INTRODUCTION

Following the introduction of acid etching of enamel,1 orthodontic brackets are now bonded routinely with resin adhesives to incisors, canines, and premolars as part of fixed appliance therapy.2–4 Modifications in adhesive formulation over the past 3 decades have led to current availability of 2 paste systems,5 no-mix adhesives,4 light-activated direct bonding materials,6,7 and brackets precoated with adhesive.8

Glass ionomer cements also have been used for orthodontic bonding, but they have inferior bond strengths compared to resin adhesives.9 However, modifications in cement formulation through the addition of resin have aimed to address the problem of the moisture sensitivity of composites and the low early mechanical strength of glass ionomers by allowing light curing, a snap set, and rapid strength development.10 Because wide variation exists in the chemical constituents and setting reaction of individual formulations, some products have been termed resin-modified glass ionomer cements, and others have been labeled as modified composites or compomers.11

Bonding of attachments to molars, rather than banding, is a less frequently adopted practice5,12 despite the periodontal advantages it confers over molar bands.13 Brackets bonded to molars have had a lower bond strength14 and a higher clinical failure rate2,15 than those brackets bonded to teeth located more anteriorly in the arch. Molar tubes bonded with either a chemically cured15 or a light-cured resin adhesive16 have exhibited failure rates of over 21%. The inferior quality of the etch pattern achieved on molars,17,18 the difficulty in achieving and maintaining adequate moisture isolation during bonding,14 inadequate adaptation of the attachment base producing an uneven adhesive layer,19 and the larger masticatory forces posteriorly in the mouth16 are potential contributors to this poor clinical performance.

When preformed cylinders of a chemically cured composite (Concise, 3M Dental Products, St Paul, Minn) were bonded to the buccal surfaces of molars, the probability of bond survival increased when the etch time was no shorter than 30 seconds, but it was not significantly improved by etching for 60 seconds.20 Although that study gives useful pointers with regard to bonding to molars, it relates exclusively to a chemically cured resin adhesive. Further information, however, is required with regard to the use of light-cured bonding agents, which allow command setting of the resin and which are probably used more commonly for this purpose.

Light-cured resin-modified glass ionomer cements and modified composites have been used to bond brackets.21,22 Each cement type has demonstrated a bracket failure rate comparable to that of resin adhesive.22,23

Recent modifications in molar tube fabrication have led to the production of bonding bases contoured to the buccal molar enamel, which facilitates placement and is likely to enhance adhesion with a bonding agent and possibly to promote better bond reliability. In addition, the bonding bases have been micro-etched, which has been shown to improve bond strength.24 The use of these new-design molar tubes in conjunction with the newer light-cured bonding agents has not been evaluated previously.

The aims of this study, therefore, were to assess the mean shear bond strength of molar tubes with micro-etched bases bonded with compomer, a resin-modified glass ionomer cement, or a light-cured resin adhesive control. The amount of adhesive remaining on the enamel surface following debonding was evaluated also. Finally, survival time of molar tubes bonded with each bonding agent was assessed following simulated mechanical fatigue in a ball mill.

MATERIALS AND METHODS

To assess debonding force, 80 sound-extracted human third molars were collected and stored in distilled water in a refrigerator following decontamination in 0.5% chloramine. The teeth were randomly divided into 4 groups of 20 teeth, each group comprising 10 maxillary and 10 mandibular third molars. Each tooth was notched in the apical third and then mounted to below the cemento-enamel junction in a block of self-curing acrylic, with the long axis of each tooth vertical.

The teeth were cleaned with pumice slurry, washed in distilled water, and dried in a stream of air. A molar tube with a micro-etched base (3M Unitek, Monrovia, Calif) was bonded to each molar. Each tube had a buccal groove indicator to facilitate placement and a contoured base to aid adaptation to the buccal enamel surface.

Tubes were kept in the manufacturer's packaging until immediately prior to bonding and were handled at all times with bonding tweezers to avoid contamination of the bonding base. Twenty tubes were bonded with each material. Four orthodontic bonding agents were investigated—a compomer (Ultra Band-Lok, Reliance Orthodontic Products Inc, Itasca, Ill); 2 resin-modified glass ionomer cements (3M Multi-Cure, 3M Unitek, and Fuji Ortho LC, GC America Inc, Chicago, Ill), and a light-cured resin adhesive (Transbond, 3M Unitek).

