1 of(2013)

European Journal of Orthodontics 35 8 236–243 © The Author 2011.

2011. Published
Published by Oxford University Press on behalf of the European Orthodontic Society.
doi:10.1093/ejo/cjr126 All rights reserved. For permissions, please email: [email protected]
Advance Access publication 2 November 2011

Roth versus MBT: does bracket prescription have an effect on

the subjective outcome of pre-adjusted edgewise treatment?
Bopelo Moesi*, Fiona Dyer** and Philip E. Benson*
*Academic Unit of Oral Health and Development, School of Clinical Dentistry, University of Sheffield and
**Department of Orthodontics, Charles Clifford Dental Hospital, Sheffield, UK

Correspondence to: Philip Benson, Reader in Orthodontics, Academic Unit of Oral Health and Development, School
of Clinical Dentistry, Sheffield S10 2TA, UK. E-mail: p.benson@sheffield.ac.uk

SUMMARY The aim was to determine if bracket prescription has any effect on the subjective outcome of
pre-adjusted edgewise treatment as judged by professionals. This retrospective observational assessment
study was undertaken in the Orthodontic Department of the Charles Clifford Dental Hospital, Sheffield,
UK. Forty sets of post-treatment study models from patients treated using a pre-adjusted edgewise
appliance (20 Roth and 20 MBT) were selected. The models were masked and shown in a random order
to nine experienced orthodontic clinicians, who were asked to assess the quality of the outcome, using a
pre-piloted questionnaire. The principal outcome measure was the Incisor and Canine Aesthetic Torque
and Tip (ICATT) score for each of the 40 post-treatment models carried out by the nine judges. A two-way
analysis of variance was undertaken with the dependent variable, total ICATT score and independent
variables, Bracket prescription (Roth or MBT) and Assessor. There were statistically significant differences
between the subjective assessments of the nine judges (P<0.001), but there was no statistically significant
difference between the two bracket prescriptions (P = 0.900). The best agreement between a clinician’s
judgment of prescription used and the actual prescription was fair (kappa statistic 0.25; CI −0.05 to 0.55).
The ability to determine which bracket prescription was used was no better than chance for the majority
of clinicians. Bracket prescription had no effect on the subjective aesthetic judgments of post-treatment
study models made by nine experienced orthodontists.

Introduction
3. Increased lingual crown torque in the lower incisor
Since the introduction of the Straight Wire Appliance™ in brackets (Andrews: −1 degrees, Roth: −1 degrees, and
the 1970s (Andrews, 1979), there have been many suggested MBT: −6 degrees)
modications to the tip and torque values used in pre- 4. Decreased tip in the upper canine brackets (Andrews:
adjusted edgewise appliances. Many of these changes 11 degrees, Roth: 13 degrees, and MBT: 8 degrees).
involve alterations of a few degrees, even though it is The developers of the appliance claim that the increased
known that torque expression in particular is affected by the palatal root torque in the upper incisors improves the under-
amount of play between the archwire and the slot torqued appearance produced by other prescriptions and the
(Archambault et al., 2010), differences in the tolerance size
increased labial root torque in the lower incisor counteracts
of manufactured brackets and archwires (Cash et al., 2004),
the forward tipping during levelling (McLaughlin et al.,
the method of ligation (Gioka and Eliades, 2004; Badawi
2001). To date, there have been no scientic studies to
et al., 2008), the initial inclination of the teeth (Archambault
support these claims.
et al., 2010), additional widening and notching of the
Several studies have shown the variations in torque values
bracket slot when placing the larger archwires (Archambault
of teeth achieved following treatment with pre-adjusted
et al., 2010), and even variations in the shape of the labial
edgewise appliances (Dellinger, 1978; Vardimon and
surface of teeth (Smith et al., 2007).
Lambertz, 1986; Ugur and Yukay, 1997). Kattner and
The MBT prescription was introduced in 1997 and
Schneider (1993) found no differences in the ideal tooth
quickly established itself as one of the most popular bracket
relationship index when they compared the study models of
prescriptions on the market. The main differences with
patients treated using a Roth prescription pre-adjusted
other bracket prescriptions are
edgewise appliance with those treated using a standard
1. Increased palatal root torque in the upper central incisor edgewise appliances. Ugur and Yukay (1997) found no
brackets (Andrews: 7 degrees, Roth: 12 degrees, and differences in the objectively measured torque values
MBT: 17 degrees) between cases treated using standard edgewise and a pre-
2. Increased palatal root torque in the upper lateral incisor adjusted Roth prescription appliance.
brackets (Andrews: 3 degrees, Roth: 8 degrees, and The aim of this study was to investigate the bracket
MBT: 10 degrees) manufacturers’ claims that when a patient is treated to a

