Effects of supplemental vibrational force on space closure, treatment duration, and occlusal outcome: A multicenter randomized clinical trial

Introduction

A multicenter parallel 3-arm randomized clinical trial was carried out in 3 university hospitals in the United Kingdom to investigate the effect of supplemental vibratory force on space closure and treatment outcome with fixed appliances.

Methods

Eighty-one subjects less than 20 years of age with mandibular incisor irregularity undergoing extraction-based fixed appliance treatment were randomly allocated to supplementary (20 minutes/day) use of an intraoral vibrational device (AcceleDent; OrthoAccel Technologies, Houston, Tex) (n = 29), an identical nonfunctional (sham) device (n = 25), or fixed-appliance only (n = 27). Space closure in the mandibular arch was measured from dental study casts taken at the start of space closure, at the next appointment, and at completion of space closure. Final records were taken at completion of treatment. Data were analyzed blindly on a per-protocol basis with descriptive statistics, 1-way analysis of variance, and linear regression modeling with 95% confidence intervals.

Results

Sixty-one subjects remained in the trial at start of space closure, with all 3 groups comparable for baseline characteristics. The overall median rate of initial mandibular arch space closure (primary outcome) was 0.89 mm per month with no difference for either the AcceleDent group (difference, −0.09 mm/month; 95% CI, −0.39 to 0.22 mm/month; P = 0.57) or the sham group (difference, −0.02 mm/month; 95% CI, −0.32 to 0.29 mm/month; P = 0.91) compared with the fixed only group. Similarly, no significant differences were identified between groups for secondary outcomes, including overall treatment duration (median, 18.6 months; P >0.05), number of visits (median, 12; P >0.05), and percentage of improvement in the Peer Assessment Rating (median, 90.0%; P >0.05).

Conclusions

Supplemental vibratory force during orthodontic treatment with fixed appliances does not affect space closure, treatment duration, total number of visits, or final occlusal outcome.

Registration

NCT02314975 .

Protocol

The protocol was not published before trial commencement.

Funding

AcceleDent units were donated by OrthoAccel Technologies; no contribution to the conduct or the writing of this study was made by the manufacturer.

Highlights

  • This was a randomized controlled trial investigating vibrational force and fixed appliance treatment.

  • Vibrational force does not increase the rate of space closure.

  • Vibrational force does not influence the final treatment outcome.

  • Vibrational force does not influence the overall treatment time.

Despite numerous innovations and advances in orthodontic appliance design and application, the average duration of comprehensive treatment with fixed appliances has remained relatively stable at just under 20 months. Accelerated orthodontic treatment is desirable, not only to limit the social and dental inconvenience of wearing fixed appliances, but also to help reduce the established risks of iatrogenic damage. Over the years, numerous innovations and adjuncts have been described that purport to speed up tooth movement and reduce overall treatment time. There is currently no robust evidence for faster tooth movement and reduced treatment time in association with any particular appliance design, bracket prescription, archwire composition, or treatment adjunct. The sole exceptions are surgical interventions, such as corticotomies or piezocision, that do seem to accelerate tooth movement, albeit on a relatively short-term basis. However, most of these surgical techniques are invasive and may not be readily acceptable to most patients. Therefore, continued efforts are directed toward the search for a safe, predictable, and acceptable method to reduce orthodontic treatment time, without compromising clinical results.

The use of supplemental vibrational force has been advocated as a method of speeding up orthodontic tooth movement. This involves the application of low-level vibration directly to the dentition as it is subjected to orthodontic force. The basic principle underlying orthodontic tooth movement is the ability of alveolar bone to respond with remodeling after the application of external force. With this principle, vibrational force has been shown to aid in the maintenance of bone mass in postmenopausal women and subjects with reduced mobility and prolonged bed rest. At the same time, data from animal models indicate increased rates of tooth movement, osteoclastic activity, and bone remodeling within the periodontium. These data have been used to inform the development of commercial vibrational appliances for clinical use, one of which is AcceleDent (OrthoAccel Technologies, Houston, Tex). This is a hands-free portable device consisting of an activator unit and a removable thermoplastic occlusal wafer that the patient bites onto. The activator unit vibrates and delivers a force of 0.2 N at a frequency of 30 Hz to the dentition. The manufacturer suggests that it should be used for 20 minutes per day to increase the speed of tooth movement and thereby reduce treatment time.

