Introduction
The purpose of this study was to quantify tooth movement among different retention protocols after the orthodontic appliances were removed.
Methods
A total of 90 patients were evaluated using the American Board of Orthodontics discrepancy index and the cast and radiograph evaluation at debond and the 1-year recall. These patients were equally divided into 3 retention protocols: upper Hawley/lower Hawley, upper Hawley/lower bonded, and upper Essix/lower bonded. The patients were then equally grouped by extraction or nonextraction treatment and case complexity. Paired t tests were used to compare the paired sample means. Analysis of variance tests were used to compare the means for more than 2 groups. A 2-sided 0.05 alpha level was used to define statistical significance.
Results
The upper Hawley/lower bonded showed the greatest amount of settling, and the upper Essix/lower bonded had the least settling, but these differences were statistically insignificant. The differences between the extraction and nonextraction treatments were not significant. The group with low discrepancy index scores showed significantly more settling than did the group with high discrepancy index scores.
Conclusions
The cast and radiograph evaluation variables that improved overall were marginal ridges, overjet, occlusal contacts, interproximal contacts, root angulation, and total cast and radiograph score. The cast and radiograph evaluation variables that worsened were alignment/rotation, buccolingual inclination, and occlusal relationship. Extraction or nonextraction treatment led to no real difference in settling. The discrepancy index, or initial case complexity, was the greatest factor in determining the improvement of occlusion or settling during the retention phase.
Settling is the natural vertical and horizontal movement of teeth into functionally stable interocclusal contacts after orthodontic treatment. It is important to remember the principles behind a stable occlusion. Andrews’ 6 keys to occlusion have been used in determining aspects of a stable occlusion. He stated that a complete and proper occlusion included molar interarch relationship, mesiodistal crown angulation, labiolingual crown inclination, absence of rotation, tight contacts, and curve of Spee. The periodontal ligament and connective tissues also influence the stability of the dentition if no circumferential supracrestal fiberotomies are performed, since they are altered during orthodontic treatment. Many authors have looked at the stability of the dentition after the orthodontic appliances have been removed and found that the overall stability and functionality of the occlusion improve during the retention period. However, some authors have found that the overall occlusion worsened during the retention period.
Studies have evaluated the change in occlusal contacts by settling during the retention period. They found that the number of occlusal contacts increased. One study showed that the increase in occlusal contacts was entirely attributed to changes in the posterior segment, whereas another study showed slight increases in the anterior segment along with the previously stated increase in the posterior segment.
If one is going to evaluate tooth movement during the retention period, it is important to know how different types of retention affect these tooth movements. It has been found that the maxillary removable Hawley appliance results in greater improvement of the occlusion through settling than do maxillary removable clear overlay retainers. One difficulty in evaluating improvement of occlusion during the retention period is to quantify positive or negative tooth movements. The cast and radiograph evaluation (CRE), formerly the objective grading system, was created by the American Board of Orthodontics (ABO) to evaluate excellence of occlusion for board certification. Alignment/rotation, marginal ridges, buccolingual inclination, overjet, occlusal contacts, occlusal relationship, interproximal contacts, and root angulation are the variables evaluated with the CRE. This method is applicable for consistently quantifying improvement or worsening after orthodontic treatment. Low CRE scores represent a better occlusion. High CRE scores represent more imperfections in the dentition. If the CRE score decreases, it shows an improvement in occlusion. If the CRE score increases, it shows that more teeth have moved out of proper position and might indicate relapse. Some studies have used the CRE to observe the changes in tooth movement from debond to different specified recall dates. One study found that the overall CRE score improved during the retention period. These authors found that the CRE variable of alignment/rotation worsened during the retention period. Marginal ridges, buccolingual inclinations, occlusal contacts, overjet, and occlusal relationships all improved. This study also used the peer assessment rating to assess initial case irregularity. They found that the peer assessment rating or initial case irregularity was not a good indicator of future settling during retention.
The literature has shown that teeth move after the orthodontic appliances are removed. We need to further understand the types of tooth movement after fixed orthodontic treatment. The purpose of this study was to quantify the differences in tooth movement, from the debond date to 1 year postdebond, among 3 retention protocols using the CRE. The sample was also separated into groups of extraction and nonextraction therapy, along with low and high case complexities, using the discrepancy index to determine whether there were any differences in tooth movement among these groups. The discrepancy index quantifies many different aspects of malocclusion and assigns point totals based on severity. For example, crowding greater than 7 mm will receive 7 points, and crowding of 3.5 mm will receive 2 points.
