R), no-mix composite resin (Right-on). Bracket site 2 (HPC), Heliosit-orthodontic with precured layer of tleliosit-orthodontic. Bracket site 3 (T), Transbond light-cured resin. The bracket site was etched for 60 seconds with the proprietary phosphoric acid etching gel. After rinsing and drying with warm air, a bracket was bonded with the composite resin. This was cured with the Heliosit white light for two periods of 15 seconds according to the manufacturer's instructions. Bracket site 4 (TPC), Transbond with precured layer of 'Transbond'. The bracket site was prepared and etched as for those in group 3. ttowever, a thin layer of Transbond resin was applied to the mesh base and precured according to the manufacturer's instructions before the routine bonding procedure described in group 3 (T).
Site of bond failure
Fig. 3. MK. Tensile tester with specimen block set in lower jaw.
Bracket site 3 (HPC), Heliosit-orthodontlc with precured layer of Heliosit-orthodontic The bracket site was prepared and etched as for those in group 2. However, a thin layer of resin was applied to the mesh base and precured according to the manufacturer's instructions, before the routine bonding procedure described for site 2 (H).
Bracket site 4 (HPR), Heliosit-orthodontic wffh precured Intermediate bonding resin (Heliobond) The bracket site was prepared and etched as for those in group 2. Before placement of the bracket with Heliosit-orthodontic as previously described, a thin layer of unfilled intermediate resin was applied to the mesh base and polymerizcd according the manufacturer's instructions. The teeth were set aside in a dry atmosphere for 24 hours before shear testing.
Shear strength testing A 0.014-inch ligature wire was placed through the attachment, and both ends were clamped in the shear tester.
The site of bond failure within the tooth-composite-mesh sandwich was assessed after shear testing with the adhesive remnant index (ARI) of Artun and Bergland." This was designated as a clinical technique for assessing the amount of composite remaining on the enamel after bond failure. The index has four categories: I. Score 0 = No adhesive remaining on the tooth. 2. Score I = Less than half of the adhesive on the tooth. 3. Score 2 = More than half of the adhesive on the tooth. 4. Score 3 = All adhesive on the tooth with a distinct impression of the bracket mesh. For these estimates, the material type used on the four bracket sites was unknown to the assessor at the time of examination.
Statistical analysis The parametric data were tested for significance with the pairwise t test. The data from the ARI were in the form of categories. Parametric tests are inappropriate, and the chisquared test for frequencies of occurrence was used for this data. In all tests of significance, alpha was set at 0.05.
RESULTS In the first e x p e r i m e n t , 4 o f the 30 teeth prepared for testing in h a r d w o o d blocks fractured during shear testing. A n o t h e r two samples fractured after testing o f the second and third bracket sites, respectively, thus providing i n c o m p l e t e data. Thus the data analyzed was
Volume 102 Number 5
Bond strength of light-cured composites
Bond Strength.(rnPa.) 16
14 ......................... m 12
421
................
E x p e r i m e n t . t ..... [ ~
Experiment .2 ............................ F'7//.zJ
7`
10 8 /I'
;
;5..;
6 4
. . . . . . . . . . . . . . .
2
~r ~~m m iN ~ H'
R
HPC '
. . . . . . . . . .
................. HPR
.......... I', T
TPC
Method of Bonding~ Fig. 4. Bond strength: Histogram showing mean for each of six methods of bonding.