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 2 of 8VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES
ROTH B. MOESI ET237
AL.

reasonable occlusal outcome, then small changes in bracket 3. A Peer Assessment Rating (PAR) score of 5 or less from
prescription can lead to visually detectable differences in the post-treatment study models;
tooth positions. 4. Placement of a 0.019 × 0.025 inch stainless steel working
archwire for at least one visit; and
The specic research questions 5. An ANB angle not less than 1 degrees and not more than
5 degrees.
From a sample of study models taken from patients with
Patient records were excluded if they were treated
skeletal 1 malocclusions treated with either upper or upper
with:
and lower premolar extractions to a good occlusal result:
1. A non-extraction approach;
1. Is there a difference in the subjective aesthetic judgements
2. Extractions other than premolars;
of orthodontists in the appearances of maxillary and
3. A functional appliance;
mandibular incisor torque or maxillary canine torque
4. Headgear; and
and tip between the Roth and MBT prescription?
5. Orthognathic surgery.
2. Are orthodontists able to distinguish if a patient was
treated with the Roth or MBT prescription? The pre-treatment records were examined by two
experienced specialist orthodontists to determine the incisor
The null hypotheses were that there are no differences in
relationship, the size of the overjet, and the degree of upper
the subjective aesthetic judgements of orthodontists as to
and lower arch crowding. Any disagreements were resolved
the appearance of torque of the maxillary and mandibular
by a third experienced orthodontist. The examiners were
incisors and the tip of the maxillary canines between cases
unaware of which bracket prescription had been used. Data
treated using a pre-adjusted edgewise appliance with a Roth
about the demographics of the patient and length of
or MBT prescription.
treatment were obtained from the clinical records.
The 40 sets of models were duplicated and cast in the
Subjects and methods same yellow stone by one investigator to ensure uniformity
of appearance. They were then allocated a computer-
The sample consisted of the post-treatment study models of
generated random number from 1 to 40.
40 patients treated in the Orthodontic Department of the
A questionnaire to capture a clinician’s subjective
Charles Clifford Dental Hospital, Shefeld, UK. This was a
assessments of the incisor torque and canine tip of each
convenience sample, chosen retrospectively, to be
model was developed through discussions with experienced
representative of a common type of orthodontic patient,
orthodontic clinicians. It was piloted by two senior specialist
treated to a good occlusal result, in a UK postgraduate
registrars and modied. The nal questionnaire (See online
teaching hospital. Twenty patients had received a pre-
supplementary material for the Appendix 1) consisted of
adjusted edgewise appliance with the Roth prescription
two questions concerning upper and lower incisor torque,
(Ovation; DENTSPLY GAC, Bohemia, New York, USA)
two questions about the torque in the right and left upper
and 20 patients had received the MBT prescription (Victory;
canine, and two questions about the right and left upper
3M, St Paul, Minnesota, USA). Conrmation of the bracket
canine tip. A seventh question asked the respondent to state
prescription used was obtained from the hospital notes, the
whether they thought that the case had been treated using an
departmental database, and by examination of clinical
MBT or a Roth prescription. Photographs showing distal,
photographs taken during treatment. The patients were
upright, and correct maxillary canine tip congurations
treated by several operators, but archwires were standardized
were provided to each clinician as an aid in the determination
within the department (Sentalloy nickel-titanium aligning
(See online supplementary material for the Appendix 1).
archwires; DENTSPLY GAC, and 0.019 × 0.025 inch ss
The responses for the rst six questions were on a 5-point
working archwires; DB Orthodontics, Silsden, West Yorks,
UK). Likert scale (See online supplementary material for the
Appendix 1). The response for question 7 was a dichotomous
An a priori sample size estimation could not be performed
(MBT or Roth).
as there were no data upon which to base the calculation;
The questionnaire was administered to nine orthodontic
however, a post hoc power analysis was undertaken once
clinicians (four consultants, two senior postgraduate
data had been collected, to determine what a suitable sample
size to detect a signicant difference might be, based on the trainees, and three other specialists). Each assessor was
masked as to the identity of the original patient, the
results of this study.
prescription used, and the number of models of each
The following inclusion criteria were applied for the
selection of the patient records: prescription.
To test reproducibility, the models were renumbered
1. Aged 20 years or under; from 1 to 40 in a new random order and three assessors
2. Two premolar extractions in the upper arch or four upper reassessed the whole sample at least 3 weeks after the initial
and lower premolar extractions; assessment.

