Introduction
The objective of this 4-arm parallel study was to evaluate the alignment efficiency and esthetic performance of 4 coated nickel-titanium archwires over an 8-week period.
Methods
Patients in the permanent dentition requiring maxillary and mandibular fixed orthodontic treatment with a preadjusted edgewise appliance were eligible for inclusion. Patients attending 4 hospital departments (United Kingdom and Italy) were randomly allocated to 1 of 4 treatment interventions: (1) BioCosmetic (Forestadent, Pforzheim, Germany), 0.017 in; (2) Titanol (Forestadent), 0.016 in; (3) TP Aesthetic (TP Orthodontics, La Porte, Ind), 0.014 in; and (4) Tooth Tone (Ortho Organizers, Calsbad, Calif) 0.016 in. Block randomization with block sizes of 4 and 8 was used to ensure an allocation ratio of 1:1:1:1. The primary outcome was alignment efficiency determined by the reduction in Little’s irregularity index (mm). Secondary outcomes were color change using the Commission Internationale de L’Eclairage L*a*b* system and percentage of coating loss. Blinding was only applicable to outcome assessment of alignment efficiency. Regression models with Sidak’s multiple comparison of means were used to analyze the data.
Results
One hundred fifty patients (300 dental arches) were allocated to the treatment interventions, including 61 male and 89 female subjects with a mean age of 16.60 years. The average duration of follow-up was 63.65 days. Baseline characteristics for the archwire groups were similar. One patient was lost to follow-up. Five percent (n = 15) of the archwires fractured: BioCosmetic, 5.3% (n = 4); Titanol, 6.8% (n = 5); TP Aesthetic, 5.3% (n = 4); and Tooth Tone, 2.7% (n = 2). We analyzed 283 dental arches for alignment efficiency. There was no statistically significant difference for mean reduction in irregularity between the archwire groups ( P = 0.627): BioCosmetic (n = 71), 3.86 mm (95% CI, 3.31-4.41); Titanol (n = 69), 4.51 mm (95% CI, 4.00-5.02); TP Aesthetic (n = 71), 4.13 mm (95% CI, 3.49-4.78); and Tooth Tone (n = 72), 4.21 mm (95% CI, 3.89-4.46). There was a statistically significant difference between archwire groups for color change ( P = 0.001) and percentage of coating loss ( P = 0.001), with BioCosmetic performing best in both parameters.
Conclusions
There was no difference between the archwires for alignment efficiency. BioCosmetic performed statistically significantly better than did the other groups for both color change and coating loss.
Registration
This trial was registered with the East Midlands NHS Research Ethics Committee (12/EM/0190).
Protocol
The protocol was not published before trial commencement.
Highlights
- •
There was no difference in efficiency of alignment for the 4 groups of esthetic archwires.
- •
There were significant color changes and coating losses for all archwires after 8 weeks.
- •
BioCosmetic performed significantly better for coating loss and color change than the other archwires.
Increased dental and orthodontic awareness has made fixed appliance orthodontic therapy more socially acceptable for patients, especially among adults. Even though more adult patients are demanding orthodontic treatment, a previous study highlighted that nearly a third of them would not be willing to undergo treatment with visible labial appliances. Young adults may be more self-conscious since less favorable judgments may be made by others based on dental appearance and orthodontic appliance type. These factors have led to the availability of a variety of esthetic appliances and auxiliary products; aligners and lingual appliances have gained in popularity. Further development in dental materials science has also seen advances in esthetic brackets and archwires. Although ceramic brackets are used widely in clinical practice, esthetic archwires may still be regarded as a “work in progress.”
Stainless steel and superelastic nickel-titanium (NiTi) archwires may be coated completely or just on their labial surface. These coatings are mainly either epoxy resin or polytetrafluoroethylene, which is of particular interest in orthodontics because of its low coefficient of friction, and it is an esthetic, stable, and antiadherent material. However, in clinical situations, delamination, ditching, cracking, and discoloration have been observed. The partial or complete loss of the surface coating can lead to detrimental mechanical and esthetic properties. In addition, exposure to the oral environment and food stains lead to color instability of the coating material.
