The objective of this research was to evaluate the anterior tooth alignment and dental arch dimension changes after orthodontic treatments with and without premolar extractions in the long-term.
Fifty-seven patients with Class I and Class II malocclusion were divided into 2 groups. Group 1 included 16 patients treated with nonextraction therapy, with mean initial (T1), final (T2), and long-term follow-up (T3) ages of 13.20 years, 15.07 years, and 50.32 years, respectively. The mean treatment (T2 − T1) and long-term follow-up (T3 − T2) times were 1.86 years and 35.25 years, respectively. Group 2 included 41 patients treated with extraction of 4 first premolars, with T1, T2, and T3 ages of 13.31 years, 15.63 years, and 53.60 years, respectively. The mean treatment and long-term follow-up times were 2.32 years and 37.96 years, respectively. The mean retention time was 2.26 years for both groups. The dental casts were obtained and digitized at T1, T2, and T3 stages. The following measurements were obtained: Little irregularity index, arch length, perimeter, and intercanine, interpremolar, and intermolar widths. Intragroup and intergroup comparisons were performed with repeated measures analysis of variance and t tests, respectively.
Anterior tooth irregularity index increased at T3 in both groups. In addition, all arch dimensions, except the intercanine width, were significantly smaller in the extraction group at T3. Both groups showed similar arch dimension changes at T3, except for the mandibular arch perimeter. The percentage of mandibular anterior tooth alignment change was significantly greater in the nonextraction than in the extraction group.
There was no difference in the changes of anterior alignment and transverse arch dimensions in patients treated with and without premolar extraction at T3, but the percentage of mandibular anterior tooth alignment changes was higher in the nonextraction than in the extraction patients at T3. The mandibular arch perimeter showed more of a decrease at T3 in extraction patients.
Anterior tooth alignment and dental arch dimensions were evaluated 37 years after treatment.
Changes were compared in extraction and nonextraction patients.
After 37 years, there was no difference in the amount of change between groups.
The percentage of mandibular tooth alignment change was greater in the nonextraction group.
Mandibular arch perimeter decreased more than in the extraction group.
Posttreatment stability is 1 of the most challenging aspects of orthodontic treatment and is a concern to all orthodontists. Moreover, long-term posttreatment stability of anterior tooth alignment is of major interest to both patients and clinicians. The anterior crowding relapse is an unforeseeable phenomenon that inevitably occurs in most treated patients. , The patient may interpret it as a treatment failure. In addition, the anterior teeth have a natural tendency to become more crowded over the years, even in untreated patients.
The long-term behavior of the anterior tooth alignment is unpredictable; no variable, such as the degree of pretreatment crowding, age, sex, and Angle classification, is useful in establishing prognosis. Typically, the arch width and length decrease after the removal of retention, regardless of the treatment had provided expansion or constriction. Two-thirds of the patients have unsatisfactory mandibular anterior alignment after retention.
Follow-up studies of treated patients show that although there is an improvement in the dentition, there is a tendency to return toward the original malocclusion many years after treatment. , This is known as relapse of orthodontic correction. It is also reported that irregularity increases are slightly greater in patients treated with mandibular premolar extractions and in patients followed over long periods.
However, many years after treatment and after retention, it is difficult to distinguish the relapse observed after orthodontic treatment and the long-term maturational changes of the aging occlusion, observed even in untreated subjects with normal occlusion. ,
Extraction in orthodontics has remained a subject of controversial debate and speculation over time. However, the 21st century introduced new features into the orthodontics specialty, and new esthetic concepts contributed to reducing the number of patients treated with extraction. , Dardengo et al stated that the frequency of tooth extraction over 32 years decreased by approximately 20%. However, teeth extraction for orthodontic purposes is still well indicated in many patients.