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 ROTH
238 VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES B. MOESI3ETof 8
AL.

Statistical analysis The strength of the agreement was determined using the
criteria suggested by Landis and Koch (1977).
Data were entered into an Excel® spreadsheet (Microsoft
2007) and PASW Statistics (SPSS Inc v 18) was used to
undertake the statistical tests. Agreement between examiners Results
for the pre-treatment characteristics was determined using
an unweighted kappa statistic. The kappa scores for agreement between examiners for the
Differences in the aesthetic outcomes for the two bracket pre-treatment occlusal characteristics were either substantial
prescriptions were examined using a total Incisor and [incisor relationship 0.80, 95 per cent condence interval
Canine Aesthetic Torque and Tip (ICATT) score. The (CI) 0.65–0.95; lower arch crowding 0.78, 95 per cent
responses to questions 1–6 were given a score ranging CI 0.60–0.96] or almost perfect (upper arch crowding 0.81,
from 0 (all four teeth inadequately torqued, severely 95 per cent CI 0.65–0.97). Table 1 shows the pre-treatment
undertorqued, and signicant distal tip) to 4 (all four teeth patient and occlusal characteristics, length of time, and
adequately torqued, best possible torque, and correct tip). number of appointments in active orthodontic treatment for
The scores for the six questions were summed to produce a the two bracket prescription groups. There was a slightly
total score for each model (minimum score 0 and maximum higher number of patients with a Class I incisor relationship
score 24). in the Roth prescription group and a slightly higher number
The reproducibility of the repeat ICATT scores was of patients with moderate lower arch crowding in the MBT
assessed using an intra-correlation coefcient for random group, but otherwise, the pre-treatment patient and occlusal
error and a paired t-test for systematic error. The agreement and treatment characteristics were very similar.
between the rst and second assessments of whether the The 40 post-treatment study models were assessed by six
assessor considered the case to have been treated using specialist orthodontists on one occasion and by three
MBT or Roth (question 7) was analysed using an unweighted specialist orthodontists on two occasions making a total of
kappa statistic. 480 separate assessments. The results of the reproducibility
The null hypothesis was that there was no difference assessment are shown in Table 2. Assessor 1 had the largest
in the scores for the two prescriptions; however, the mean difference between the two ICATT readings (−1.1),
assessor was also included in the analysis as an independent which was statistically signicant (P = 0.035) suggesting a
variable to take into account inter-examiner differences. systematic error. The random error showed moderate
The distribution of the data was examined and found to agreement. The other two assessors had lower mean
be normally distributed but truncated. This was because the differences in their repeat ICATT scores, with no systematic
peak of the distribution was towards the higher scores; error and substantial agreement for random error.
however, as it was impossible to achieve a score larger than
24, although the curve was diminishing, it did not return to Table 1 Pre-treatment patient and occlusal characteristics,
length of time, and number of appointments in active orthodontic
the horizontal axis. Several ways of transforming the data
treatment. SD, standard deviation.
were attempted, but no suitable method was found and we
were unable to determine a non-parametric equivalent of
the two-way analysis of variance (ANOVA), which allowed MBT (n = 20) Roth (n = 20)