Color measurements are conventionally carried out using spectrophotometers or colorimeters in industry. Using these instruments to measure the surface color of an archwire is problematic due to the relatively large and flat area required for a valid measurement. Calibrated images from single reflex lens digital cameras can be used in conjunction with the Commission Internationale de L’Eclairage L*a*b* (CIELAB) color space system to measure the surface color of dental materials. The CIELAB color space is 3 dimensional and allows color measurements to be plotted within the space; the Euclidean distance between the 2 points is the color difference.
The magnitude of perceptible and acceptable color differences is still not well defined in dental research. A study using nondental materials suggested that a color difference value of 1 would be detected 50% of the time, but this may not apply to dental research, since many of the colors assessed were outside the range for tooth colors. When assessing metal ceramic crowns, a previous study showed that a mean color difference between 0.4 units was perceptible between adjacent crowns with a range of 0 to 0.9 units. The same study also showed that the acceptability threshold was higher in comparison and ranged from 1.1 to 2.1 color difference units. Others have shown acceptability limits of 2.72 and 3.3 units.
Objectives and hypotheses
The aim of this study was to evaluate the alignment efficiency and esthetic performance of 4 coated NiTi archwires and compare these groups. Our null hypothesis was that there would be no difference in alignment efficiency, coating loss, and color change between the archwire groups.
Material and methods
Trial design
This was a parallel-group randomized clinical trial with a 1:1:1:1 allocation ratio.
Participants, eligibility criteria, and settings
This was a multicenter clinical trial, with 4 orthodontic departments participating in the study: Guy’s Hospital (London, United Kingdom), Royal Alexandra Children’s Hospital (Brighton, United Kingdom), Addenbrooke’s Hospital (Cambridge, United Kingdom), and Second University of Naples (Naples, Italy). Approval for this project was obtained from the East Midlands Research Ethics Committee (12/EM/0190) and from the Second University of Naples Ethics Committee (Protocol No.346). NHS Research and Development approval was also obtained at each of the 3 treatment centers in the United Kingdom.
The inclusion criteria were patients in the permanent dentition, requiring maxillary and mandibular orthodontic treatment with preadjusted edgewise appliances. Patients were excluded for any of the following: (1) unable to provide valid consent, (2) craniofacial syndromes, (3) teeth blocked out of the arch or ectopic teeth not allowing bracket placement and ligation at bond-up, or (4) hypodontia resulting in more than 1 tooth missing in any quadrant.
Interventions
The interventions investigated were 4 preformed coated NiTi archwires: (1) BioCosmetic (Forestadent, Pforzheim, Germany), 0.017 in; (2) Titanol (Forestadent), 0.016 in; (3) TP Aesthetic (TP Orthodontics, La Porte, Ind), 0.014 in; and (4) Tooth Tone (Ortho Organizers, Calsbad, Calif) 0.016 in. The patients had their maxillary and mandibular arches bonded with a preadjusted edgewise appliance (0.022 × 0.028-in slot), and the allocated archwires were ligated to all teeth at bond-up. Noncolored elastomeric modules were used throughout. The patients were recalled 4 weeks after bond-up for religation of the archwires and at 8 weeks after bond-up for archwire removal. On completion of the trial, the patients carried on with treatment using uncoated archwires.
Outcomes
The primary outcome was alignment efficiency; this was calculated by taking the difference in Little’s irregularity index. Measurements were taken of contact point displacements between canine and canine for maxillary and mandibular models on pretreatment and postintervention dental casts using digital calipers. The sums of the 5 readings were taken per arch. Measurements were then entered into an Excel spreadsheet (Microsoft, Redmond, Wash). The difference between the 2 measurements gave us the reductions in irregularity and alignment efficiency. All measurements were carried out blindly by the same examiner (Z.E.).