Clinical benefits from supplemental vibration have been reported in case reports and nonrandomized retrospective cohort studies. These investigations have shown increases in the rate of orthodontic tooth movement and reductions in treatment time, but their nonrandomized and retrospective design exposes them to potential bias and exaggerated treatment effects. There are data from randomized studies demonstrating statistically significant effects of supplemental vibration when delivered using either AcceleDent or a vibrating toothbrush during orthodontic treatment. These data are at both the clinical and biochemical levels, but again, the methodologic design of these studies predisposes them to a high risk of bias. These encouraging results have not been confirmed by other randomized clinical trials investigating rates of tooth movement; these trials found no significant benefit from supplemental vibrational force. However, these trials have only reported on the initial alignment phase with fixed appliances, and no robust evidence exists to date in relation to rates of space closure or overall treatment time when using fixed appliances with supplemental vibration.

Specific objectives and hypothesis

The aim of this study was to investigate the effect of AcceleDent appliance usage on the outcome of fixed appliance orthodontic treatment. The primary outcome measure for this component of the trial was initial rate of mandibular arch space closure, whereas secondary outcomes included overall rate of mandibular space closure, treatment duration, number of visits, appliance breakages and Peer Assessment Rating (PAR) reduction during treatment. The null hypothesis was that the use of supplemental vibrational force does not improve the rate of mandibular arch space closure, overall treatment duration, or outcome in subjects undergoing comprehensive extraction treatment with fixed appliances.

Material and methods

Trial design and any changes after trial commencement

Data for this investigation were gathered from the follow-up of a 3-arm parallel randomized controlled trial comparing the effect of supplemental vibrational force on orthodontic tooth alignment and are reported according to the CONSORT statement. Ethical approval was obtained from the National Research Ethics Service of the United Kingdom (South East London REC 3: 11/LO/0056), and written informed consent was received from all parents, guardians, and subjects. This trial was registered at the European Clinical Trials Database (EudraCT, 2014-004211-37) on September 29, 2014, and ClinicalTrials.gov ( NCT02314975 ) on November 25, 2014. No changes to the methodology occurred after trial commencement.

Participants, eligibility criteria, and settings

Participants were recruited from subjects referred to the orthodontic departments at King’s College London Dental Institute (Guy’s Hospital), the Royal Alexandra Children’s Hospital, Brighton, Sussex; and William Harvey Hospital, Ashford, Kent, United Kingdom. The former is based in a dental school, and the latter two are based in regional hospitals. All offer comprehensive orthodontic treatment for children and adults. Eligibility criteria have been previously described and included: (1) age less than 20 years at the start of treatment, (2) medically fit and well, (3) in the permanent dentition, (4) mandibular incisor irregularity, and (5) bilateral mandibular first premolar extractions as part of the treatment plan.

Interventions

Participants were randomly assigned to 1 of 3 groups: (1) preadjusted edgewise fixed appliance treatment with daily use of an AcceleDent vibrational device (Accel group); (2) preadjusted edgewise fixed appliance treatment with daily use of a nonfunctional AcceleDent device (sham group) provided by the manufacturer; or (3) preadjusted edgewise fixed appliance treatment alone (fixed only group). The subjects allocated to adjunctive devices were given direct verbal and written instructions on operation and usage, instructed to use it for 20 minutes per day at a time of their choosing, and informed that a timer was part of the device that allowed the investigator to monitor their compliance.