Table I. Mean ( -+- I SD) bond strength (mega-pascals) of Right-on (R), Heliosit-Orthodontic (H), Heliosit-
orthodontic with precured composite (HPC), and Heliosit-orthodontic with intermediate resin (IIPR)
I Mean SD
R
I
8.45 2.17
derived from the four sites located on 24 teeth. The mean bond strength for each of the bonding systems is shown (Table I and Fig. 4) and the significance of their cross comparison listed in Table II. The highest bond strengths were achieved by precuring material onto the bracket base before carrying out the direct bonding procedure (HPC 9.95 mPa, HPR 8.87 mPa). Heliosit-orthodontic, used without any precuring treatment, consistently gave the weakest bonds (p < 0.001). Right-on had a similar bond strength (8.45 mPa) to Heliosit-orthodontic when this material was pretreated with Heliobond, the intermediate bonding resin (8.87 mPa). The ARI as an in vitro method of assessing the amount of material on the tooth surface is highly reproducible (r = 0.99). The ARI scores for each of the 96 sites is shown (Table III and Fig. 5) and the significance of the cross comparison of the four bonding systems is tabulated in Table IV. Most material remains on the enamel when using Right-on and Heliosit-orthodontic. In both methods that involved precuring material onto the bracket base, the reverse was found with the site of bond failure in the latter two groups associated with the adhesive-enamel interface. The second series of experiments, using a similar
"
I
I
7.22 2.61
,,pR
9.95 2.82
8.87 2.75
II. Significance levels derived from a pairwise crosscomparison of the bond strengths of Right-on, Heliosit-Orthodontic, Heliositorthodontic with precured composite, and Heliosit-orthodontic with intermediate resin (ttPR)
Table
I H HPC HPR
R P < O.OOl P < 0.001 NS
I
" P < 0.001 P < 0.001
I
.pc P < 0.001
method on 20 teeth, included Transbond and Transbond with a precured resin layer. The remaining two sites were occupied by Right-on and Heliosit-orthodontic with a precured layer of that material. Thus these latter two bonding systems were common to both sets of experiments. Only one sample fractured, after testing the second bracket site in the sequence. Therefore data were available from 19 teeth with four sites on each tooth. The bond strengths from these tests are shown (Table V and Fig. 4) with the significance of the crosscomparisons shown in Table VI. The bond strengths of Right-on and Heliosit-orthodontic (HPC) were consis-
422
Bradburn and Pender
Am. J. Orthod. Denmfac. Orthop. November 1992
Mean A.R.I. score. aT
. . . . . . . . . . . . . . .
i
~
Experiment 1
~
~
Experiment 2
:'
2.6
1.5
0.5
I
I
I
Method of Bonding. Fig. 5. ARI scores: Histogram showing mean ARI scores for six methods of bonding.
Table III. Sites of fracture described by the ARI of brackets bonded with Right-on, Heliosit-orthodontic,
Heliosit-orthodontic with precurcd composite, and Heliosit-orthodontic with intermediate resin 1
1 R tt ltPC It PR
I 1 5 5
4 3 12 15
Table IV. Significance levels derived from a
pairwise crosscomparison of site of fracture assessed by the ARI of brackets bonded with Right-on, Heliosit-orthodontic, Heliositorthodontic with precured composite, and Heliosit-orthodontic with intermediate resin
I H HPC HPR
R NS P < 0.001 P < 0.001
I
" P < 0.001 P < 0.001
I NS
tent with those found in the previous experiment. Transbond compared favorably with Right-on (Table VI and Fig. 4). Transbond precured with resin showed the greatest bond strength, which was also significantly greater than precured Heliosit-Orthodontic (HPC~. Thus higher bond strengths seemed to be associated more with the process of precuring on the bracket base than with a particular material.