There is a lack in the orthodontic literature regarding what type of treatment has greater long-term stability and also what are the changes in the dental arches when comparing extraction and nonextraction treatments in the long-term. Most of the follow-up studies focused on morphologic changes in the mandibular arch evaluating only patients treated nonextraction , or with premolar extraction. , In addition, few studies focus mainly on changes in the maxillary arch, , and studies that focus on changes in both the maxillary and mandibular arches have a mean postretention follow-up time of 20 years. Only one study had a longer follow-up, but the authors evaluated only extraction patients. In addition, follow-up studies which compared the stability between nonextraction and extraction treatments evaluated only in the short-term , or in the long-term, but no more than 25 years after retention. , , , There is no known study evaluating the maxillary and mandibular crowding and dental arch dimensions’ changes between extraction and nonextraction treatments over more than 35 years after retention.
The objective of this study was to test the null hypothesis that there is no difference in the long-term changes in anterior tooth alignment and dental arch dimensions 35 years after retention between patients treated with and without premolar extractions.
Material and methods
This retrospective study was approved by the Ethics Research Committee of Bauru Dental School (protocol no. 71629217500005417, decision no 2268347), and all subjects signed informed consent.
The sample size calculation was based on an alpha significance level of 5% and a beta of 20% to achieve an 80% test power to detect a mean difference of 1.3 mm, with a standard deviation (SD) of 1.26 mm for the mandibular irregularity index. Thus, the sample size calculation determined that 16 subjects were needed for each group.
The sample was obtained from the files of the Department of Orthodontics, Bauru Dental School, University of São Paulo, Bauru, São Paulo, Brazil, from May 2017 to June 2019. The subjects were recalled, and the dental models were obtained.
The sample comprised dental casts from 57 patients with Class I or Class II malocclusion treated with nonextraction orthodontic therapy or 4 first premolar extraction. The dental casts were obtained at 3 different stages: pretreatment (T1), posttreatment (T2), and at a mean of 37 years long-term posttreatment (T3; an average of 35 years after retention). The inclusion criteria were: (1) Class I or Class II malocclusion at the beginning of orthodontic treatment; (2) treatment protocol nonextraction or with the extraction of 4 first premolars; (3) complete orthodontic treatment with full maxillary and mandibular fixed appliances (0.022 × 0.028-in slot); (4) all permanent teeth erupted up to the first molars at T1 stage; (5) no tooth agenesis or anomalies; and (6) maxillary removable appliance (Hawley plate) worn for 1 year, mandibular fixed canine-to-canine retainer worn for at least 1 year, and a maximum of 3 years after treatment, without retention at the time of follow-up. The mean retention time was 2.26 years.
The sample was divided into 2 groups. Group 1 comprised 16 subjects (10 girls, 6 boys) who received nonextraction orthodontic treatment. The mean initial maxillary and mandibular anterior crowding were: 8.54 mm (SD, 5.02) and 4.27 mm (SD, 2.73), respectively. The mean T1 age was 13.10 years (SD, 0.82), the mean treatment time (T2 − T1) was 1.82 years (SD, 0.82), and the long-term follow-up evaluation time (T3 − T2) was 35.25 years (SD, 6.11). Six patients presented Class I malocclusion, and 10 had Class II malocclusion.
Group 2 comprised 41 subjects (26 girls, 25 boys) treated with extraction of 4 first premolars. The mean initial maxillary and mandibular anterior crowding values were 9.67 mm (SD, 4.11) and 8.82 mm (SD, 3.99), respectively. The mean T1 age was 13.31 years (SD, 1.97), the mean treatment time was 2.32 years (SD, 0.59), and the long-term follow-up evaluation time was 37.96 years (SD, 4.54). Twenty-three patients had Class I malocclusion, and 18 presented Class II malocclusion.
In patients treated with extractions, the retraction of anterior teeth was performed by sliding mechanics with elastic chains. No patient underwent interproximal enamel reduction, rapid maxillary expansion, or fiberotomy to avoid postretention rotational relapse as part of the treatment plan. Class II elastics were used when necessary, especially in the Class II malocclusion patients treated with 4 premolars extraction.