Bonding method and fixed appliances were standardized (precoated Victory series brackets; MBT prescription; 3M Unitek, Monrovia, Calif; bonding the mandibular permanent second molars) with a predetermined sequence of 0.014-in, 0.018-in, 0.017 × 0.025-in nickel-titanium, and 0.019 × 0.025-in stainless steel archwires. Data collection relating to the alignment phase of treatment took place at the start of treatment (baseline), placement of 0.018-in nickel-titanium wires (initial alignment), and placement of 0.019 × 0.025-in stainless steel wires (completion of alignment); these data have been previously reported. All appointments were made as part of the routine orthodontic treatment provided in the participating departments and scheduled at approximately 6-week intervals. No biteplanes, auxiliary arches, or headgears were used during space closure, but intermaxillary elastics were permitted as prescribed. All subjects were treated by senior orthodontists (A.T.D., N.J., C.S., J.G., M.T.C.) or postgraduate specialist trainees (N.R.W., M.A.) under their direct supervision.

For this component of the trial, space closure was initiated at the first visit after placement of a 19 × 25-in stainless steel working archwire (completion of alignment) and undertaken using 9-mm nickel-titanium coil springs attached from the first molar to hooks placed on the archwire between lateral incisor and canine, and stretched to no more than twice their length, as per the manufacturer’s instructions. Data were collected at the start of mandibular space closure (T1), at the first visit after initiation of space closure (T2), at the end of space closure in the mandibular arch (T3), and at completion of treatment on removal of the fixed appliances (T4). Coil springs were checked during routine adjustments between T1 and T3 and retied. If there was any sign of damage, they were replaced with a spring of the same dimensions. Specifically, dated mandibular (T1-T3) and both maxillary and mandibular (T4) alginate impressions were taken for the generation of dental study casts.

For mandibular arch space closure, space was measured using digital calipers (IP67 150 mm; Mitutoyo, Andover, United Kingdom) by placing the caliper tip from the most concave contact point to the contact point between the mandibular second premolar and canine bilaterally and calculating a mean value for each subject. A sample of 20 subjects was randomly chosen and remeasured by the same assessor (M.A.) after 2 weeks for repeatability. Repeatability and agreement of the measurements were assessed with the concordance correlation coefficient and the Bland-Altman method. Monthly rates of mandibular arch space closure were calculated by dividing the mean space closure value by the exact number of space closure days divided by 30 (days).

All subjects in the trial had first premolar extractions in the mandibular arch. In the maxillary arch, all subjects had 1 tooth extracted in each quadrant, but these extraction patterns varied and were classified as those with premolar, canine, or incisor extractions or a combination.

For the PAR index, dental casts taken at baseline and T4 were scored by a calibrated examiner (Y.K.).

Outcomes (primary and secondary)

The primary outcome measure for this component of the trial was initial rate of mandibular arch space closure (T1-T2, calculated as millmeters/month). Secondary outcomes included overall rate of mandibular arch-space closure (millimeters/month), overall treatment duration (months), overall number of visits, number of appliance breakages, and both absolute and relative percentages of PAR reduction during treatment.

Sample size calculation

The primary outcome for this component of the trial was the initial rate of mandibular arch space closure. No formal sample size calculation was performed for this component because it was a follow-up examination of a previous randomized clinical trial. However, a previous randomized trial investigating 3 methods of orthodontic space closure calculated that 11 subjects per group (33 subjects in total) would yield a power of 90% to detect a clinically significant difference in space closure at quadrant level (0.75 ± 0.50 mm/month) with α = 5%, which indicated that the primary outcome for this component of the trial was adequately powered.

Randomization

The randomization sequence was computer generated using GraphPad online software ( www.graphpad.com/quickcalcs/index.cfm ) with participant allocation undertaken centrally at King’s College London, independently from the clinical operators after recruitment (allocation concealment). No restricted randomization or stratification was used.

Blinding

By the nature of the trial intervention, subjects and treating clinicians were aware of treatment group allocations. Dental casts were coded so that all measurements were undertaken blindly. All dental cast linear measurements were carried out blindly by 1 investigator (M.A.). PAR scoring was also conducted blindly for all dental casts by a calibrated examiner (Y.K.).

Statistical analysis

Statistical analysis was conducted on a per-protocol basis and blinded with a coded data set, where the code was broken after final provision of the analysis results. Data normality was checked via visual inspection of distributional diagrams and formal testing with the Shapiro-Wilk test. Since all outcomes were nonnormally distributed ( P <0.05), descriptive statistics consisted of medians and interquartile ranges (IQRs). Initial crude differences among randomized groups were calculated with the Kruskal-Wallis 1-way analysis of variance.