2
3
Total score
Mean score
7 4 6 4
12 16 I 0
54 59 27 18
2.25 2.45 1.12 0.96
Table VII and Fig. 5 record the ARI scores for the 76 sites examined, and Table VIII shows the significance of the erosscomparison of these scores between the four methods of direct bonding used in this second part of the study. The site of bond failure with precured IIeliosit (HPC) occurred more often at the adhesive enamel interface, which is consistent with the previous experiment. Transbond, in common with Right-0n and Heliosit-orthodontic in the previous experiment, showed bond failure more often associated with the adhesive-bracket interface. The relationship between bond failure and the site of failure for precuring with Transbond (TPC) is less clear with a significant number of cohesive failures located within the composite resin. DISCUSSION
The clinical trial by Lovius et al., 6 showed that Iteliosit-orthodontie failed more often than Right-on on all teeth other than incisors. The laboratory trial by Pender et al., 7 confirmed tleliosit-orthodontic as a weaker material than Right-on and that anterior bond
Voh,me 102 Number 5
Bond strength of light-cured composites
423
Table V. Mean (+_ 1 SD) bond strength (mega-pascals) of Right-on, Heliosit-orthodontic with precured
composite, Transbond (T), and Transbond with precured composite (TPC)
Mean SD
8.22 2.20
10.53 1.53
strengths were greater for both materials than those for posterior teeth. Thus a similar pattern emerged with both in vivo and in vitro testing of the light-cured material. In the clinical trial, 6 it was suggested that poor moisture control, poor access with the Heliosit light, and an increased amount of aprismatic enamel on posterior teeth might be responsible for higher clinical failure rates. In the laboratory trials, access with the Heliosit light and moisture control were not significant factors. For aprismatic enamel to be an important consideration in bond failure, the micromechanical bond with enamel would have to fail more readily with tleliosit-orthodontic on posterior teeth than with other materials. However, the in vitro study of Pender et al. 7 showed that this was not the case as the weaker bond strength of Heliosit-orthodontic appeared to lead to failure at the bracket adhesive interface. Knoll et al. I-" concluded that the weak link in the bond to posterior teeth was caused by bracket design and variation in resin performance because of nonuniformity of resin thickness beneath the brackets. O'Brien et al. t3 are in agreement with Knoll and have suggested that the failure site of the direct bonding system is a product of base design and the adhesive material used. Most in vitro investigations have shown that the most common failure site is the bracket adhesive interface for metal brackets. ~''~4-'-" However, O'Brien et al.,~3 in an in vitro investigation with grooved bracket bases and an experimental visible light-cured resin, did not agree with these findings. It has been observed that when bond strengths are higher, failure will more often occur at the enamel adhesive interface. 7"t2"2~This shift in failure site has also been observed in cases in which brackets have been given special surface treatment such as etching. ~5.24.:5 In this study, Right-on showed significantly more failures at the mesh adhesive interface. Inadequate mixing and curing of the two components of the adhesive behind the mesh base may be responsible for this. ~9"-'6 Minimum film thickness is important for no-mix cements since complete polymerization is unlikely to occur in thicker films where mixing is inadequate. -'7 Barnes :s and Maijer et al. :9 have both commented on the role of oxygen inhibition of free radical polymerization. Air entrapment behind the mesh of a metal
8.74 2.16
13.5 ! .78
Table VI. Significance levels derived from a pairwise crosscomparison of the bond strengths of Right-on, Heliosit-orthodontic with precured composite, Transbond, and Transbond with precured composite
I ttPC T TPC
R
Liverpool, England This study examines methods of improving the bond strengths of the light-activated composites, Heliosit-orthodontic, and Transbond. Begg brackets were bonded onto each of the four surfaces of 50 molar teeth that had been previously extracted. The bonding systems used were Right-on (R), Heliosit-orthodontic (H), Heliosit with a precured composite resin on the mesh (HPC), Heliosit with a precured intermediate bonding resin on the mesh (HPR), Transbond (T), and Transbond with a precured layer of resin on the mesh (TPC). The shear bond strengths were tested on a M5K tensile tester, crosshead speed 0.5 mm/min. After bracket removal, the enamel surface of the bonding site was examined and assessed with the adhesive remnant index (ARI). The shear bond strength in mean mPa __ 1 SD was found to be significantly greater for HPC, HPR, and TPC than for H (P < 0.001). The ARI scores suggest that bond failure is associated more with the micromechanical bond with enamel for HPC and HPR. The results indicate that the chemical properties of the two light-activated adhesives were improved by precuring on the mesh base of the bracket before bonding. (AM J ORTHOD DENTOFACORTHOP 1992;102:418-26.)