All dental casts were digitized using an R700 3-dimensional scanner (3Shape, Copenhagen, Denmark). The dental casts measurements were performed using the OrthoAnalyzer 3-dimensional software (3Shape A/S). The following measurements were obtained for each set of dental casts: All measurements were linear, in millimeters, and were performed in both maxillary and mandibular arches by a single calibrated examiner (P.C.).
Little irregularity index ( Fig 1 ): the sum of the linear displacements of the anatomic contact points of each incisor from the adjacent tooth anatomic contact point. In this present study, the Little irregularity index was modified and also used to measure the irregularity of the maxillary anterior teeth. ,
Intercanine width ( Fig 2 ; black arrows ): the linear distance between the cusp tips of the right and left canines.
Interpremolar width ( Fig 2 ; black arrows ): the linear distance between the cusp tips of the second premolars.
Intermolar width ( Fig 2 ; black arrows ): the linear distance between the cusp tips of the first molars.
Arch length ( Fig 2 ; red arrows ): perpendicular length from the midpoint between the maxillary and mandibular central incisors to the line drawn between the mesial anatomic contact points of the first molars.
Arch perimeter ( Fig 2 ; yellow arrows ): the sum of the 4 segments from the mesial aspect of the right permanent first molar to the mesial aspect of the contralateral tooth.
The cusp tips were estimated in the dental casts presenting tooth with excessive dental wear.
The examiner (P.C.) was previously trained and calibrated with an experienced professor (K.M.S.F.). Only after proper calibration, the research measurements were performed.
The treatment changes were obtained from T2 − T1 values and the long-term posttreatment changes from T3 − T2 values. The percentage of changes of the anterior tooth alignment was obtained from the amount of change of the irregularity index from T2 to T3 in relation to the amount of correction with treatment (T2 − T1).
One month after the first measurement, 30% of the dental casts were randomly selected and remeasured by the same examiner (P.C.). Random and systematic errors were calculated according to Dahlberg’s formula and with dependent t tests, at P <0.05.
The normal distribution of the data was checked with the Shapiro-Wilk test. The intragroup comparison of the T1, T2, and T3 stages were performed with the repeated measures analysis of variance and Tukey tests when necessary. The intergroup comparability of sex distribution and type of malocclusion was performed with the chi-square test. The intergroup comparability of T1, T2, and T3 ages, the treatment, and long-term posttreatment evaluation times, were performed by independent t tests.
The intergroup comparisons of all variables studied at the 3 stages evaluated (T1, T2, and T3) and changes with treatment (T2 − T1) and long-term posttreatment (T3 − T2) were performed with independent t tests.
All statistical analyses were performed using the Statistica for Windows software (version 10.0; StatSoft, Tulsa, Okla) at P <0.05.
The random errors varied from 0.16 mm (Md 3-3 width) to 0.58 mm (Mx arch perimeter) and were within the acceptable limits. , Only 1 variable showed statistically significant systematic error ( Table I ).
|Variables||First measurement (n = 37)||Second measurement (n = 37)||Dahlberg||P|
|Maxillary dental casts measurements|
|Mx 3-3 width||33.83||2.42||33.81||2.49||0.21||0.750|
|Mx 5-5 width||43.20||2.84||43.19||2.49||0.19||0.835|
|Mx 6-6 width||48.90||2.44||48.77||2.47||0.25||0.031 ∗|
|Mx arch length||22.84||3.88||22.90||3.84||0.18||0.121|
|Mx arch perimeter||66.85||6.37||67.70||6.42||0.58||0.254|
|Mandibular dental casts measurements|
|Md 3-3 width||26.03||1.67||26.00||1.71||0.16||0.419|
|Md 5-5 width||36.84||2.90||36.76||2.94||0.26||0.193|
|Md 6-6 width||42.16||2.46||42.04||2.47||0.26||0.077|
|Md arch length||18.13||3.36||18.05||3.32||0.18||0.081|
|Md arch perimeter||57.70||7.15||57.70||7.08||0.36||0.964|
The groups were comparable regarding ages in all stages, long-term follow-up, retention time, distribution of sex, and type of malocclusion ( Table II ). The treatment time was significantly longer in the extraction group than in the nonextraction group ( Table II ).