Subsequent linear regression models were fitted with independent variables for either the randomization group (crude analysis) or additional confounders, including the possible influence of study center. Choice of the latter was based on both clinical judgment and whether bivariable model fit improved with a criterion-based method using a model with just the dependent variable. Assumptions of linear regression for all fitted models were checked including graphic and statistical tests for homoscedasticity of residuals, multicollinearity of predictors, and model misspecification. Results are reported as unstandardized coefficients and 95% confidence intervals (CIs) with α set at 5%.

Post hoc explorative analyses were conducted to investigate any systematic differences across centers in terms of average time interval between appointments. Additionally, interactions of randomized intervention effects with baseline severity of irregularity, baseline extraction spaces, and baseline PAR scores were investigated (with cutoffs of 7 mm, 7 mm, and 30 points, respectively). The main (per protocol) analysis that was conducted excluded subjects who reported not using their Accel or sham appliance (n = 9), and cases of early fixed appliance removal at subject request (n = 2), subjects with more than 3 missed appointments (n = 5), those with more than 5 episodes of fixed appliance breakage (n = 2), 1 patient with an impacted maxillary canine, and 1 patient undergoing orthognathic surgery. A separate sensitivity analysis was performed with the intention-to-treat sample by including all excluded subjects with available data and compared with the main analysis for robustness.

Material and methods

Trial design and any changes after trial commencement

Data for this investigation were gathered from the follow-up of a 3-arm parallel randomized controlled trial comparing the effect of supplemental vibrational force on orthodontic tooth alignment and are reported according to the CONSORT statement. Ethical approval was obtained from the National Research Ethics Service of the United Kingdom (South East London REC 3: 11/LO/0056), and written informed consent was received from all parents, guardians, and subjects. This trial was registered at the European Clinical Trials Database (EudraCT, 2014-004211-37) on September 29, 2014, and ClinicalTrials.gov ( NCT02314975 ) on November 25, 2014. No changes to the methodology occurred after trial commencement.

Participants, eligibility criteria, and settings

Participants were recruited from subjects referred to the orthodontic departments at King’s College London Dental Institute (Guy’s Hospital), the Royal Alexandra Children’s Hospital, Brighton, Sussex; and William Harvey Hospital, Ashford, Kent, United Kingdom. The former is based in a dental school, and the latter two are based in regional hospitals. All offer comprehensive orthodontic treatment for children and adults. Eligibility criteria have been previously described and included: (1) age less than 20 years at the start of treatment, (2) medically fit and well, (3) in the permanent dentition, (4) mandibular incisor irregularity, and (5) bilateral mandibular first premolar extractions as part of the treatment plan.

Interventions

Participants were randomly assigned to 1 of 3 groups: (1) preadjusted edgewise fixed appliance treatment with daily use of an AcceleDent vibrational device (Accel group); (2) preadjusted edgewise fixed appliance treatment with daily use of a nonfunctional AcceleDent device (sham group) provided by the manufacturer; or (3) preadjusted edgewise fixed appliance treatment alone (fixed only group). The subjects allocated to adjunctive devices were given direct verbal and written instructions on operation and usage, instructed to use it for 20 minutes per day at a time of their choosing, and informed that a timer was part of the device that allowed the investigator to monitor their compliance.

Bonding method and fixed appliances were standardized (precoated Victory series brackets; MBT prescription; 3M Unitek, Monrovia, Calif; bonding the mandibular permanent second molars) with a predetermined sequence of 0.014-in, 0.018-in, 0.017 × 0.025-in nickel-titanium, and 0.019 × 0.025-in stainless steel archwires. Data collection relating to the alignment phase of treatment took place at the start of treatment (baseline), placement of 0.018-in nickel-titanium wires (initial alignment), and placement of 0.019 × 0.025-in stainless steel wires (completion of alignment); these data have been previously reported. All appointments were made as part of the routine orthodontic treatment provided in the participating departments and scheduled at approximately 6-week intervals. No biteplanes, auxiliary arches, or headgears were used during space closure, but intermaxillary elastics were permitted as prescribed. All subjects were treated by senior orthodontists (A.T.D., N.J., C.S., J.G., M.T.C.) or postgraduate specialist trainees (N.R.W., M.A.) under their direct supervision.