T h e development of adhesives that will satisfactorily bond orthodontic attachments directly to enamel has been greatly influenced by the research work directed toward improving adhesive properties of materials used in conservative dentistry.~ Light-cured materials are an example of such product development. While a strong and durable bond is required, the problem of removing the bracket without damaging the labial enamel must not be overlooked. The ideal bonding material sandwich should fail during debonding by the clean separation of the resin from the etched enamel. 2 The in vitro use of light-cured materials for orthodontic bonding was first described in 1979, 3 and laboratory trials were completed 5 years later. 4 In the direct bonding technique, the material is cured under metalbased brackets by transillumination as thetooth structure transmits visible light very well. 3's In vitro investigations have suggested that light-cured materials compare favorably with chemically cured adhesives. 4 In a clinical investigation, 6 the survivorship of the chemically activated bonding agent Right-on (R) (TP From the Department of Clinical Dental Sciences, School of Dentistry, University of Liverpool. 8 / 1/32325 ,,
418
~ o .
Product, La Porte, Ind.) was compared with the lightactivated material Heliosit-orthodontic (H) (Vivadent, Schaan-Lichtenstein). The study compared the failure rate of 1366 attachments bonded by graduate students. This was 16% for the chemically cured material and 23% for the light-cured material, with both significantly greater on posterior teeth compared with anterior teeth. After this clinical trial, 6 an in vitro study 7 investigated failure of bonds made with Concise (3M HealthCare, St. Paul, Minn.) and Right-on, two chemically activated systems, and the light-activated system Heliosit-orthodontic. The light-activated composite provided the weakest bond to enamel, and bonds on molar teeth were weaker than bonds on incisors, canines, and premolars. The clinical trial 6 suggested that the increased failure rates in the posterior segments were due to difficult access or salivary contamination. This hypothesis appears unjustified. A clinical investigation s using an experimental hybrid type of light-cured material with a filler loading of 62.8% demonstrated a failure rate of 4.7% compared with a chemically cured material at 6%. Before applying the filled resin phase to the bracket, a low viscosity resin phase was precured onto the bracket base in a procedure first prescribed by Tavas. 9
Volume102
Number 5
Bond strength of light-cured composites 419
3
Fig. 1. Diagrammatic representation of method of allocating identification number to bracket site.
OBJECTIVES The objective of this study was to investigate further the properties of Heliosit-orthodontic, and then to compare these with the light-activated material Transbond (T) (Unitek, 3M, St. Paul, Minn.). The molar tooth was chosen as the in vitro model system since the previously lower bond strengths found in shear 7 would permit a more critical assay. The aim was to assess the behavior of light-cured materials and their use with metal brackets and, in particular, to investigate the effects of pretreatment of the bracket base on bond strength of these materials. In addition, the site of bond failure under shear was to be assessed in the context of bond strength with these different modalities and thus further describe bonding to both bracket mesh and etched enamel.
MATERIALS AND METHODS Extracted human molar teeth with sound enamel were stored in distilled water at room temperature. Thirty teeth were mounted in dental stone within hardwood blocks .7 Each of the four comers of the molar teeth was numbered for the purposes of identification. For each specimen, the identification number was advanced in a downward and forward direction throughout this series. Thus on the first tooth the mesiobuccal bonding site was allocated the number 1, while the distobuccal bonding site received this identification number on the second in the series (Fig. 1). The bonding sites on each tooth were then polished with a rubber cup and a pumice and water paste. The surface was then rinsed with water and dried with warm air. Small curved Begg brackets (TP Products, La Porte, Ind.) were cleaned in 70% alcohol, followed by a rinsing in distilled water, and then allowed to dry. The brackets (nominal surface area 9.92 mm'-) were then bonded to the "appropriate bracket site" by one of the following four methods. The bonding procedures were carried out by the same operator using a standard technique. To a~,oid deficiencies around the bracket margins, an excess of material
Fig. 2. Molar tooth mounted in stone in hardwood block with Begg bracket bonded to each of four bonding sites.
