J. Orthod.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Journal of Orthodontics, Vol. 31, No. 1, 34-40, March 2004 doi:10.1179/146531204225011355
© 2004 British Orthodontic Society
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thind, B. S.
Right arrow Articles by Lloyd, C. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thind, B. S.
Right arrow Articles by Lloyd, C. H.

Article

An ex vivo assessment of gingivally offset lower premolar brackets

B. S. Thind, C. J. Larmour, D. R. Stirrups and C. H. Lloyd

University of Dundee Dental School, Dundee, UK

Address for correspondence: B. S. Thind, Unit of Dental and Oral Health, Orthodontics Section, Dundee Dental School, Park Place, University of Dundee, Dundee DD1 4HN, UK. Email: bikram_chaman{at}yahoo.com

Received March 6, 2003; accepted June 5, 2003


   Abstract
Top
 Abstract
Introduction
 Aims
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
Objectives: To compare the force to failure of standard premolar brackets to that of gingivally offset brackets and evaluate the site of bond failure between the two bracket types through the use of the Adhesive Remnant Index (ARI).

Design: An ex vivo study.

Setting: Dental Materials Science Laboratory, Dundee Dental School, Dundee.

Materials and methods: Forty extracted lower premolar teeth (caries free, extracted as part of orthodontic treatment, all donors living in a non-fluoridated area), divided into two equal size sample groups, as follows: Group 1: Victory SeriesTM (3M Unitek, Monrovia CA, USA) lower premolar brackets bonded to buccal surfaces with Transbond XT (3M Unitek, Monrovia CA). Group 2: Victory SeriesTM Gingivally Offset Bicuspid Brackets (3M Unitek, Monrovia CA) bonded to buccal surfaces with Transbond XT (3M Unitek, Monrovia CA). Force was applied in the occluso-gingival direction using an Instron Model 4469 Universal Testing Machine (Instron Ltd, High Wycombe, UK) operating at a cross-head speed of 0.5 mm/min and its value at failure determined. Following debond, the site of bond failure and ARI were recorded.

Outcome: Force to failure, site of bond failure and adhesive remnant index.

Results: The Weibull analysis gave higher values for the force to failure at 5% level (200 v. 159 N) and at all other levels of probability of failure for the gingivally offset bracket. The non-parametric survival analysis using Gehan–Wilcoxon tests with Breslow’s algorithm (p < 0.0001) showed significant difference in force to failure between bracket types. Chi-square tests showed no significant (p = 0.55) relationship between the site of bond failure and the bracket types.

Conclusion: Ex vivo testing suggests that there is a significant difference in the force to failure between gingivally offset and standard lower premolar brackets when force application is from an occluso-gingival direction. The site of failure (as given by the ARI) is insensitive to bracket types and force to failure.

Key words: Dentistry, orthodontics, orthodontic bracket bonding, gingival offset brackets


   Introduction
Top
 Abstract
 Introduction
Aims
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
The debonding of orthodontic brackets is a common problem in orthodontic practice. Ideally, a bracket should remain bonded to a tooth until the end of treatment and resist complex occlusal forces whilst in service. Bond failure rates between 2.7% and 6% have been reported.1,Go2Go

Various reasons for this bond failure have been cited. These include the bonding agent applied,3Go bonding technique used,4Go design of the bracket base,5Go etch time and concentration of etchant.6,Go7Go

Another important factor is occlusal trauma. This may stem from the type of malocclusion, e.g. cusp to cusp occlusion where brackets may fail when the patient brings the posterior teeth into contact, or from excessive occlusal forces when the patient fails to follow the instructions provided. Previous studies8Go have shown that lower premolar brackets are the most vulnerable to occlusal forces and most likely to debond during treatment. In order to reduce this problem a redesigned lower premolar bracket is now available, on which the wings of the bracket are offset gingivally to provide a greater bracket base area, occlusal to bracket wings. The manufacturer (3M Unitek, Monrovia CA, USA) claims that this arrangement should provide greater resistance to debonding, thus reducing the incidence of bond failure.


   Aims
Top
 Abstract
 Introduction
 Aims
Material and methods
 Results
 Discussion
 Conclusions
 References
 
The main aim of this ex vivo study was to compare the force to failure of the gingivally offset brackets lower premolar with that of standard brackets, by means of probabilistic approach (Weibull analysis). The null hypothesis is that there are no differences between gingivally offset brackets and standard brackets with respect to:

The second aim was to evaluate the site of bond failure between two bracket types. It is of interest to determine whether any change in the force to failure that results from bracket selection is accompanied by change in the site at which failure occurs. The ARI was selected as a simple, but informative semi-quantitative method to provide this evidence. The null hypothesis is that there is no difference in the distribution of ARI scores between bracket types or force to failure levels.