For this component of the trial, space closure was initiated at the first visit after placement of a 19 × 25-in stainless steel working archwire (completion of alignment) and undertaken using 9-mm nickel-titanium coil springs attached from the first molar to hooks placed on the archwire between lateral incisor and canine, and stretched to no more than twice their length, as per the manufacturer’s instructions. Data were collected at the start of mandibular space closure (T1), at the first visit after initiation of space closure (T2), at the end of space closure in the mandibular arch (T3), and at completion of treatment on removal of the fixed appliances (T4). Coil springs were checked during routine adjustments between T1 and T3 and retied. If there was any sign of damage, they were replaced with a spring of the same dimensions. Specifically, dated mandibular (T1-T3) and both maxillary and mandibular (T4) alginate impressions were taken for the generation of dental study casts.

For mandibular arch space closure, space was measured using digital calipers (IP67 150 mm; Mitutoyo, Andover, United Kingdom) by placing the caliper tip from the most concave contact point to the contact point between the mandibular second premolar and canine bilaterally and calculating a mean value for each subject. A sample of 20 subjects was randomly chosen and remeasured by the same assessor (M.A.) after 2 weeks for repeatability. Repeatability and agreement of the measurements were assessed with the concordance correlation coefficient and the Bland-Altman method. Monthly rates of mandibular arch space closure were calculated by dividing the mean space closure value by the exact number of space closure days divided by 30 (days).

All subjects in the trial had first premolar extractions in the mandibular arch. In the maxillary arch, all subjects had 1 tooth extracted in each quadrant, but these extraction patterns varied and were classified as those with premolar, canine, or incisor extractions or a combination.

For the PAR index, dental casts taken at baseline and T4 were scored by a calibrated examiner (Y.K.).

Outcomes (primary and secondary)

The primary outcome measure for this component of the trial was initial rate of mandibular arch space closure (T1-T2, calculated as millmeters/month). Secondary outcomes included overall rate of mandibular arch-space closure (millimeters/month), overall treatment duration (months), overall number of visits, number of appliance breakages, and both absolute and relative percentages of PAR reduction during treatment.

Sample size calculation

The primary outcome for this component of the trial was the initial rate of mandibular arch space closure. No formal sample size calculation was performed for this component because it was a follow-up examination of a previous randomized clinical trial. However, a previous randomized trial investigating 3 methods of orthodontic space closure calculated that 11 subjects per group (33 subjects in total) would yield a power of 90% to detect a clinically significant difference in space closure at quadrant level (0.75 ± 0.50 mm/month) with α = 5%, which indicated that the primary outcome for this component of the trial was adequately powered.

Randomization

The randomization sequence was computer generated using GraphPad online software ( www.graphpad.com/quickcalcs/index.cfm ) with participant allocation undertaken centrally at King’s College London, independently from the clinical operators after recruitment (allocation concealment). No restricted randomization or stratification was used.

Blinding

By the nature of the trial intervention, subjects and treating clinicians were aware of treatment group allocations. Dental casts were coded so that all measurements were undertaken blindly. All dental cast linear measurements were carried out blindly by 1 investigator (M.A.). PAR scoring was also conducted blindly for all dental casts by a calibrated examiner (Y.K.).

Statistical analysis

Statistical analysis was conducted on a per-protocol basis and blinded with a coded data set, where the code was broken after final provision of the analysis results. Data normality was checked via visual inspection of distributional diagrams and formal testing with the Shapiro-Wilk test. Since all outcomes were nonnormally distributed ( P <0.05), descriptive statistics consisted of medians and interquartile ranges (IQRs). Initial crude differences among randomized groups were calculated with the Kruskal-Wallis 1-way analysis of variance.