Material and methods
Records were evaluated from patients treated at the University of Oklahoma graduate orthodontic clinic between 2002 and 2010 to find equal numbers of patients for each retention protocol group. The subjects were selected based on availability of the records and also to be equally divided among retention protocol groups and type of treatment. This study was approved by the institutional review board.
The 90 patients were divided into 3 retention groups: upper Hawley/lower Hawley, upper Hawley/lower bonded canine to canine, and upper Essix/lower bonded canine to canine. Each group included 15 extraction and 15 nonextraction subjects for a total of 30 in each of the 3 groups. Of the 90 patients in the sample, 45 had extractions and 45 did not. The different retention groups were compared according to the 8 variables of the CRE at debond and the 1-year recall. The subjects were also separated into extraction and nonextraction groups, and groups with low (10-20) and high (>20) discrepancy index scores. The patients were instructed to wear the removable retainer full time for 6 months and then only at night thereafter.
The patient records needed for this study were pretreatment study models and cephalometric radiographs, debond study models and panoramic radiographs, and 1-year posttreatment models and panoramic radiographs.
The exclusion criteria for this study included patients with cleft lip and palate or other craniofacial syndromes, incomplete records, multiple missing teeth, continued habits that cause malocclusion, scores less than 10 on the discrepancy index, and implants or dental prosthetic bridges.
The sample included 49 female (54.5%) and 41 male (45.5%) subjects. Of the 90 patients, there were 77 whites (85%), 3 Asians (3%), 6 Hispanics (8%), and 4 blacks (4%). The average age was 15.2 years, with a range of 11.1 to 34.8 years. The average length of treatment time was 24.7 months (SD, 2.1 years) with a range of 16 to 35 months. The average length of retention time was 12.3 months (SD, 1 year) with a range of 8 to 18 months.
All subjects were evaluated with the ABO discrepancy index. This analysis quantifies the complexity of orthodontic treatment for each patient. Of the 90 patients, 57 had a discrepancy index score in the range of 10 to 20. There were 33 who scored 20 and over. All subjects analyzed in this study were board-qualifying cases with regard to case complexity measured by the discrepancy index. The upper Hawley/lower Hawley group had 19 Angle Class I, 8 Class II, and 3 Class III subjects. The upper Hawley/lower bonded group had 7 Class I, 18 Class II, and 2 Class III patients. The upper Essix/lower bonded group had 13 Class I, 14 Class II, and 3 Class III subjects. The upper Hawley/lower Hawley group had the most Class I patients. The upper Hawley/lower bonded group had the most Class II subjects.
All patients were measured according to the ABO’s CRE. The following aspects of occlusion were measured: alignment/rotation, marginal ridges, buccolingual inclination, overjet, occlusal contacts, occlusal relationship, interproximal contacts, and root angulation.
Statistical analysis
To test for error of measurement, 10 subjects were selected at random and rescored with the CRE. This score was compared with the original score, and the difference of each variable was calculated. Dahlberg’s formula, <SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='S2=∑d2/2n,’>S2=∑d2/2n,S2=∑d2/2n,
S 2 = ∑ d 2 / 2 n ,
was used to calculate the error of the measurements.
Software (version 9.2; SAS, Cary, NC) was used for the statistical analysis. Paired t tests were used to compare the paired sample means. Analysis of variance tests were used to compare the means for more than 2 groups. A 2-sided 0.05 alpha level was used to define statistical significance.