Ultra Band-Lok is a single-component compomer resin formed by combining a composite resin with glass ionomer particles. Supplied in sealed capsules from which it is applied to the bonding base, it hardens only through photopolymerization. 3M Multi-Cure is a powder-liquid–based resin-modified glass ionomer cement that sets via an acid-base reaction of the glass ionomer component, a free radical addition polymerization reaction activated by visible blue light, and a self-cure of the resin monomer. Fuji Ortho LC is a resin-modified glass ionomer cement supplied in foil-wrapped capsules, each containing powder and liquid in separate compartments. Following mechanical mixing of the components for 10 seconds, an acid-base reaction is initiated. Similar to 3M Multi-Cure, there is also a self-curing of the resin monomer and a light-activated polymerization. Transbond is a single-component, visible light–cured resin adhesive, supplied in capsules.

Enamel etching for 30 seconds with 37% orthophosphoric acid gel was undertaken on the midbuccal aspect of each molar before bonding with Ultra Band-Lok, Transbond or Fuji Ortho LC. The enamel surface was maintained moist on teeth to be bonded with Fuji Ortho LC, this procedure having been shown to yield optimal bond strength.25 A 30-second etch time was adopted, as it has been shown to produce a consistent bond strength when bonding to the buccal surfaces of molars.20 For teeth to be bonded with 3M Multi-Cure, no enamel etching was undertaken, in accordance with the manufacturer's recommendations. Instead, the enamel surface was rubbed lightly dry with a cotton wool roll but was maintained moist. Otherwise, cement mixing, where appropriate, and light curing were performed according to the manufacturer's instructions. In order to eliminate the influence of operator variability on bond reliability, a single operator carried out all bonding procedures.

Excess cement or adhesive was removed from around the periphery of each molar tube immediately following tube placement. For each bonding agent, light curing was undertaken for 40 seconds: 10 seconds each from the mesial, distal, incisal and gingival aspects of each tube using an Ortholux light source (3M Unitek).

Following curing, all specimens were transferred to a humidor set at 37°C for 24 hours prior to measuring the shear debonding force using a Nene N3000 testing machine (Nene Instruments Ltd., Wellingborough, Northants, UK) with a crosshead speed of 1 mm/min. A piece of 0.8-mm stainless steel orthodontic wire was placed under the gingival aspect of each molar tube and connected to the upper arm of the Nene testing machine. To calculate bond strength, the debonding force values (N) were converted to MPa (N/mm²) by taking account of surface area data provided by the manufacturer.

After each bonded tube failed, the amount of adhesive remaining on the enamel surface was coded using the criteria proposed in the adhesive remnant index (ARI) of Årtun and Bergland:²⁶

• 0 = no adhesive remains on the tooth surface.

• 1 = less than half the adhesive remains on the tooth surface.

• 2 = more than half the adhesive remains on the tooth surface.

• 3 = all the adhesive remains on the tooth surface.

To assess survival time, another group of 40 extracted human third molars was collected. These teeth were treated and stored in a manner identical to that used on the teeth to be debonded. They were divided into 4 groups, each with 5 maxillary and 5 mandibular third molars. The palatal or lingual enamel surface of each tooth to be bonded with each bonding agent was coded with a diamond bur to facilitate identification later. Ten tubes were bonded with each material in accordance with manufacturers' instructions, as outlined previously, and specimens were then stored in a humidor at 37°C for 24 hours.

Specimens were subjected to mechanical stress for 50 hours in a ball mill containing 470 g of ceramic spheres and 250 mL of distilled water at 37°C. After each hour of testing at 100 revolutions/min, the failed specimens (those where the tube debonded) were removed from the mill. The distilled water was then replaced with a fresh sample at 37°C, and testing recommenced.