Subsequent linear regression models were fitted with independent variables for either the randomization group (crude analysis) or additional confounders, including the possible influence of study center. Choice of the latter was based on both clinical judgment and whether bivariable model fit improved with a criterion-based method using a model with just the dependent variable. Assumptions of linear regression for all fitted models were checked including graphic and statistical tests for homoscedasticity of residuals, multicollinearity of predictors, and model misspecification. Results are reported as unstandardized coefficients and 95% confidence intervals (CIs) with α set at 5%.

Post hoc explorative analyses were conducted to investigate any systematic differences across centers in terms of average time interval between appointments. Additionally, interactions of randomized intervention effects with baseline severity of irregularity, baseline extraction spaces, and baseline PAR scores were investigated (with cutoffs of 7 mm, 7 mm, and 30 points, respectively). The main (per protocol) analysis that was conducted excluded subjects who reported not using their Accel or sham appliance (n = 9), and cases of early fixed appliance removal at subject request (n = 2), subjects with more than 3 missed appointments (n = 5), those with more than 5 episodes of fixed appliance breakage (n = 2), 1 patient with an impacted maxillary canine, and 1 patient undergoing orthognathic surgery. A separate sensitivity analysis was performed with the intention-to-treat sample by including all excluded subjects with available data and compared with the main analysis for robustness.

Results

A CONSORT diagram demonstrating subject flow through the trial is shown in Figure 1 . Eighty-one subjects were recruited into the trial between July 2011 and May 2014, with 29 allocated to the Accel group, 25 to the Accel sham group, and 27 to the fixed only group. The total randomized sample consisted of 40 boys and 41 girls with a mean age of 14.1 years (SD, 1.7). The mean ages were 13.9 years (SD, 1.6) for subjects allocated to the Accel group, 14.1 years (SD, 1.9) for the Accel sham group, and 14.4 years (SD, 1.8) for the fixed only group.

Fig 1
CONSORT diagram showing the flow of subjects through the trial. B , Baseline; T1 , start of space closure; T2 , first visit after T1; T3 , end of space closure; T4 , completion of treatment. *One subject was not analyzed for PAR score because the final model was not available. +In the Accel and Accel sham groups, 1 subject was lost from each group because they discontinued using the device; in the fixed only group, 1 subject was lost through early removal of the fixed appliance; all were lost before the completion of incisor alignment. In total, 6 subjects were lost from Brighton, 8 subjects were lost from Guy’s, and 6 subjects were lost from Canterbury.

Table I shows baseline demographics of subjects investigated in this component of the trial. A total of 61 subjects remained in the trial at T1, including 22 in the Accel group, 19 in the Accel sham group, and 20 in the fixed only group. These 3 groups were comparable for all patient characteristics at baseline ( Table I ).

Table I
Baseline characteristics of trial subjects
Characteristic Overall Accel group Sham group Fixed only group
Patients, n 61 22 19 20
Center Brighton, n (%) 27 (44%) 8 (36%) 11 (58%) 8 (40%)
Center Guy’s, n (%) 10 (16%) 4 (18%) 1 (5%) 5 (25%)
Center Canterbury, n (%) 24 (39%) 10 (45%) 7 (37%) 7 (35%)
Male, n (%) 30 (49%) 11 (50%) 8 (42%) 11 (55%)
Age (y), mean (SD) 13.9 (1.5) 13.6 (1.4) 13.9 (1.5) 14.3 (1.7)
Baseline irregularity, mean (SD) 7.8 (3.2) 7.5 (3.2) 7.9 (3.3) 8.0 (3.3)
Baseline PAR , mean (SD) 33.0 (11.0) 34.0 (11.7) 30.9 (9.6) 33.9 (11.9)
Class I, n (%) 22 (36%) 8 (36%) 7 (37%) 7 (35%)
Class II, n (%) 22 (36%) 9 (41%) 6 (32%) 7 (35%)
Class III, n (%) 17 (28%) 5 (23%) 6 (32%) 6 (30%)
Premolar extractions, n (%) 53 (87%) 21 (95%) 17 (89%) 15 (75%)
Canine/incisor extractions, n (%) 2 (3%) 0 (0%) 0 (0%) 2 (10%)
Mixed extractions, n (%) 6 (10%) 1 (5%) 2 (11%) 3 (15%)
Extraction spaces, mean (SD) 8.0 (2.3) 8.3 (2.4) 8.2 (2.2) 7.4 (2.4)

59 not 61 subjects (21, 19, and 19 subjects in the Accel, sham, and fixed only groups, respectively).