Results
Overall differences of the 90 subjects in CRE scores from debond to the 1-year recall for alignment/rotation, marginal ridges, buccolingual inclination, overjet, occlusal contacts, interproximal contacts, root angulation, and total CRE score proved to be statistically significant. The difference from debond to the recall for occlusal relationship showed statistical insignificance. The variables that improved in occlusion were marginal ridges, overjet, occlusal contacts, interproximal contacts, root angulation, and CRE total. The variables that worsened were alignment/rotation, buccolingual inclination, and occlusal relationship. The total CRE score improved from 29.8 to 25.2 points. This was an improvement of 4.6 points from debond to the 1-year recall ( Table I ).
| Variable | Average at T1 | Average at T2 | P value |
|---|---|---|---|
| Alignment/rotation | 7.43 | 8.46 | 0.0028 ∗ |
| Marginal ridges | 4.28 | 2.99 | <0.0001 ∗ |
| Buccolingual inclination | 2.57 | 2.99 | 0.0080 ∗ |
| Overjet | 4.00 | 2.62 | <0.0001 ∗ |
| Occlusal contacts | 4.89 | 2.57 | <0.0001 ∗ |
| Occlusal relationship | 3.08 | 3.48 | 0.1337 |
| Interproximal contacts | 1.47 | 0.57 | <0.0001 ∗ |
| Root angulation | 2.03 | 1.67 | <0.0001 ∗ |
| CRE total | 29.80 | 25.21 | <0.0001 ∗ |
Table II shows the comparison of variables among the different retention groups. There were statistical differences for marginal ridges, occlusal contacts, and interproximal contacts among the different retention groups. There was no statistical difference between the retention groups for alignment/rotation, buccolingual inclination, overjet, occlusal relationship, and root angulation. There was no statistical change from debond to the recall for the total CRE score among the 3 retention groups.
| Variable | Retention group | Mean of difference (T2-T1) | P value |
|---|---|---|---|
| Alignment/rotation | Upper Hawley/lower Hawley | 1.87 | 0.1323 |
| Upper Hawley/lower bonded 3-3 | 0.23 | ||
| Upper Essix/lower bonded 3-3 | 0.97 | ||
| Marginal ridges | Upper Hawley/lower Hawley | −2.23 | 0.0182 ∗ |
| Upper Hawley/lower bonded 3-3 | −1.13 | ||
| Upper Essix/lower bonded 3-3 | −0.50 | ||
| Buccolingual inclination | Upper Hawley/lower Hawley | 0.83 | 0.1450 |
| Upper Hawley/lower bonded 3-3 | 0.33 | ||
| Upper Essix/lower bonded 3-3 | 0.10 | ||
| Overjet | Upper Hawley/lower Hawley | −1.13 | 0.5437 |
| Upper Hawley/lower bonded 3-3 | −1.87 | ||
| Upper Essix/lower bonded 3-3 | −1.13 | ||
| Occlusal contacts | Upper Hawley/lower Hawley | −2.13 | 0.8190 |
| Upper Hawley/lower bonded 3-3 | −2.67 | ||
| Upper Essix/lower bonded 3-3 | −2.17 | ||
| Occlusal relationship | Upper Hawley/lower Hawley | 0.43 | 0.9846 |
| Upper Hawley/lower bonded 3-3 | 0.33 | ||
| Upper Essix/lower bonded 3-3 | 0.43 | ||
| Interproximal contacts | Upper Hawley/lower Hawley | −1.60 | 0.0162 ∗ |
| Upper Hawley/lower bonded 3-3 | −1.00 | ||
| Upper Essix/lower bonded 3-3 | −0.10 | ||
| Root angulation | Upper Hawley/lower Hawley | −0.57 | 0.1740 |
| Upper Hawley/lower bonded 3-3 | −0.37 | ||
| Upper Essix/lower bonded 3-3 | −0.17 | ||
| Total CRE | Upper Hawley/lower Hawley | −4.57 | 0.2784 |
| Upper Hawley/lower bonded 3-3 | −6.53 | ||
| Upper Essix/lower bonded 3-3 | −2.67 |
At debond, the alignment/rotation variable was the largest contributor to the overall CRE score at 25%. The interproximal contact variable contributed the least at 5% of the overall CRE score. At the recall, the alignment/rotation variable was the largest contributor to the overall CRE score at 34%. The interproximal contact variable contributed the least at 2% of the overall CRE score.
The variables of each retention group were compared at debond and recall ( Table III ). All 3 groups improved in total CRE scores. The upper Hawley/lower bonded retainer group demonstrated the greatest improvement in occlusion with 5 variables that showed statistical improvement in the occlusion. The upper Hawley/lower Hawley group had 4 variables that had statistical improvements, but 2 variables showed statistical worsening. The upper Essix/lower bonded retainer had 2 variables that showed statistically significant improvements. This group showed the least improvement among the 3 retention protocols ( Fig ).