an analysis of two independent variables (assessor and

Incisor relationship
bracket). Following statistical advice, it was decided that  Class I 3 8
since the data were truncated–normal, then a two-way  Class II division 1 8 7
ANOVA was acceptable; however, to conrm this, a non-  Class II division 2 4 3
 Class II intermediate 2 0
parametric Mann–Whitney U-test was conducted to analyse  Class III 3 2
the difference in ICATT scores for each of the assessors to Overjet (mm)
examine if this agreed with the results of the overall  Mean (SD) 4.0 (1.7) 3.7 (2.3)
Upper arch alignment (mm)
ANOVA. The dependent variable for the two-way ANOVA  No or mild crowding (0–4) 9 10
was the total ICATT score and the independent variables  Moderate crowding (5–8) 7 6
were bracket prescription (Roth or MBT) and assessor.  Severe crowding (>9) 4 4
Lower arch alignment (mm)
The mean values of all nine assessors’ judgements for the  No or mild crowding (0–4) 10 15
six characteristics and the total ICATT scores were calculated  Moderate crowding (5–8) 9 5
and a Mann–Whitney U-test carried out to determine any  Severe crowding (>9) 1 0
Age at start of treatment (years)
differences. As multiple comparisons were undertaken, the  Mean (SD) 15.1 (1.8) 14.4 (2.7)
signicance level was set at P < 0.01. Length of treatment (months)
For question 7, the agreement between the assessor ’s  Mean (SD) 24.4 (8.1) 23.3 (6.3)
Nos of appts in appliances
judgement about which prescription was used and the actual  Mean (SD) 17.1 (3.7) 16.0 (3.8)
prescription used was assessed using the kappa statistic.

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 4 of 8VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES
ROTH B. MOESI ET239
AL.

The levels of agreement for the three assessors repeat individual judges, none of which showed a signicant
assessments of whether the case was treated with Roth or difference between the two brackets systems (P values
MBT prescription are shown in Table 3. Assessor 2 showed ranged from 0.120 for assessor 5 to 0.978 for assessor 1).
the best agreement between the rst and second viewings, Since the overall ANOVA showed no effect for bracket and
but this was only moderate (kappa = 0.52) and the 95 per this agrees with the nine individual assessors’ Mann–Whitney
cent CIs were wide (lower limit kappa = 0.20; slight U-tests, then it can be concluded that bracket prescription
agreement to upper limit kappa = 0.84; almost perfect had no signicant effect on the aesthetic scores.
agreement). Table 5 shows the descriptive data for the mean aesthetic
Table 4 shows the results of the two-way ANOVA. There scores provided by the nine assessors for the 20 cases
were signicant differences between the individual assessors’ treated using MBT prescription and 20 cases treated using
scores (P < 0.001); however, the scores for the two bracket Roth prescription. There were no statistically signicant
systems were not signicantly different (P = 0.900). This differences between the judgements made for the two
was conrmed by the Mann–Whitney U-tests for the prescriptions for any of the six attributes or total ICATT
scores.
Examination of the individual scores showed that
Table 2 Reproducibility of three judges repeat incisor and canine
aesthetic torque and tip (ICATT) scores including mean difference,
generally, the majority of the assessors agreed on the
standard deviation (SD) of the differences, 95 per cent condence outcomes of good and bad cases; however, there were wide
intervals (CIs), P-value of paired t-test for systematic error, and discrepancies in the subjective assessments of some cases.
intra-class correlation coefcient for random error. For example, case 4 received a total ICATT score of 2 from
assessor 9 and a score of 20 from assessor 7. For other
Judge Mean SD of 95% CI P-value ICC cases, there was reasonable consensus, e.g. case 38 had a
difference differences minimum score of 18 from assessor 4, a maximum score of
Lower Upper 24 from assessor 6, and six assessors gave it the same total
ICATT score of 22. To determine if there were any
1 −1.1 3.1 −2.0 −0.1 0.035 0.53 differences between the cases, they were arranged according
2 0.4 2.5 −0.4 1.2 0.314 0.71 to their average total score per case (Figure 1). No case
7 −0.4 3.3 −1.4 0.7 0.470 0.63
obtained a mean maximum score (24). The best case was
treated with the Roth prescription and the worst case with
Table 3 Agreement of three judges repeat assessments of the MBT prescription. However, from the best 25 per cent
whether the case was treated with Roth or MBT prescription of cases (top 10 cases), 6 were treated with MBT and 4 with
including kappa statistic, 95 per cent condence intervals (CIs), Roth. From the worst 25 per cent of cases (bottom 10 cases),
and strength of agreement according to the criteria of Landis and 5 were treated with each prescription. Most of the cases
Koch (1977). (80 per cent) had scores between 14 and 20. To determine if
the poorly ranked cases skewed the data, the comparison
Judge Kappa statistic 95% CI Strength of agreement was repeated with the 10 poorly ranked cases of each
prescription excluded; however, there were still no
Lower Upper statistically signicant differences between the two
prescriptions for any of the variables (total ICATT Mann–
1 0.04 −0.27 0.34 Slight Whitney U-test P = 0.845).
2 0.52 0.20 0.84 Moderate
7 0.34 0.03 0.65 Fair The assessors were asked to predict whether the cases
were treated with MBT or Roth prescriptions. Table 6 shows

Table 4 Results of two-way analysis of variance with dependent variable total incisor and canine aesthetic torque and tip (ICATT) score
and two independent variables of bracket and assessor.