The 2 secondary outcome measures were assessed from the retrieved wires: percentage of coating loss from each wire and color change of the archwire coating. To measure coating loss, each archwire was placed against a green background and scanned using a color scanner (V750; Seiko Epson, Nagano, Japan). A resolution of 400 dpi was used; this provided sufficient image quality to carry out the required measurements ( Fig 1 ). The images were imported into AutoCAD 2014 software (Autodesk, San Rafael, Calif), and an arc was drawn over the archwire using the “draw arc” tool. The archwire lengths and coating loss lengths were obtained using the “arc dimension” tool. The measurements were made at an increased magnification to minimize error. Coating loss was determined using the following equation:
For color change measurements, a standardized photographic setup was created to produce images of the wires. A digital single reflex lens camera (D2x; Nikon, Tokyo, Japan) with a 60-mm macro lens (Nikon), shutter speed of 1/125 of a second, and aperture of f/2.0 was used throughout. The camera was attached to an adjustable mount at a distance of 60 cm from the copy stand. A 45/0 geometry setup was used with the object (archwire) at 0° to the detector (camera), with illumination from 2 lights at 45° to the object. All photographs were captured in “raw” format to prevent data loss and calibrated using a grey card (Fotowand, Südwalde, Germany) with 17.68% reflectance. The images ( Fig 2 ) were opened in Photoshop Creative Suite 6 (Adobe Systems, San Jose, Calif). Color measurements were made using the L*a*b* slider tool. There were 3 measurements made for each archwire at randomly selected points, and the average of these values was used. These were then compared with the respective unused control archwire.
Sample size calculation
The sample size was based on the ability to detect a clinically significant difference of 1 mm of irregularity between the archwire groups over an 8-week period with a standard deviation of 2.1 mm based on a previous study. It was determined that 138 patients (276 dental arches) would be required to detect a clinically significant difference between the groups with power of 80% and a significance level of 0.05. The sample was increased to 150 patients to allow for dropouts.
Interim analyses and stopping guidelines
Not applicable.
Randomization and allocation
The command “ralloc” was used in Stata software (version 14.1; StataCorp, College Station, Tex) to randomize the assignment of wires to patients and arches in each hospital. Block randomization was used with block sizes of 4 and 8 to achieve a 1:1:1:1 allocation ratio. The archwires were placed in a sealed opaque envelope for allocation. This was performed before trial commencement by contributors who were not involved in the treatment of patients (D.B., J.D.). The envelopes were held in a central location in each department by the clinic manager and opened once the patient was ready for treatment.
Blinding
Operator and patient blinding was not possible because the archwire coatings had different appearances: the TP Aesthetic archwire was coated only on the labial surface. Assessment was blinded for alignment because this was measured on study casts; however, outcome assessment was not blinded for coating loss and color change since it might have been possible to distinguish between the groups.
Statistical analysis
Data were analyzed using the Stata data analysis and statistical software. The analyst (M.S.) was blinded to the treatment groups. Reductions in irregularity, color change, and percentage of coating loss were the dependent variables. Archwire type, dental arch, extraction or nonextraction, and sex were the independent variables. To allow for each patient having 2 wires and the patients being clustered within hospital, the reduction in irregularity and color changes were analyzed using a multilevel mixed-effects linear regression, program “mixed.” The percentage of coating loss was analyzed using fractional response regression, program “fracreg,” again allowing for clustering. Other investigators used analysis of variance for statistical analysis; however, the percentage of loss is between 0% and 100%, and it has been argued that analysis of variance requires an unbounded continuous dependent variable. Our investigation therefore used the former analysis. The main effects in each model were compared using Šidák’s multiple comparison of means where appropriate.