All subjects had mandibular first premolar extractions; these extraction categories relate to teeth that were extracted in the maxillary arch.

Outcomes and estimation

The mean time periods were 68 ± 28 days from T1 to T2 and 172 ± 79 days from T1 to T3. For the primary outcome of the initial rate of mandibular arch space closure, the median across all randomized subjects was 0.89 mm per month (IQR, 0.56-1.33 mm/month) with no significant differences among groups ( P = 0.61; Fig 2 ; Table II ) . In addition, no significant differences among groups were identified for any secondary outcomes, including overall rate of mandibular arch space closure (median [IQR], 0.74 [0.56-1.00] mm/month), overall treatment duration (median [IQR], 18.57 [16.3-23.9] months), number of visits (median [IQR], 12 [10-16] visits), number of breakages (median [IQR], 2 [1-3] breakages), final PAR score (median [IQR], 3 [2-4] points), absolute (median [IQR], 28 [21-35] points), or percentage of improvement in PAR score (median [IQR], 90.0% [84.6%-93.8%]) ( P >0.05 in all instances).

Fig 2
Cumulative graph showing predicted marginal treatment effects for primary and secondary outcomes of the trial. Results are plotted as unstandardized regression coefficients with 95% confidence intervals ( blue ) based on the adjusted models from Table III . The graph has been augmented with contours of effect magnitude ( grey shades ) based on the standard deviations in the reference group (fixed only) for each outcome.

Table II
Descriptive statistics and crude differences across groups
Accel group Sham group Fixed only group P
n Median IQR n Median IQR n Median IQR
Initial rate of space closure T1-T2 (mm/mo) 22 0.82 0.48, 1.33 19 0.89 0.56, 1.13 20 0.95 0.65, 1.47 0.61
Monthly space closure rate T1-T3 (mm/mo) 22 0.82 0.62, 1.06 19 0.68 0.57, 0.80 19 0.76 0.52, 1.01 0.42
Treatment duration (mo) 22 20.45 17.63, 25.23 19 16.73 15.33, 22.07 20 17.64 14.87, 23.89 0.07
Visits (n) 22 14.5 11.0, 17.0 19 11.0 10.1, 14.0 20 11.0 9.5, 16.0 0.13
Breakages (n) 22 2.0 1.0, 3.0 19 2.0 1.0, 3.0 20 2.0 1.0, 2.0 0.95
PAR at T4 21 3.0 2.0, 6.0 19 3.0 2.0, 4.0 19 3.0 2.0, 4.0 0.60
PAR improvement 21 28.0 21.0, 38.0 19 27.0 20.0, 35.0 19 29.0 24.0, 31.0 0.91
PAR Improvement (%) 21 88.9 80.9, 94.4 19 90.0 87.5, 92.0 19 91.2 85.0, 93.5 0.74
Data are given as medians and interquartile ranges (IQR) and not as means and standard deviations, because they were not normally distributed.

Kruskal-Wallis test.

These findings were also confirmed by regression analyses with either crude (including only experimental groups and study center effects) or adjusted (experimental groups, study center effects, and confounders) models ( Table III ). No differences were found in initial space closure rates between either the Accel group or the sham and fixed only groups ( P = 0.57 and P = 0.91, respectively). The only factors that significantly influenced space closure rate were patient sex, extraction category in the maxillary arch, and the amount of initial space to be closed (with boys, extraction of maxillary anterior teeth and increased baseline space were positively associated with closure rate; P <0.05 in all 3 cases; Table III ) .