Source Type III sum of squares df Mean square F P

Corrected model 1181.250* 9 131.3 7.4 <0.001

Intercept 90123.4 1 90123.4 5114.5 <0.001
Bracket 0.3 1 0.3 0.0 0.900
Assessor 1181.0 8 147.6 8.4 <0.001
Error 6167.4 350 17.6
Total 97472.0 360
Corrected total 7348.6 359

*R2 = 0.161 (adjusted R2 = 0.139).

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 ROTH
240 VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES B. MOESI5ETof 8
AL.

Table 5 Descriptive statistics for the six attributes and total incisor and canine aesthetic torque and tip (ICATT) scores for the 20 cases
treated using Roth prescription and 20 cases treated using MBT prescription including means, standard deviations (SDs), 95 per cent
condence intervals (CIs), ranges, and P-values (Mann–Whitney U-test).

Attribute Prescription Mean score SD 95% CI Minimum Maximum P-value

Lower Upper

Upper incisor torque (scores 0–4) Roth 2.6 1.2 2.0 3.1 1 4 0.828
MBT 2.7 1.3 2.0 3.3 0 4
Lower incisor torque (scores 0–4) Roth 3.2 1.3 2.5 3.8 0 4 0.773
MBT 3.4 0.8 3.0 3.8 1 4
Upper right canine torque (scores 0–4) Roth 2.3 0.6 2.0 2.5 1 3 0.587
MBT 2.2 0.5 2.0 2.4 1 3
Upper left canine torque (scores 0–4) Roth 2.4 0.6 2.1 2.6 1 3 0.430
MBT 2.4 0.7 2.1 2.7 1 3
Upper right canine tip (scores 0–4) Roth 2.7 1.1 2.2 3.2 0 4 0.725
MBT 2.6 1.0 2.2 3.1 0 4
Upper left canine tip (scores 0–4) Roth 2.9 1.2 2.3 3.4 0 4 0.490
MBT 2.5 1.0 2.1 2.9 1 4
Total ICATT score (scores 0–24) Roth 15.9 3.0 14.5 17.2 8 20 0.957
MBT 15.8 3.5 14.3 17.3 9 22

Figure 1 Graph showing the frequency of mean total incisor and canine aesthetic torque and tip
(ICATT) scores (minimum 0 and maximum 24) from the nine judges for the cases treated with either
the Roth (n = 20) or the MBT prescriptions (n = 20).

the agreement between the assessments of the nine assessors the patient was in the 0.019 × 0.025 inch stainless steel
about which prescription they thought had been used and archwire, as determined from the clinical record, against the
the actual prescription used. The best kappa statistic mean total ICATT scores from the nine examiners for all
achieved for assessor validity was a fair agreement (0.25); 40 cases. The Pearson’s product correlation coefcient was
however, the CI ranged from poor agreement (−0.05) to also calculated. There was no obvious visual relationship
moderate agreement (0.55). Six assessors achieved slight between the two from the scatter plot (Figure 2) and the
agreement, whereas two assessors achieved poor agreement. correlation coefcient was both weak (r = 0.106) and
The agreement between assessor judgement and the actual non-signicant.
prescription used for treating each case was not statistically
signicant for any of the assessors.
Discussion
To determine if the length of time the patient had a
rectangular stainless steel archwire in place affected the This study found no differences in the subjective assessments
subjective outcome, we examined a scatter plot of the time carried out by experienced orthodontic clinicians of the

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 6 of 8VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES
ROTH B. MOESI ET241
AL.