|Variables||Group 1 nonextraction (n = 16)||Group 2 extraction (n = 41)||P|
|Age T1, y||13.20 (0.82)||13.31 (1.97)||0.841 ( t test)|
|Age T2, y||15.07 (1.16)||15.63 (2.14)||0.322 ( t test)|
|Age T3, y||50.32 (6.05)||53.60 (5.51)||0.054 ( t test)|
|Treatment Time, y||1.86 (0.82)||2.32 (0.59)||0.022 ∗ ( t test)|
|Long-term follow-up evaluation, y||35.25 (6.11)||37.96 (4.54)||0.071 ( t test)|
|Retention time, y||2.26 (1.17)||2.26 (1.26)||0.983 ( t test)|
|Sex||0.948 † (chi-square)|
|Type of malocclusion||0.206 ‡ (chi-square)|
In the nonextraction group, the maxillary Little irregularity index was significantly corrected with treatment and showed a statistically significant increase at T3 ( Table III ). The maxillary arch length and arch perimeter showed a statistically significant decrease from T2 to T3 ( Table III ). The mandibular Little irregularity index showed a statistically significant correction at T2, followed by a significant increase at T3, returning to values similar to the T1 stage ( Table III ). The mandibular arch length and perimeter showed a statistically significant decrease from T2 to T3 ( Table III ).
|Maxillary dental casts measurements|
|Mx Little||8.54 (5.02) a||1.40 (1.07) b||3.98 (2.35) c||<0.001 ∗|
|Mx 3-3 width||33.93 (2.70)||33.59 (1.57)||33.59 (1.73)||0.865|
|Mx 5-5 width||44.85 (4.06)||46.60 (2.32)||46.18 (2.63)||0.116|
|Mx 6-6 width||48.88 (4.03)||50.57 (2.67)||50.44 (2.81)||0.214|
|Mx arch length||26.98 (2.76) a||25.91 (1.69) a||23.83 (1.24) b||<0.001 ∗|
|Mx arch perimeter||75.23 (3.84) a||75.17 (3.69) a||71.73 (2.57) b||<0.001 ∗|
|Mandibular dental casts measurements|
|Md Little||4.27 (2.73) a||1.17 (0.89) b||3.78 (2.18) a||<0.001 ∗|
|Md 3-3 width||26.57 (1.76)||26.67 (2.81)||25.52 (2.24)||0.217|
|Md 5-5 width||39.04 (3.52)||39.53 (1.96)||39.14 (4.06)||0.678|
|Md 6-6 width||44.02 (3.55)||43.58 (2.63)||44.50 (2.95)||0.264|
|Md arch length||22.09 (1.50) a||21.71 (1.27) a||20.02 (1.03) b||<0.001 ∗|
|Md arch perimeter||65.65 (3.88) a||65.37 (3.04) a||63.16 (3.33) b||0.016 ∗|
In the extraction group, the maxillary Little irregularity index was significantly corrected at the T2 stage, followed by a statistically significant increase at T3 ( Table IV ). The maxillary intercanine width showed a statistically significant decrease from T2 to T3 ( Table IV ). The maxillary interpremolar and intermolar widths, arch length, and perimeter decreased significantly with treatment and continued to decrease significantly from T2 to T3 ( Table IV ). The mandibular Little irregularity index was significantly corrected at T2 and showed a significant increase at T3, although not returning to the T1 values ( Table IV ). The mandibular intercanine width had a statistically significant increase at T2 and a significant decrease at T3 ( Table IV ). The mandibular interpremolar width significantly decreased at T2 and continued to decrease significantly at T3 ( Table IV ). The mandibular intermolar width had a statistically significant decrease at T2 that remained stable at T3 ( Table IV ). The mandibular arch length and perimeter had a statistically significant decrease at T2 that continued to decrease at T3 ( Table IV ).