   Material and methods
Top
 Abstract
 Introduction
 Aims
 Material and methods
Results
 Discussion
 Conclusions
 References
 
Forty sound lower premolar teeth, extracted for orthodontic purposes from patients under the age of 18 years living in an area with a non-fluoridated water supply, were collected and stored in distilled water. They were mounted in resin blocks with the long axis of each tooth set vertically. These specimens were divided randomly into two equal size groups. This sample size complies with the recommendation of Fox et al.9Go that at least 20 teeth per test be used for ex vivo bond strength testing:

Group 1. Standard pre-adjusted edgewise lower premolar brackets (Victory SeriesTM, 3M Unitek, Monrovia CA, USA). Bonding product — Transbond XT (3M Unitek, Monrovia CA, USA). This group served as the control. Sample size = 20.
Group 2. Gingivally offset brackets lower premolar brackets (Victory SeriesTM Gingivally Offset Bicuspid Bracket, 3M Unitek, Monrovia CA, USA). Bonding product — Transbond XT (3M Unitek, Monrovia CA, USA). Sample size = 20.

A single operator carried out all bonding. The materials were used according to the manufacturer’s instructions: The teeth were pumiced using fluoride-free pumice and water for 15 seconds with a rubber cup, then rinsed with water and dried in a stream of oil-free compressed air. The teeth were then etched for 30 seconds with orthophosphoric acid etchant (37%) provided by the manufacturer, washed for 60 seconds then dried using oil-free compressed air. A thin layer of primer was applied to each tooth with a microbrush. The bracket was loaded with adhesive paste and placed on the buccal surface with light pressure exerted to extrude any excess adhesive paste. This excess was removed with a probe. The composite was cured with a 60-second light exposure (Visilux 2TM, 3M-ESPE, St Paul, MN, USA). The application of light was 15 seconds each from occlusal, gingival, mesial and distal direction. The bonded specimens were stored in distilled water at 37 °C for 1 week before determining the force to failure.

An Instron Model 4469 Universal Testing Machine (Instron Ltd, High Wycombe, UK) was used to measure the force to debond. To simulate the intra-oral failure of brackets due to occlusal trauma, they were debonded with the force applied in an occluso-gingival direction using a flat-ended steel rod. One end of the rod was fixed rigidly to the moving cross-head. The test jig by which the specimen was attached to the stationary anvil allowed the specimen position to be adjusted in the x–y plane and then locked into position. This enabled the other end of the rod to be placed precisely and consistently over the bracket between the occlusal tie wings of each bracket, to ensure the distance between the surface of the tooth and point of application of force was same for each specimen (Figure 1Go). A cross-head speed of 0.5 mm/min was used and the force required to dislodge the bracket measured to a resolution of 0.1 N. Debonding tests were conducted in air at ambient laboratory temperature.



View larger version (31K):
[in this window]
[in a new window]
  		Figure 1 The specimen and its positioning in the loading apparatus, viewed in the mesial direction. This shows the adjustment in the x direction to position the bracket under the loading rod. A similar adjustment exists in the y direction, for the same purpose

 
Following debond, each tooth was examined by a single operator at x 10 magnification. The site of bond failure was recorded along with the Adhesive Remnant Index. (ARI).10Go In order to check the reliability of ARI scoring 20 random samples were scored by the first and the second authors independently on two separate occasions to calculate the inter- and intra-rater reliability. There was 100% agreement on both occasions by the first author and a single disagreement between occasions by the second author giving his intra-rater kappa of 0.77 indicating substantial agreement.11Go In regard to inter-rater reliability, on the first occasion a single disagreement between authors gave a kappa value of 0.77. However, on the second occasion, there was 100% agreement between the two authors.

The statistical analyses were carried out using Unistat 5.0 (Unistat Ltd, London, UK). Weibull analysis was carried out to relate the probability of bond failure to the applied force. This analysis has been advocated previously.9Go Non-parametric survival analysis was carried out. The ARI data were analysed with chi-squared tests.


   Results
Top
 Abstract
 Introduction
 Aims
 Material and methods
 Results
Discussion
 Conclusions
 References
 
The results of the study are presented graphically in Figure 2Go. The histogram shows the distribution of force to failure values of both bracket types in Newtons. The standard brackets debonded consistently at lower force level than gingivally offset brackets.