post-treatment study models from patients who had been The study models were a convenience sample chosen
treated with premolar extractions and pre-adjusted edgewise retrospectively. This could lead to potential bias; however,
xed appliances using a Roth prescription compared with specic inclusion criteria were used to produce representative
an MBT prescription. In fact, the clinicians’ determination samples of skeletal 1 orthodontic cases treated with upper or
of which prescription had been used was, in most individuals, upper and lower premolar extractions to a good standard in
no better than chance. It therefore appears that for treating a postgraduate teaching clinic. A recognized quality
skeletal Class I cases, with at least two premolars extracted, outcome (post-treatment PAR score of 5 or less) was one
it does not matter whether one treats a case with the MBT inclusion criterion as the objective was not to determine
or Roth prescription as the subjective outcome is the same. whether the MBT or Roth prescriptions were more effective
This study did not examine the issue of stability, which can at producing a good occlusal result; a prospective
only be determined with long-term follow up of patients. randomized design would be an appropriate study design to
the answer that question, but to assess if differences between
Table 6 The individual agreements for the nine judges as to the prescription could be detected by experienced clinicians
whether the case was treated with Roth or MBT prescription with once a good occlusal result had been achieved. The use of
the actual bracket prescriptions used including kappa statistic, 95
per cent condence intervals (CIs), and strength of agreement these relatively strict criteria, particularly the outcome
according to the criteria of Landis and Koch (1977). measure, would make it very difcult to obtain a sample
of consecutively started patients as many cases would
Judge Kappa statistic 95% CI Strength of agreement
be excluded. The details of the patient demographics,
malocclusion, and treatment characteristics show that the
Lower Upper two patient samples were very similar and all the models
were masked to avoid assessment bias.
1 −0.05 −0.37 0.27 Poor The method of assessing the outcome of treatment used
2 0.20 −0.07 0.47 Slight in this study was developed to be clinically relevant, quick,
3 0 −0.31 0.31 Slight and easily applied to the teeth that are potentially most
4 0.25 −0.05 0.55 Fair
5 0.10 −0.19 0.39 Slight affected by the changes in the prescription values.
6 0 −0.32 0.32 Slight Investigators who have used more objective methods of
7 −0.05 −0.36 0.26 Poor measurement have also been unable to nd signicant
8 0.10 −0.21 0.41 Slight
9 0.05 −0.26 0.36 Slight differences between appliances (Ugur and Yukay, 1997). It
might be true that more contemporary objective and precise

Figure 2 Scatter plot of time in 0.019 × 0.025 inch stainless steel archwire against mean total incisor and canine
aesthetic torque and tip (ICATT) scores from the nine judges.

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 ROTH
242 VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES B. MOESI7ETof 8
AL.