Table III
Results of the crude and adjusted regression models for primary and secondary outcomes
Crude model: only group Adjusted modeling: group and confounding factors
b 95% CI P b 95% CI P
Initial rate of space closure (T1-T2)
Group (reference, fixed only)
Accel −0.10 −0.44, 0.25 0.57 −0.07 −0.39, 0.25 0.67
Sham −0.03 −0.39, 0.32 0.85 −0.02 −0.34, 0.31 0.92
Sex NT 0.29 0.02, 0.56 0.04
Malocclusion NT 0.12 −0.05, 0.28 0.15
Extraction NT 0.24 0.02, 0.47 0.03
Spaces T3 NT 0.08 0.03, 0.14 0.005
Overall rate of space closure (T1-T3)
Group (reference, fixed only)
Accel 0.11 −0.19 ,0.40 0.47 0.11 −0.17, 0.39 0.43
Sham −0.11 −0.41, 0.190 0.45 −0.12 −0.41, 0.16 0.39
Sex NT 0.18 −0.05, 0.41 0.12
Malocclusion NT 0.15 0.01, 0.29 0.04
Extraction NT 0.16 −0.03, 0.35 0.11
Spaces T3 NT 0.05 0.00, 0.10 0.03
Treatment duration
Group (reference, fixed only)
Accel 3.00 −0.39, 6.40 0.08 2.19 −0.70, 5.07 0.13
Sham −0.27 −3.79, 3.26 0.88 0.85 −2.21, 3.90 0.58
Sex NT −0.84 −3.24, 1.57 0.49
PAR T1 NT 0.11 −0.01, 0.22 0.07
Center (reference, Brighton)
Guys NT 3.21 2.06, 9.14 0.002
Ashford NT −4.31 2.84, 8.27 <0.001
Number of visits
Group (reference, fixed only)
Accel 1.45 −0.69, 3.59 0.18 1.01 −0.89, 2.90 0.29
Sham −0.97 −3.20, 1.25 0.39 −0.11 −2.10, 1.88 0.91
Sex NT 0.10 −0.92, 1.13 0.84
Irregularity T1 NT −0.13 −0.40, 0.14 0.33
PAR T1 NT 0.05 −0.03, 0.12 0.20
Center (reference, Brighton)
Guys NT 4.24 1.81, 6.67 0.001
Ashford NT 2.81 0.97, 4.64 0.003
Breakages (n)
Group (reference, fixed only)
Accel 0.16 −1.10, 1.42 0.80 0.10 −1.12, 1.32 0.87
Sham −0.09 −1.40, 1.21 0.89 −0.31 −1.60, 0.97 0.63
Malocclusion NT 0.63 −0.03, 1.29 0.06
PAR T1 NT −0.01 −0.06, 0.04 0.68
Center (reference, Brighton)
Guys NT −0.85 −2.36, 0.66 0.26
Ashford NT 1.04 −0.15, 2.23 0.09
Change in PAR
Group (reference, fixed only)
Accel −0.51 −7.43, 6.42 0.88 −0.35 −1.60, 0.90 0.57
Sham −2.37 −9.46, 4.72 0.51 0.69 −0.62, 2.01 0.30
PAR T1 NT 0.98 0.93, 1.03 <0.001
Center (reference, Brighton)
Guys NT 0.86 −0.67, 2.39 0.26
Ashford NT −1.37 −2.54, −0.19 0.02
Change in PAR (%)
Group (reference, fixed only)
Accel −2.22 −6.52, 2.08 0.31 −1.54 −5.10, 2.03 0.39
Sham 1.19 −3.22, 5.59 0.59 2.78 −0.98, 6.54 0.14
PAR T1 NT 0.31 0.17, 0.45 <0.001
Center (reference, Brighton)
Guys NT 3.21 −1.15, 7.57 0.15
Ashford NT −4.31 −7.67, −0.96 0.01
Only gold members can continue reading. Log In or Register to continue

Dec 12, 2018 | Posted by in Orthodontics | Comments Off on Effects of supplemental vibrational force on space closure, treatment duration, and occlusal outcome: A multicenter randomized clinical trial
Premium Wordpress Themes by UFO Themes