View larger version (14K):
[in this window]
[in a new window]
  		Figure 2 Histogram showing the frequency distribution of force to failure of the two bracket types

 
The Shapiro–Wilk test12Go of normality was carried out on both groups of force to failure data. As Table 1Go shows, Group 2 has results compatible with a normal distribution. Group 1 did not and further examination of the data revealed one extreme outlier. (A check of the original machine output confirmed this value and excluded a data entry error). Group 1 was only just non-significant for non-normality, but Tiku13Go has shown that Shapiro–Wilk test is not particularly powerful. It is therefore unsafe to assume normality when the test produces marginally non-significant results.14Go


View this table:
[in this window]
[in a new window]
  		Table 1 Shapiro–Wilk test for normality
 
Re-application of the Shapiro–Wilk test of normality with the exclusion of this single value shows compatibility with normal distribution. However, it was an abnormally low value and the authors feel that it is inappropriate to exclude this ‘inconvenient’ data point from further analysis.

Weibull analyses15Go were carried out and Weibull plots for the two groups are shown in Figure 3Go and the statistics in Table 2Go. Group 1 has a slightly lower Weibull modulus. Given the non-normality of some of the data, median and 95% CI exact conservative intervals about the median are provided in preference to the mean and normal distribution-based confidence intervals. Examination of these confidence intervals shows that they do not overlap for Groups 1 and 2. Thus, Groups 1 and 2 differ from each other with respect to the applied force to failure. The force to failure required for various probabilities of failure are in Table 3Go.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 3 The Weibull plots for the force to failure for the two bracket types. — offset bracket, — normal bracket

 

View this table:
[in this window]
[in a new window]
  		Table 2 Weibull analysis
 

View this table:
[in this window]
[in a new window]
  		Table 3 Force required for specific probabilities of failure
 
Non-parametric survival analysis was carried out using Gehan–Wilcoxon tests with Breslow’s algorithm16Go to confirm the deductions from the confidence intervals. Tests were carried out comparing Groups 1 and 2. The results are in Table 4Go. Additionally, the test was repeated for Groups 1 and 2 excluding the Group 1 outlier. This shows that excluding the outlier does not significantly affect the result and that the outlier is not responsible for most of the differences found. The null hypothesis that the force to failure at 5% level of probability of failure is same for both bracket types is rejected.


View this table:
[in this window]
[in a new window]
  		Table 4 Survival analysis
 
The bracket resin interface was the more common site of bond failure for both the groups. Thirty-seven specimens failed at the bracket resin interface and remaining three failed at enamel resin interface (Table 5Go). Table 6Go shows the Chi-square test results for bracket types and ARI scores, indicating no significant difference. The null hypothesis that there is no difference in the distribution of ARI scores between bracket types is accepted.


View this table:
[in this window]
[in a new window]
  		Table 5 Site of bond failure of each group
 

View this table:
[in this window]
[in a new window]
  		Table 6 Chi-square test for bracket type and ARI scores
 
Further investigation was carried out in order to determine whether there is a relationship between ARI scores and the amount of force required to debond brackets. The specimens were redistributed into ARI score groups. The number of specimens in one group was too small for statistical testing. Therefore, it was not possible to test the null hypothesis. The median and inter quartile intervals for the ARI score groups are shown in Table 7Go.


View this table:
[in this window]
[in a new window]
  		Table 7 ARI scores and related force to failure for all specimens
 

   Discussion
Top
 Abstract
 Introduction
 Aims
 Material and methods
 Results
 Discussion
Conclusions
 References
 
The experiment
The gingivally offset lower premolar bracket has been designed by the manufacturer to reduce bracket failure from occlusal forces. The results of this ex vivo study suggest that a gingivally offset lower premolar bracket has a significantly higher force to failure than conventional lower premolar brackets. This difference could be due the greater bracket base area; this bracket has 25% greater bracket base area than the standard bracket. (According to the manufacturer, the bracket base area of the standard bracket is 10.57 mm2 and that of the gingivally offset bracket is 13.96 mm2). Previous studies17Go have reported that the bracket base area has minimal effect on bond strength, but this conclusion is based upon the stress at failure and not the force to failure. While failure is initiated when the stress experienced by the bond reaches a critical value, this stress is the result of the trauma force being distributed over the bracket base area. It is the maximum trauma force that can be withstood before failure that is clinically relevant and any comparison of brackets that have different designs should be made using the force to failure as the criterion.