methods of measurements, such as 3D laser scanning, may were from the same supplier; but the brackets were made
be able to detect differences between appliances; by different manufacturers and one may have been
nevertheless, we would argue that for a specialty, which is manufactured to a better tolerance than the other. However,
aiming to produce the best aesthetic result, a difference that if small changes to bracket prescription make a difference
is not detectable by the human eye is of little importance. to tooth position, then one might expect that the bracket
Kattner and Schneider (1993) examined the post- with the lower tolerance would produce a better aesthetic
treatment study models of 120 patients treated using a outcome, which experienced clinicians would be able to
standard edgewise appliance and pre-adjusted edgewise detect, but they could not. Other reasons for the nding that
Roth prescription appliance by two specialist orthodontists. bracket prescription made no difference to the subjective
The investigators did not nd any signicant differences in assessments of treatment outcome might include the
the outcomes between appliance systems; however, they possibility that clinicians manipulated the working archwire
did nd differences between the two clinicians. The clinician to introduce more torque into the cases treated with the
judged to have better occlusal outcomes routinely took Roth prescription, but very few clinicians stated in the
longer to nish cases and more often used a full-sized patient record that this had been carried out. Another reason
archwire than the practitioner with the lower scores. We for the lack of difference might be due to inaccuracies in
found no difference in the length of active treatment bracket placement, leading to inaccuracies in tip and torque
between patients treated with the MBT and Roth expression. Some assessors made comments to this effect
prescriptions and we also examined our sample to see if the about some of the cases.
length of time in the largest archwire used in our cases It is possible that tip is more fully expressed in a pre-
(0.019 × 0.025 inch stainless steel) inuenced the outcome adjusted edgewise appliance than torque. There is a
but were unable to nd any relationship. The average length 5 degrees difference in canine tip between the MBT and Roth
of time the patients in our sample were in a 0.019 × 0.025 prescriptions and the expectation was that the patients treated
inch ss archwire was 11 months. The case that scored the using the MBT prescription would have more upright upper
highest mean total ICATT score was in this archwire for canine teeth compared with the more mesially tipped canines
13 months; however, 10 from the best 20 cases used the of patients treated using the Roth prescription. The results
0.019 × 0.025 inch ss archwire for less than 11 months. of this study suggest that the two appliance prescriptions
It is possible that it is not the length of time in the largest could not be differentiated on the basis of canine tip.
archwire used, but the size of archwire that affects the The questionnaire was piloted before starting the
outcome. Finishing cases in wires that do not substantially investigation to test the relevance and ease of use; however,
ll the bracket slot will not fully express the torque values. the design of study could be criticized for a number of
The cases in this study were nished with 0.019 × 0.025 reasons. Firstly, we used experienced clinicians to make a
inch ss archwires in a 0.022 × 0.028 inch bracket slots. Ugur judgement from study models, whereas it might be more
and Yukay (1997) examined cases nished with 0.016 × appropriate to ask lay people, patients, or parents to assess
0.022 inch ss archwires in a 0.018 × 0.030 inch bracket the smile aesthetics. Studies have suggested that lay people
slots. The loss of torque of a 0.016 × 0.022 inch archwire in assess the smile differently to clinicians (Flores-Mir et al.,
a 0.018 inch bracket slot is approximately 14 degrees, 2004); however, orthodontists tend to be more critical than
which is similar to that of a 0.019 × 0.025 inch archwire in lay people or even general dentists (Kokich et al., 2006), so
a 0.022 inch bracket slot (Sebanc et al., 1984; Gioka and using non-specialists would potentially make it even less
Eliades, 2004; Badawi et al., 2008). Increasing the archwire likely that a signicant difference could be found.
size to dimensions 0.017 × 0.025 inch in a 0.018 inch Another criticism of this investigation is that no
bracket slot decreases the play to 6 degrees, but this is the calculation to determine a suitable sample size needed to
same magnitude as the difference between the torque values detect a clinically signicant difference if one truly exists
of the Roth and MBT prescriptions. Kattner and Schneider was performed before carrying out the study. This was not
(1993) did not nd any differences between cases nished undertaken because there were no data upon which to base
using 0.017 × 0.025 inch ss archwires compared with cases the calculation. It is, however, possible to use the actual
nished using 0.016 × 0.022 inch ss archwires. It is possible data from the study to determine how many patients would
that if the patients were treated with full-sized 0.021 × 0.028 be required to show a signicant difference. The largest
inch ss archwires, which have a greater potential for difference in the subjective judgments between the two
full torque expression, then differences between bracket bracket systems was for the torque in the upper right canine.
prescriptions might be detected; however, the routine use of The mean difference between the scores for the patients
full size archwires is not advocated by those who developed treated using Roth prescription and those treated using the
the MBT appliance (McLaughlin et al., 2001). MBT prescription was 0.07 (SD 0.60). This gives a
It has been shown that there are differences in the standardized difference of 0.12 (Altman, 1991). Using the
tolerance size of manufactured brackets and archwires nomogram for continuous data (two independent groups)
(Cash et al., 2004). The archwires used in the patient sample and assuming that this was a representative sample, we

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017
 8 of 8VERSUS MBT PRESCRIPTIONS: SUBJECT OUTCOMES
ROTH B. MOESI ET243
AL.