was used and this was extruded around the entire periphery of the base on seating. A probe was then used to remove all excess material before polymerization (Fig. 2). Three adhesives were used. Right-on is a chemically activated system comprising a filled diacrylatc resin and an unfilled low viscosity resin containing the activator that is applied to the bracket base and enamel. Polymerization is initiated when the two resins come into contact. The other two materials were light-activated. Heliosit-orthodontic is a diacrylate micro-filled composite resin. It consists ofa methone dimcthyacrylate resin with a 14% filler of pyrolytic sillicium dioxide. Transbond is a microfine composite, which uses a hybrid filler of submicron silica. The average particle size is 3 microns, and the filler loading is approximately 82%. For systems using white light, a diketone is used in conjunction with an organic amine. The dikctonc absorbs blue light in the 420 to 450 nm range, and an excited triplet state is produced, which, together with the amine, results in ion radicals to initiate polymerization.'~
Bracket site 1 (R), No-mix composite resin (Right-on) The bonding site was etched for 60 seconds with Righton etching fluid containing phosphoric acid, rinsed with copious amounts of water and dried with warm air. A bracket was bonded with the composite resin according to the manufacturer's instructions.
Bracket site 2 (H), Heliosit-orthodontic light-cured resin The bracket site was etched for 90 seconds with the proprietary phosphoric acid etching fluid provided. After rinsing and drying, a bracket was bonded with the composite resin. This was cured for two periods of 20 seconds according to the manufacturer's instructions.
420
Am. J. Orthod. Oentofac. Orthop.
Bradburn and Pender
November 1992
/
1
The specimen block was set in the lower jaw of the MK tensile tester (JJ Instruments, United Kingdom) (Fig. 3). With a crosshead speed of 0.5 mm/min, the attachments were shear tested to failure. Shear testing was along an axis parallel to the long axis of the tooth. The point at which failure occurred was noted in Newtons and bond strength calculated in mega Pascals by dividing the force in Newtons by the area of the mesh in milllimeters squared. In a second series of experiments 20 molars were prepared for bonding in a similar manner to that already described. These were used to test the bond strength of the light-cured material Transbond. The following four methods were used. Bracket site 1 (R), no-mix composite resin (Right-on). Bracket site 2 (HPC), Heliosit-orthodontic with precured layer of tleliosit-orthodontic. Bracket site 3 (T), Transbond light-cured resin. The bracket site was etched for 60 seconds with the proprietary phosphoric acid etching gel. After rinsing and drying with warm air, a bracket was bonded with the composite resin. This was cured with the Heliosit white light for two periods of 15 seconds according to the manufacturer's instructions. Bracket site 4 (TPC), Transbond with precured layer of 'Transbond'. The bracket site was prepared and etched as for those in group 3. ttowever, a thin layer of Transbond resin was applied to the mesh base and precured according to the manufacturer's instructions before the routine bonding procedure described in group 3 (T).
Site of bond failure
Fig. 3. MK. Tensile tester with specimen block set in lower jaw.
Bracket site 3 (HPC), Heliosit-orthodontlc with precured layer of Heliosit-orthodontic The bracket site was prepared and etched as for those in group 2. However, a thin layer of resin was applied to the mesh base and precured according to the manufacturer's instructions, before the routine bonding procedure described for site 2 (H).
Bracket site 4 (HPR), Heliosit-orthodontic wffh precured Intermediate bonding resin (Heliobond) The bracket site was prepared and etched as for those in group 2. Before placement of the bracket with Heliosit-orthodontic as previously described, a thin layer of unfilled intermediate resin was applied to the mesh base and polymerizcd according the manufacturer's instructions. The teeth were set aside in a dry atmosphere for 24 hours before shear testing.
Shear strength testing A 0.014-inch ligature wire was placed through the attachment, and both ends were clamped in the shear tester.
The site of bond failure within the tooth-composite-mesh sandwich was assessed after shear testing with the adhesive remnant index (ARI) of Artun and Bergland." This was designated as a clinical technique for assessing the amount of composite remaining on the enamel after bond failure. The index has four categories: I. Score 0 = No adhesive remaining on the tooth. 2. Score I = Less than half of the adhesive on the tooth. 3. Score 2 = More than half of the adhesive on the tooth. 4. Score 3 = All adhesive on the tooth with a distinct impression of the bracket mesh. For these estimates, the material type used on the four bracket sites was unknown to the assessor at the time of examination.
Statistical analysis The parametric data were tested for significance with the pairwise t test. The data from the ARI were in the form of categories. Parametric tests are inappropriate, and the chisquared test for frequencies of occurrence was used for this data. In all tests of significance, alpha was set at 0.05.