Reynolds18Go has suggested that a minimum bond stress of 5.9 to 7.8 Nmm–2 would be adequate for most clinical orthodontic needs. Arguably, it is inappropriate to use stress instead of force to quantify failure that takes place at the resin/bracket interface since this interface is convoluted. Nevertheless, the force is transmitted through the resin to the resin/enamel interface that is considered planar and at which stress to cause failure is the accepted criterion (as, for example, in restorative dentistry).19,Go20Go Reynolds’ stress range is equivalent to a force to failure range of 63–109 N for the brackets used in the present study. The results presented here show that when the debonding force applied in occluso-gingival direction, the force to failure is much higher than this ‘recommended’ level. When the site of failure is considered, the majority of these were at the bracket/resin interface. This is consistent with other studies21,Go22Go that have shown that brackets bonded with adhesive resins systems cured with visible light tend to fail at the bracket/ resin interface. Though the force to failure showed variability (pooled data — minimum force 125 N, maximum 328 N), overall no relationship was found when the force failure was compared with the site of failure for the two groups, as evident from Table 7Go. The tensile bond stress has been recommended23Go not exceed 14.5 Nmm–2 to prevent enamel fracture. This can be translated to a force to failure of 153 and 202 N for the standard and gingivally offset brackets, respectively. Though such values would result in a low chance of failure (5%), no enamel fractures were seen and the bracket/resin interface was the predominant failure site.

Conclusions drawn from any laboratory test must be applied with caution since it is normal to minimize the number variables through the design of the ex vivo experiment to allow the effect of change in a specific variable to be studied. Often, it is possible to simulate conditions that are close to those in clinical use, but the potential for unrecognized factors to modify the outcome should always be borne in mind. Thus, the results of this ex vivo force to failure study should be treated as a first positive indication of a superior performance by gingival offset brackets. In the clinical situation, the forces that act on the bracket are more complex.24Go A clinical trial is the logical next step and, in progress, to compare in vivo failure rates.

Statistical methods
The statistical methods used were determined by the need to avoid tests that depend upon the normal distribution of the data. A single extreme outlier in one group was on the low side of force required to produce failure. Since a low force for failure would be clinically significant its exclusion could have biased the results towards better performance. In fact, as the subsequent analysis showed, its inclusion or exclusion did not significantly affect the outcome. Medians and distribution-free exact estimates of 95% CI were used to avoid the problems of non-normal distribution for conventional means and parametric confidence intervals. The use of confidence intervals is preferable to statistical testing, since inspection shows whether overlapping intervals exist and, hence, if differences are likely. Certainly, they usefully pointed the way for subsequent analysis.

Failure time and survival analysis statistical techniques are directly applicable to this study, as the force to failure can be treated in the same way as time to failure. The parametric method of the Weibull analysis is preferred in the absence of missing or censored data, but is still to a degree dependent on the goodness of fit of the data to a Weibull distribution. Goodness of fit tests showed adequate, but as is usual with small samples, not great fit. Therefore, a non-parametric survival analysis, Gehan–Wilcoxon tests with Breslow’s algorithm was used as well. The Weibull moduli suggest little scattering of the results in the groups not withstanding the outlier. The normalizing parameter (or characteristic force) is a measure of central tendency used in Weibull analysis, instead of the mean and is a more reliable indicator of expected force to failure. Breslow’s algorithm was chosen as it is more robust than that originally used by Gehan. All the statistical methods used, confidence intervals, Weibull analysis and survival analysis produced consistent conclusions.


   Conclusions
Top
 Abstract
 Introduction
 Aims
 Material and methods
 Results
 Discussion
 Conclusions
References
 

  1. There was a significant lower force required to debond conventional brackets when compared with gingivally offset brackets.
  2. There was no significant relationship between the ARI scores and the bracket types.


   Acknowledgments
 
We would like to thank 3M Unitek for providing the materials used in this study, the Orthodontic laboratory at Dundee Dental Hospital for help in preparing the specimens and Mr Chic Gibson for production of the test jig.

Contributors

Bikram Singh Thind was responsible for study design, conducting the experiment, data interpretation, drafting of the article and final approval of the article. Colin J. Larmour assisted with protocol of the study and was responsible for preparation of the specimens and proofreading of the article before submission. David R. Stirrups was responsible for the data interpretation, statistical analysis and contributed to the writing of the article. Charles Lloyd was responsible for design of the apparatus and test procedures, supervising the production/commissioning of equipment and contributed to the writing of the article. Bikram Singh Thind and David R. Stirrups are the guarantors.