Reproducibility
Reproducibility of the dental cast measurements was assessed by calculating the intraexaminer correlation by taking 20 measurements 2 weeks apart and calculating the correlation coefficient. The result of the coefficient of concordance of Lin was 0.959. This translates to a substantial strength of agreement. Measurements for color change and percentage of coating loss were repeated by the same examiner (A.U.) on 2 separate occasions, 2 weeks apart, for 35 archwires. All 3 parameters of color (L*a*b*) were evaluated separately for reproducibility. The Bland-Altman method was used to assess the 95% level of agreement between the 2 sets of measurements. The differences between the sets of measurements for all parameters were within acceptable limits and with minimal bias ( Table I ).
| Parameter | Difference | 95% limits of agreement | ||
|---|---|---|---|---|
| Average | SD | Lower | Upper | |
| L* | 0.295 | 0.490 | –0.666 | 1.256 |
| a* | 0.010 | 0.236 | –0.452 | 0.471 |
| b* | 0.143 | 0.532 | –0.900 | 1.185 |
| % coating loss | –0.149 | 0.548 | –1.223 | 0.926 |
Material and methods
Trial design
This was a parallel-group randomized clinical trial with a 1:1:1:1 allocation ratio.
Participants, eligibility criteria, and settings
This was a multicenter clinical trial, with 4 orthodontic departments participating in the study: Guy’s Hospital (London, United Kingdom), Royal Alexandra Children’s Hospital (Brighton, United Kingdom), Addenbrooke’s Hospital (Cambridge, United Kingdom), and Second University of Naples (Naples, Italy). Approval for this project was obtained from the East Midlands Research Ethics Committee (12/EM/0190) and from the Second University of Naples Ethics Committee (Protocol No.346). NHS Research and Development approval was also obtained at each of the 3 treatment centers in the United Kingdom.
The inclusion criteria were patients in the permanent dentition, requiring maxillary and mandibular orthodontic treatment with preadjusted edgewise appliances. Patients were excluded for any of the following: (1) unable to provide valid consent, (2) craniofacial syndromes, (3) teeth blocked out of the arch or ectopic teeth not allowing bracket placement and ligation at bond-up, or (4) hypodontia resulting in more than 1 tooth missing in any quadrant.
Interventions
The interventions investigated were 4 preformed coated NiTi archwires: (1) BioCosmetic (Forestadent, Pforzheim, Germany), 0.017 in; (2) Titanol (Forestadent), 0.016 in; (3) TP Aesthetic (TP Orthodontics, La Porte, Ind), 0.014 in; and (4) Tooth Tone (Ortho Organizers, Calsbad, Calif) 0.016 in. The patients had their maxillary and mandibular arches bonded with a preadjusted edgewise appliance (0.022 × 0.028-in slot), and the allocated archwires were ligated to all teeth at bond-up. Noncolored elastomeric modules were used throughout. The patients were recalled 4 weeks after bond-up for religation of the archwires and at 8 weeks after bond-up for archwire removal. On completion of the trial, the patients carried on with treatment using uncoated archwires.
Outcomes
The primary outcome was alignment efficiency; this was calculated by taking the difference in Little’s irregularity index. Measurements were taken of contact point displacements between canine and canine for maxillary and mandibular models on pretreatment and postintervention dental casts using digital calipers. The sums of the 5 readings were taken per arch. Measurements were then entered into an Excel spreadsheet (Microsoft, Redmond, Wash). The difference between the 2 measurements gave us the reductions in irregularity and alignment efficiency. All measurements were carried out blindly by the same examiner (Z.E.).