would require approximately 3000 patients to detect a Supplementary material

signicant difference between the two bracket systems for
Supplementary material is available at European Journal of
this assessment alone and probably many more for the other
Orthodontics online.
assessments. Even if we were able to use an objective
measurement of tooth angulations, instead of a subjective
assessment, previous studies suggest that the variation Acknowledgements
in the measurements (Dellinger, 1978; Vardimon and
The authors would like to thank the nine orthodontists
Lambertz, 1986) would be such that sample sizes into the
who took part in this study, as well as Dr Kathleen Baster of
thousands would be required in order to nd signicant
the Statistical Services Unit, University of Shefeld for
differences and then these might not be detectable to the
help with the statistical analysis and Dr Nicola Parkin for
human eye.
resolving disagreements between PEB and FD on the pre-
A further potential criticism of the study is that the
treatment characteristics.
patients were treated by different clinicians, albeit from one
centre. Operator variability might therefore have masked
any differences in bracket prescription. We would argue that References
an original objective of the pre-adjusted edgewise appliance Altman D G 1991 Practical statistics for medical research. Chapman &
was to reduce the amount of wire bending required when Hall, London
treating patients and therefore, promote more consistent Andrews L F 1979 The straight-wire appliance. British Journal of
treatment outcomes both within- and between-individual Orthodontics 6: 125–143
operators. Using the study models from patients treated by Archambault A, Lacoursiere R, Badawi H, Major P W, Carey J, Flores-Mir C
2010 Torque expression in stainless steel orthodontic brackets. a
different operators is more representative of what happens systematic review. Angle Orthodontist 80: 201–210
in the real world, but examining study models from single Badawi H M, Toogood R W, Carey J P, Heo G, Major P W 2008 Torque
operators might be the basis for further investigations. expression of self-ligating brackets. American Journal of Orthodontics
Another potential weakness is that each assessor was asked and Dentofacial Orthopedics 133: 721–728
to carry out 280 judgements and that fatigue might have Cash A C, Good S A, Curtis R V, McDonald F 2004 An evaluation of slot
size in orthodontic brackets—are standards as expected? Angle
affected the results. No time restraints were placed on the Orthodontist 74: 450–453
assessor to carry out the work and they were advised to Dellinger E L 1978 A scientic assessment of the straight-wire appliance.
take regular breaks throughout the assessment period. American Journal of Orthodontics 73: 290–299
As previously stated, we used reasonably strict and Flores-Mir C, Silva E, Barriga M I, Lagravere M O, Major P W 2004 Lay
limited inclusion and exclusion criteria in the selection person’s perception of smile aesthetics in dental and facial views.
Journal of Orthodontics 31: 204–209
of models for this investigation. Further work involving
Gioka C, Eliades T 2004 Materials-induced variation in the torque
patients with skeletal 2 and skeletal 3 malocclusions as expression of preadjusted appliances. American Journal of Orthodontics
well as those treated non-extraction is required to and Dentofacial Orthopedics 125: 323–328
confirm that small changes in pre-adjusted edgewise Kattner P F, Schneider B J 1993 Comparison of Roth appliance and
bracket prescriptions has no effect on the aesthetic standard edgewise appliance treatment results. American Journal of
Orthodontics and Dentofacial Orthopedics 103: 24–32
judgements of the nal results of orthodontic treatment.
Kokich V O, Kokich V G, Kiyak H A 2006 Perceptions of dental
When one considers the many factors that affect torque professionals and laypersons to altered dental esthetics: asymmetric and
expression, not least of which is the amount of variation symmetric situations. American Journal of Orthodontics and Dentofacial
in the torsional or engagement angle between the bracket Orthopedics 130: 141–151
slot and the wire (Archambault et al., 2010), then it is Landis J R, Koch G G 1977 The measurement of observer agreement for
categorical data. Biometrics 33: 159–174
perhaps not surprising that small changes in pre-adjusted
McLaughlin R P, Bennett J C, Trevisi H 2001 Systemized orthodontic
edgewise bracket prescriptions fail to make any clinically treatment mechanics. Mosby, London
detectable differences to the subjective appearance of Sebanc J, Brantley W A, Pincsak J J, Conover J P 1984 Variability of
the final result. effective root torque as a function of edge bevel on orthodontic arch
In conclusion, this study has found that bracket wires. American Journal of Orthodontics 86: 43–51
prescription had no effect on the subjective aesthetic Smith R N, Karmo M, Russell J, Brook A H 2007 The variability of the
curvature of the labial surface of the upper anterior teeth along the facial
judgements made by nine experienced orthodontists from axis of the clinical crown. Archives of Oral Biology 52: 1037–1042
the post-treatment study models of patients treated with Ugur T, Yukay F 1997 Normal faciolingual inclinations of tooth crowns
premolar extractions and a pre-adjusted edgewise xed compared with treatment groups of standard and pretorqued brackets.
appliance system using either a Roth or a MBT prescription. American Journal of Orthodontics and Dentofacial Orthopedics 112:
50–57
In the majority of cases, the ability of the clinicians to
Vardimon A D, Lambertz W 1986 Statistical evaluation of torque angles in
determine which bracket prescription was used was no reference to straight-wire appliance (SWA) theories. American Journal
better than chance in the majority of cases. of Orthodontics 89: 56–66

Downloaded from https://academic.oup.com/ejo/article-abstract/35/2/236/489796

by guest
on 11 November 2017