The mean shear debonding force values for the bonding agents tested were compared using analysis of variance followed by a Tukey's multiple comparison procedure. A Weibull analysis27,28 was used to calculate probability of failure at given values of applied force. The Weibull analysis may be more applicable to the evaluation of debonding force data than conventional comparison of mean and standard deviation values only, because it takes account of debonding force values at the extremes of the distribution. Generation of a Weibull modulus for each bonding agent allows numerical evaluation of its reliability.

For ARI scores, analyses were based on the χ² test followed by a multiple comparison procedure using the Fisher's exact test. For these analyses, it was necessary to combine score 0 and 1 categories and score 2 and 3 categories because of the small expected frequencies. A log rank test was used to compare survival time distributions of tubes bonded with each bonding agent in the ball mill experiment.

RESULTS

Shear bond strength data for each bonding agent are given in Table 1. The mean shear bond strength of tubes bonded with Transbond (3.04 MPa) was significantly less than those bonded with 3M Multi-Cure (P = .0036) or Fuji Ortho LC (P < .0001). Tubes bonded with Ultra Band-Lok also had a significantly lower mean shear bond strength than those bonded with Fuji Ortho LC (P = .020).

TABLE 1. Shear Bond Strength Values for 20 Molar Tubes Bonded With Each Bonding Agent

View Fullscreen

Weibull data are also shown in Table 1 and displayed graphically in Figure 1. The highest Weibull modulus was recorded with Fuji Ortho LC, indicating the greatest bond reliability with this bonding agent. The high values of correlation coefficient of linearized least squares fit indicate that the data fit closely the Weibull distribution function. Figure 1 indicates that, for a given probability of failure, significantly less force would be required to dislodge a molar tube bonded with Transbond compared to one bonded with either of the other bonding agents.

View Figure

• Download as Powerpoint
• Download Figure

The distribution of ARI scores for each bonding agent is given in Table 2. The only significant difference in distribution of ARI scores existed between tubes bonded with 3M Multi-Cure and those bonded with Transbond (P < .001).

TABLE 2. Distribution of ARI Scores for 20 Molar Tubes Bonded with Each Bonding Agent^a

View Fullscreen

Only 1 molar tube—bonded with Transbond—debonded in the ball mill after 5 hours. No other bond failures occurred, indicating no apparent difference between the bonding agents when subjected to mechanical stress in the ball mill over a 50-hour test period.

DISCUSSION

This laboratory investigation evaluated the mean shear bond strength of stainless steel molar tubes with micro-etched bases bonded with a compomer, a resin-modified glass ionomer cement, or a light-cured resin adhesive control. Although ground-enamel surfaces have been used in an attempt to produce a more standardized bonding substrate,29,30 this procedure was not adopted in this study or in a previous study20 on bonding to molars, because it is regarded as unrepresentative of clinical practice.31 No other study has assessed bonded molar tubes in a similar manner to that adopted here, although other studies have used similar bonding agents for bonding brackets.32–34

The mean shear bond strength of molar tubes bonded with either of the resin-modified glass ionomer cements, 3M Multi-Cure or Fuji Ortho LC, was significantly greater than that achieved with a light-cured resin adhesive, Transbond. This contrasts with the findings of previous studies, which have found the mean shear bond strength of brackets bonded with Fuji Ortho LC to be significantly less than33 or comparable to that of Transbond.33,35 Variation in tooth type, bracket base design, and enamel preparation are likely to account in part for these differences. The mean shear bond strength of tubes bonded with the compomer, Ultra Band-Lok, was also significantly less than that obtained with Fuji Ortho LC. No significant difference has been found, however, in the mean shear bond strength of brackets bonded with the compomer Dyract Orthodontic (De Trey, Dentsply, Konstanz, Germany) and those bonded with Fuji Ortho LC.33 Differences in cement formulation between Ultra Band-Lok and Dyract Orthodontic may contribute to the variation in outcome between these 2 studies.