The 2 secondary outcome measures were assessed from the retrieved wires: percentage of coating loss from each wire and color change of the archwire coating. To measure coating loss, each archwire was placed against a green background and scanned using a color scanner (V750; Seiko Epson, Nagano, Japan). A resolution of 400 dpi was used; this provided sufficient image quality to carry out the required measurements ( Fig 1 ). The images were imported into AutoCAD 2014 software (Autodesk, San Rafael, Calif), and an arc was drawn over the archwire using the “draw arc” tool. The archwire lengths and coating loss lengths were obtained using the “arc dimension” tool. The measurements were made at an increased magnification to minimize error. Coating loss was determined using the following equation:
For color change measurements, a standardized photographic setup was created to produce images of the wires. A digital single reflex lens camera (D2x; Nikon, Tokyo, Japan) with a 60-mm macro lens (Nikon), shutter speed of 1/125 of a second, and aperture of f/2.0 was used throughout. The camera was attached to an adjustable mount at a distance of 60 cm from the copy stand. A 45/0 geometry setup was used with the object (archwire) at 0° to the detector (camera), with illumination from 2 lights at 45° to the object. All photographs were captured in “raw” format to prevent data loss and calibrated using a grey card (Fotowand, Südwalde, Germany) with 17.68% reflectance. The images ( Fig 2 ) were opened in Photoshop Creative Suite 6 (Adobe Systems, San Jose, Calif). Color measurements were made using the L*a*b* slider tool. There were 3 measurements made for each archwire at randomly selected points, and the average of these values was used. These were then compared with the respective unused control archwire.
Sample size calculation
The sample size was based on the ability to detect a clinically significant difference of 1 mm of irregularity between the archwire groups over an 8-week period with a standard deviation of 2.1 mm based on a previous study. It was determined that 138 patients (276 dental arches) would be required to detect a clinically significant difference between the groups with power of 80% and a significance level of 0.05. The sample was increased to 150 patients to allow for dropouts.
Interim analyses and stopping guidelines
Not applicable.
Randomization and allocation
The command “ralloc” was used in Stata software (version 14.1; StataCorp, College Station, Tex) to randomize the assignment of wires to patients and arches in each hospital. Block randomization was used with block sizes of 4 and 8 to achieve a 1:1:1:1 allocation ratio. The archwires were placed in a sealed opaque envelope for allocation. This was performed before trial commencement by contributors who were not involved in the treatment of patients (D.B., J.D.). The envelopes were held in a central location in each department by the clinic manager and opened once the patient was ready for treatment.
Blinding
Operator and patient blinding was not possible because the archwire coatings had different appearances: the TP Aesthetic archwire was coated only on the labial surface. Assessment was blinded for alignment because this was measured on study casts; however, outcome assessment was not blinded for coating loss and color change since it might have been possible to distinguish between the groups.
Statistical analysis
Data were analyzed using the Stata data analysis and statistical software. The analyst (M.S.) was blinded to the treatment groups. Reductions in irregularity, color change, and percentage of coating loss were the dependent variables. Archwire type, dental arch, extraction or nonextraction, and sex were the independent variables. To allow for each patient having 2 wires and the patients being clustered within hospital, the reduction in irregularity and color changes were analyzed using a multilevel mixed-effects linear regression, program “mixed.” The percentage of coating loss was analyzed using fractional response regression, program “fracreg,” again allowing for clustering. Other investigators used analysis of variance for statistical analysis; however, the percentage of loss is between 0% and 100%, and it has been argued that analysis of variance requires an unbounded continuous dependent variable. Our investigation therefore used the former analysis. The main effects in each model were compared using Šidák’s multiple comparison of means where appropriate.
Reproducibility
Reproducibility of the dental cast measurements was assessed by calculating the intraexaminer correlation by taking 20 measurements 2 weeks apart and calculating the correlation coefficient. The result of the coefficient of concordance of Lin was 0.959. This translates to a substantial strength of agreement. Measurements for color change and percentage of coating loss were repeated by the same examiner (A.U.) on 2 separate occasions, 2 weeks apart, for 35 archwires. All 3 parameters of color (L*a*b*) were evaluated separately for reproducibility. The Bland-Altman method was used to assess the 95% level of agreement between the 2 sets of measurements. The differences between the sets of measurements for all parameters were within acceptable limits and with minimal bias ( Table I ).