RESULTS In the first e x p e r i m e n t , 4 o f the 30 teeth prepared for testing in h a r d w o o d blocks fractured during shear testing. A n o t h e r two samples fractured after testing o f the second and third bracket sites, respectively, thus providing i n c o m p l e t e data. Thus the data analyzed was
Volume 102 Number 5
Bond strength of light-cured composites
Bond Strength.(rnPa.) 16
14 ......................... m 12
421
................
E x p e r i m e n t . t ..... [ ~
Experiment .2 ............................ F'7//.zJ
7`
10 8 /I'
;
;5..;
6 4
. . . . . . . . . . . . . . .
2
~r ~~m m iN ~ H'
R
HPC '
. . . . . . . . . .
................. HPR
.......... I', T
TPC
Method of Bonding~ Fig. 4. Bond strength: Histogram showing mean for each of six methods of bonding.
Table I. Mean ( -+- I SD) bond strength (mega-pascals) of Right-on (R), Heliosit-Orthodontic (H), Heliosit-
orthodontic with precured composite (HPC), and Heliosit-orthodontic with intermediate resin (IIPR)
I Mean SD
R
I
8.45 2.17
derived from the four sites located on 24 teeth. The mean bond strength for each of the bonding systems is shown (Table I and Fig. 4) and the significance of their cross comparison listed in Table II. The highest bond strengths were achieved by precuring material onto the bracket base before carrying out the direct bonding procedure (HPC 9.95 mPa, HPR 8.87 mPa). Heliosit-orthodontic, used without any precuring treatment, consistently gave the weakest bonds (p < 0.001). Right-on had a similar bond strength (8.45 mPa) to Heliosit-orthodontic when this material was pretreated with Heliobond, the intermediate bonding resin (8.87 mPa). The ARI as an in vitro method of assessing the amount of material on the tooth surface is highly reproducible (r = 0.99). The ARI scores for each of the 96 sites is shown (Table III and Fig. 5) and the significance of the cross comparison of the four bonding systems is tabulated in Table IV. Most material remains on the enamel when using Right-on and Heliosit-orthodontic. In both methods that involved precuring material onto the bracket base, the reverse was found with the site of bond failure in the latter two groups associated with the adhesive-enamel interface. The second series of experiments, using a similar
"
I
I
7.22 2.61
,,pR
9.95 2.82
8.87 2.75
II. Significance levels derived from a pairwise crosscomparison of the bond strengths of Right-on, Heliosit-Orthodontic, Heliositorthodontic with precured composite, and Heliosit-orthodontic with intermediate resin (ttPR)
Table
I H HPC HPR
R P < O.OOl P < 0.001 NS
I
" P < 0.001 P < 0.001
I
.pc P < 0.001
method on 20 teeth, included Transbond and Transbond with a precured resin layer. The remaining two sites were occupied by Right-on and Heliosit-orthodontic with a precured layer of that material. Thus these latter two bonding systems were common to both sets of experiments. Only one sample fractured, after testing the second bracket site in the sequence. Therefore data were available from 19 teeth with four sites on each tooth. The bond strengths from these tests are shown (Table V and Fig. 4) with the significance of the crosscomparisons shown in Table VI. The bond strengths of Right-on and Heliosit-orthodontic (HPC) were consis-
422
Bradburn and Pender
Am. J. Orthod. Denmfac. Orthop. November 1992
Mean A.R.I. score. aT
. . . . . . . . . . . . . . .
i
~
Experiment 1
~
~
Experiment 2
:'
2.6
1.5
0.5
I
I
I
Method of Bonding. Fig. 5. ARI scores: Histogram showing mean ARI scores for six methods of bonding.
Table III. Sites of fracture described by the ARI of brackets bonded with Right-on, Heliosit-orthodontic,
Heliosit-orthodontic with precurcd composite, and Heliosit-orthodontic with intermediate resin 1
1 R tt ltPC It PR
I 1 5 5
4 3 12 15
Table IV. Significance levels derived from a
pairwise crosscomparison of site of fracture assessed by the ARI of brackets bonded with Right-on, Heliosit-orthodontic, Heliositorthodontic with precured composite, and Heliosit-orthodontic with intermediate resin
I H HPC HPR
R NS P < 0.001 P < 0.001
I
" P < 0.001 P < 0.001
I NS
tent with those found in the previous experiment. Transbond compared favorably with Right-on (Table VI and Fig. 4). Transbond precured with resin showed the greatest bond strength, which was also significantly greater than precured Heliosit-Orthodontic (HPC~. Thus higher bond strengths seemed to be associated more with the process of precuring on the bracket base than with a particular material.
2
3
Total score
Mean score
7 4 6 4
12 16 I 0
54 59 27 18
2.25 2.45 1.12 0.96
Table VII and Fig. 5 record the ARI scores for the 76 sites examined, and Table VIII shows the significance of the erosscomparison of these scores between the four methods of direct bonding used in this second part of the study. The site of bond failure with precured IIeliosit (HPC) occurred more often at the adhesive enamel interface, which is consistent with the previous experiment. Transbond, in common with Right-0n and Heliosit-orthodontic in the previous experiment, showed bond failure more often associated with the adhesive-bracket interface. The relationship between bond failure and the site of failure for precuring with Transbond (TPC) is less clear with a significant number of cohesive failures located within the composite resin. DISCUSSION
The clinical trial by Lovius et al., 6 showed that Iteliosit-orthodontie failed more often than Right-on on all teeth other than incisors. The laboratory trial by Pender et al., 7 confirmed tleliosit-orthodontic as a weaker material than Right-on and that anterior bond
Voh,me 102 Number 5
Bond strength of light-cured composites
423
Table V. Mean (+_ 1 SD) bond strength (mega-pascals) of Right-on, Heliosit-orthodontic with precured
composite, Transbond (T), and Transbond with precured composite (TPC)
Mean SD
8.22 2.20
10.53 1.53
strengths were greater for both materials than those for posterior teeth. Thus a similar pattern emerged with both in vivo and in vitro testing of the light-cured material. In the clinical trial, 6 it was suggested that poor moisture control, poor access with the Heliosit light, and an increased amount of aprismatic enamel on posterior teeth might be responsible for higher clinical failure rates. In the laboratory trials, access with the Heliosit light and moisture control were not significant factors. For aprismatic enamel to be an important consideration in bond failure, the micromechanical bond with enamel would have to fail more readily with tleliosit-orthodontic on posterior teeth than with other materials. However, the in vitro study of Pender et al. 7 showed that this was not the case as the weaker bond strength of Heliosit-orthodontic appeared to lead to failure at the bracket adhesive interface. Knoll et al. I-" concluded that the weak link in the bond to posterior teeth was caused by bracket design and variation in resin performance because of nonuniformity of resin thickness beneath the brackets. O'Brien et al. t3 are in agreement with Knoll and have suggested that the failure site of the direct bonding system is a product of base design and the adhesive material used. Most in vitro investigations have shown that the most common failure site is the bracket adhesive interface for metal brackets. ~''~4-'-" However, O'Brien et al.,~3 in an in vitro investigation with grooved bracket bases and an experimental visible light-cured resin, did not agree with these findings. It has been observed that when bond strengths are higher, failure will more often occur at the enamel adhesive interface. 7"t2"2~This shift in failure site has also been observed in cases in which brackets have been given special surface treatment such as etching. ~5.24.:5 In this study, Right-on showed significantly more failures at the mesh adhesive interface. Inadequate mixing and curing of the two components of the adhesive behind the mesh base may be responsible for this. ~9"-'6 Minimum film thickness is important for no-mix cements since complete polymerization is unlikely to occur in thicker films where mixing is inadequate. -'7 Barnes :s and Maijer et al. :9 have both commented on the role of oxygen inhibition of free radical polymerization. Air entrapment behind the mesh of a metal
8.74 2.16
13.5 ! .78
Table VI. Significance levels derived from a pairwise crosscomparison of the bond strengths of Right-on, Heliosit-orthodontic with precured composite, Transbond, and Transbond with precured composite
I ttPC T TPC
R