A minimum bond strength of 5.9 MPa to 7.8 MPa has been suggested as adequate for most clinical orthodontic needs.36 On that basis, only tubes bonded with Fuji Ortho LC and possibly 3M Multi-Cure would be expected to give adequate performance clinically. Tubes bonded to molars, however, are likely to be subjected to greater occlusal forces than attachments bonded to teeth located more anteriorly in the mouth.16 Caution, therefore, must be exercised in extrapolation to the clinical situation based on bond strength data alone, as these data are of limited value.37

Of greater significance to the clinician is the need to ascertain whether the bond strength of each tube/adhesive combination can be exhibited reliably. It has been recommended that Weibull analysis be applied routinely to bond strength data on orthodontic bonding agents to obtain this information.31 The lowest Weibull modulus was obtained with Transbond, indicating poor bond reliability with this adhesive. Because Fuji Ortho LC exhibited the highest Weibull modulus (3.60 MPa), tubes bonded with this cement are likely to perform more reliably than those bonded with any of the other adhesives.

Following debonding, there was only a significant difference in ARI scores between tubes bonded with 3M Multi-Cure and those bonded with Transbond. Scores of 0 or 1 were predominant with 3M Multi-Cure, whereas mostly scores of 3 were recorded with Transbond, the latter confirming the findings of other studies.33,38 Less clean-up time is, therefore, likely to be required if tubes are bonded with 3M Multi-Cure rather than with Transbond. Tubes bonded with Ultra Band-Lok mostly had ARI scores of 2 or 3, indicating that most of the adhesive remained on the tooth surface at debond. This is not surprising, because this cement is a compomer and is likely to have ARI scores similar to those of Transbond. A compomer and the light-cured resin adhesive Transbond have been shown to exhibit similar ARI scores previously.33,38 In line with other investigations,32,34 Fuji Ortho LC mainly displayed ARI scores of 2 or 3, although the bonding base design and its surface characteristics have differed between these studies.

Although the results of bond strength tests and the Weibull analysis on orthodontic bonding agents are of interest, testing is invariably undertaken in carefully controlled conditions that bear little resemblance to the clinical situation.37,39 In the mouth, bond failure may result from the cyclical nature of mechanical, thermal and chemical processes that induce material fatigue.39 Efforts should be made to make laboratory evaluations more closely mimic the oral environment to provide more meaningful data for the practicing orthodontist.33,37

Bonded specimens have been subjected to thermal insult31 or to mechanical stress,32,38 the latter applied in a ball mill. Diverse forces of varying magnitude operate in the ball mill,40,41 with the impact force and mechanical action of the ceramic spheres likely to generate slow crack propagation in the bonding agent, which eventually leads to bond failure. In the present study, only 1 tube—bonded with Transbond—failed at 5 hours in the 50-hour ball mill trial. Although it may be unwise to attach undue importance to this finding, it is interesting that Transbond also had the lowest mean shear bond strength and the lowest Weibull modulus of the adhesives tested, indicating its greater likelihood of bond failure. A longer trial period would have been required to obtain data that are more precise in this regard. At 50 hours, however, there was no significant difference in bonded molar tube survival between the groups tested. Results obtained using the ball mill technique have indicated its usefulness in predicting the clinical performance of orthodontic band cements and bracket bonding agents.22,33,39 On that basis, tubes bonded with any of the bonding agents used in this study may have a similar performance clinically. One must realize, however, that in the study reported here, tubes were bonded to third molars, whereas they are likely to be bonded to first or second molars in the mouth. A clinical trial would be required to further verify the findings of this laboratory investigation.

CONCLUSIONS

The mean shear bond strength of molar tubes bonded with Transbond was significantly less than that of those bonded with 3M Multi-Cure or Fuji Ortho LC. Molar tubes bonded with Ultra Band-Lok had a significantly lower mean shear bond strength than tubes bonded with Fuji Ortho LC. The Weibull analysis indicated that, for a given force, the probability of bond failure was greater for tubes bonded with Transbond than for those bonded with any of the other bonding agents.

The distribution of ARI scores differed significantly for tubes bonded with 3M Multi-Cure compared to those bonded with Transbond. There was no significant difference in survival time of tubes bonded with any of the adhesives. Ultra Band-Lok, 3M Multi-Cure, and Fuji Ortho LC appear to be viable alternatives to Transbond for bonding molar tubes.