   References
Top
 Abstract
 Introduction
 Aims
 Material and methods
 Results
 Discussion
 Conclusions
 References
 
1 Ash S, Hay N. Adhesive pre-coated brackets, a comparative clinical study. Br J Orthod 1996; 23: 325–9.[Abstract]

2 Millett DT, Hallgren A, Cattanach D, et al. A 5-year clinical review of bond failure with light-cured resin adhesive. Angle Orthod 1998; 68: 351–6.[Medline]

3 O’Brien KD, Read MJF, Sandison RJ, Roberts, CT. A visible light-activated direct-bonding material: an in vivo comparative study. Am J Orthod Dentofac Orthop 1989; 95: 348–51.[CrossRef][Medline]

4 Zachrisson BU, Brobakken BO. Clinical comparison of direct versus indirect bonding with different bracket types and adhesives. Am J Orthod 1978; 74: 62–78.[CrossRef][Medline]

5 Smith NR, Reynolds IR. A comparison of three bracket bases: an in vitro study. Br J Orthod 1991; 18: 29–35.[Abstract]

6 Kinch AP, Taylor H, Warltier R, Oliver RG, Newcombe RG. A clinical trial comparing the failure rates of directly bonded brackets using etch times of 15 or 60 seconds. Am J Orthod Dentofac Orthop 1988; 94: 476–83.[CrossRef][Medline]

7 Brannstrom M, Nordenvall KJ, Malmgren O. The effect of various pretreatment methods of the enamel in bonding procedures. Am J Orthod 1978; 74: 522–30.[CrossRef][Medline]

8 Larmour CJ. An audit of orthodontic appliance breakages. RCS Eng Orthod Audit Newsletter 2000; 13: 3.

9 Fox NA, McCabe JF, Buckley JG. A critique of bond strength force testing in orthodontics. Br J Orthod 1994; 21: 33–43.[Abstract]

10 Artun J, Bergland S. Clinical trials with crystal growth conditioning as an alternative to acid-etch enamel pretreatment. Am J Orthod 1984; 85: 333–40.[CrossRef][Medline]

11 Landis JR, Koch GG. The measurement of observer agreement for categorial data, Biometrics 1977; 33: 159–74.[CrossRef][Medline]

12 Shapiro SS, Wilk MN. An analysis of variance test for normality. Biometrika 1965; 52: 606–9.

13 Tiku ML, A new statistic for testing suspected outliers, Commun Stat Theory 1975; 3: 485–93.

14 Shapiro SS. How to Test Normality and Other Distributional Assumptions. Milwaukee: American Society for Quality Control, 1980.

15 Gehan EA. Generalized Wilcoxon test for comparing arbitrarily single-censored samples. Biometrika 1965; 52: 203–23.[Abstract/Free Full Text]

16 Breslow N. A Generalized Kruskal–Wallis test for comparing K samples subject to unequal patterns of censorship. Biometrika 1970; 57: 579–94.[Abstract/Free Full Text]

17 Dickinson PT, Powers JM. Evaluation of fourteen direct bonding bases. Am J Orthod 1980; 78: 630–9.[CrossRef][Medline]

18 Reynolds IR. A review of direct orthodontic bonding. Br J Orthod 1975; 2: 171–8.

19 Larmour CJ, Stirrups DR. An ex vivo assessment of a resin-modified glass ionomer cement in relation to bonding technique. J Orthod 2001; 28: 207–10.[Abstract/Free Full Text]

20 Cardoso PE, Sadek FT, Goracci C, Ferrari M. Adhesion testing with the microtensile method: effects of dental substrate and adhesive system on bond strength measurements. J Adhes Dent 2002; 4: 291–7.[Medline]

21 ISO TS 11405. Dental Materials — Testing Methods of Adhesion to Tooth Structure. ISO Geneva: 2003.

22 Bishara SE, Olsen ME, Damon P, Jakobsen JR. Evaluation of a new light-cured orthodontic bonding adhesive. Am J Orthod Dentofac Orthop 1998; 114: 80–7.[CrossRef][Medline]

23 Bowen RL, Rodriguez MS. Tensile strength and modulus of elasticity of tooth structure and several restorative materials. J Am Dent Assoc 1962; 64: 378–87.

24 Katona TR. A comparison of the stresses developed in tension, shear peel, and torsion strength testing of direct bonded orthodontic brackets. Am J Orthod Dentofac Orthop 1997; 112: 44–251.





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thind, B. S.
Right arrow Articles by Lloyd, C. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thind, B. S.
Right arrow Articles by Lloyd, C. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS