- Email: [email protected]
Effects of posttreatment skeletal maturity measured with the cervical vertebral maturation method on incisor alignment relapse
ORIGINAL ARTICLE
Effects of posttreatment skeletal maturity measured with the cervical vertebral maturation method on incisor alignment relapse Piotr Fudalej,a Laura E. Rothe,b and Anne-Marie Bollenc Warsaw, Poland, Spokane, and Seattle, Wash Introduction: Our aim was to test the hypothesis that relapse of incisor alignment is associated with skeletal maturity at the end of treatment, as assessed with the cervical vertebral maturation (CVM) method. Methods: This was a case-control study with information from the postretention database at the University of Washington. Mandibular incisor irregularity (II) at least 10 years out of retention (T3) was used to define the subjects (II ⬎6 mm, relapse group) and the controls (II ⬍3.5 mm, stable group). The following model measurements were made: II at pretreatment (T1), II at posttreatment (T2), and intercanine width at T1 and T2. On cephalograms taken T2, the CVM status was determined. Logistic regression analyses were used to determine the association between relapse and CVM status after treatment. The models were adjusted for potentially confounding variables (II at pretreatment and posttreatment, intercanine width change during treatment, sex, age at T2, and treatment alternatives). Results: No association between CVM stage at T2 and relapse was found (P ⫽ 0.89). Both groups had similar distributions of the CVM stages (P ⬎0.05). Pretreatment II and postretention time were found to be correlated with long-term incisor stability (P ⫽ 0.007 and 0.034, respectively). Sex was not related to relapse (P ⫽ 0.33). Conclusions: Maturity of craniofacial structures at the end of treatment evaluated with the CVM method is not associated with long-term stability of incisor alignment. (Am J Orthod Dentofacial Orthop 2008;134:238-44)
L
ong-term stability of orthodontically aligned teeth frequently poses a serious challenge. Many studies1-12 have demonstrated that relapse, particularly of mandibular incisor alignment, affects more than half of the patients 10 years after discontinuation of retention.6 Changes from 10 to 20 years out of retention further increase incisor irregularity (II) so that less than 30% of patients have satisfactory dental alignment 20 years after discontinuing retention.13 Among the potential causes of incisor relapse discussed in recent literature reviews, the growth of dentofacial structures merits attention.14,15 When a typical orthodontic treatment is completed, variable skeletal maturity of craniofacial structures remains. Since II, in both orthodontically treated and untreated subjects, has the highest rate of increase during the late a
Assistant professor, Center for Craniofacial Disorders, National Research Institute for Mother and Child, Warsaw, Poland. b Private practice, Spokane, Wash. c Professor, Department of Orthodontics; University of Washington, Seattle. Reprint requests to: Piotr Fudalej, Center for Craniofacial Disorders, National Research Institute for Mother and Child, Kasprzaka Str. 17a, Warsaw, Poland; e-mail, [email protected]. Submitted, July 2006; revised and accepted, September 2006. 0889-5406/$34.00 Copyright © 2008 by the American Association of Orthodontists. doi:10.1016/j.ajodo.2006.09.060
238
teens and early twenties,13,16 circumpubertal growth is suspected to be related to crowding.17-20 A finding that craniofacial growth continues, although at a limited rate, after maturity aroused additional interest in the search for an association between craniofacial growth and long-term stability of incisor alignment.21-26 Driscoll-Gilliland et al17 followed a sample of treated and untreated subjects from 14 to 29 years of age and found a similar increase of incisor irregularity in both groups. They demonstrated that the development of crowding was correlated with vertical facial growth and eruption of the incisors. Richardson18 examined skeletal and dental changes from 13 to 18 years of age and concluded that mandibular growth rotation contributed to the development of crowding. Perera19 investigated craniofacial changes in untreated subjects from 11 to 20 years of age and found that mandibular rotation and mandibular incisor crowding were associated, and the correlation coefficient equaled 0.51. Ormiston et al12 compared 45 subjects with stable incisor alignment and 41 subjects with unstable incisor alignment and found that both male sex and sustained period of growth were associated with increased instability. Opposite conclusions were arrived at by Sinclair and Little,27 who examined a sample of untreated normal subjects at the ages of 9 to 10, 12 to 13, and 19 to 20 but found no cephalometric variable associated with II. Fudalej and
American Journal of Orthodontics and Dentofacial Orthopedics Volume 134, Number 2
Årtun28 evaluated an association between relapse and 3 facial types (short, average, and long) considered to have different types of mandibular rotation. No correlation was found. Also, Williams and Andersen,29 who examined craniofacial changes 4 years after discontinuation of retention, concluded that mandibular rotation was not correlated with crowding. Nanda and Nanda,30 in the summary of their research regarding facial growth, stated that the onset of puberty is different in children with different growth patterns. A child with a skeletal deep bite has the growth spurt on average 1.5 to 2 years later than one with an open bite. They recommended that deep-bite patients have longer retention periods than skeletal open-bite subjects. Their conclusion was that duration of retention should depend on the patient’s maturation status: ie, anticipated craniofacial growth. Chronologic and dental ages, although easy to determine, are poorly correlated with craniofacial growth.31 Hand-wrist radiographic evaluation is accurate but requires an additional x-ray.31 The cervical vertebral maturation (CVM) method combines ease and accuracy.32 It was proposed as a synthetic way to assess skeletal maturity that is directly correlated with growth potential. The method is based on the morphologic alterations of the cervical vertebrae during growth. The consistent sequence of morphologic changes of the vertebral bodies allows us to distinguish the CVM stages that correspond to a patient’s skeletal maturity. CVM stage 1 implies that the pubertal growth spurt is likely within 2 years after this stage, whereas a person in stage 5 had the pubertal growth spurt approximately 2 years before. Determination of CVM stage at the end of orthodontic treatment permits assessment of skeletal maturity and expected craniofacial growth. Therefore, our aim in this study was to test the hypothesis that relapse of incisor alignment is associated with skeletal maturity assessed with the CVM method. MATERIAL AND METHODS
We used a sample from another study on long-term incisor stability, and a detailed description of the selection process is reported elsewhere.11 The postretention collection in the Department of Orthodontics at the University of Washington was screened, and Little’s II index33 was measured on all models made at least 10 years out of retention (T3). Inclusion criteria were (1) patients or relapse group: II ⬎6 mm; and (2) controls or stable group: II ⬍3.5 mm. Subjects older than 18 at the end of treatment (T2) were excluded.
Fudalej, Rothe, and Bollen 239
Fig. Schematic representation of landmarks identified and traced on the second, third, and fourth cervical vertebral bodies.
For the subjects included in this study, II was measured on the models taken at the initial orthodontic examination (T1) and at T2. Patient information and treatment history for all subjects was obtained from the database. Age at the start of treatment and sex were recorded. Treatment time was determined by subtracting the date of the T1 records from the date of the T2 records. Retention time was estimated by subtracting the date of the T2 records from the patient-reported end-of-retention date. The postretention time was estimated by subtracting the reported end-of-retention date from the T3 record date. II was measured as the sum of the linear displacements of the anatomic contact points of each mandibular incisor from the adjacent tooth anatomic point on all study models by one investigator (L.E.R.). II was remeasured on 165 study models by another investigator (P.F.). Intercanine width was measured as the distance between the cusp tips of the mandibular canines at T1 and T2 on all models. The measurements were made to the nearest 0.01 mm with a digital caliper (Fred V. Fowler, Newton, Mass). Lateral cephalograms taken at T2 were used to determine CVM status.34 On each cephalogram, the second, third, and fourth cervical vertebrae (C2, C3, and C4) were identified. The traced landmarks and performed measurements are shown in the Figure and
240 Fudalej, Rothe, and Bollen
Table I.
American Journal of Orthodontics and Dentofacial Orthopedics August 2008
Landmarks and measurements
Landmark or measurement
Description
C2LP, C2D, C2LA C3UP, C3UA C3LP, C3D, C3LA C4UP, C4UA C4LP, C4D, C4LA C2Conc C3Conc C4Conc C3BAR
The most posterior, the deepest, and the most anterior points on the lower border of the body of C2 The most superior points of the posterior and anterior borders of the body of C3 The most posterior, the deepest, and the most anterior points on the lower border of the body of C3 The most superior points of the posterior and anterior borders of the body of C4; The most posterior, the deepest, and the most anterior points on the lower border of the body of C4 Distance from the line connecting C2LP and C2LA to the deepest point on the lower border of the vertebra, C2D Distance from the line connecting C3LP and C3LA to the deepest point on the lower border of the vertebra, C3D Distance from the line connecting C4LP and C4LA to the deepest point on the lower border of the vertebra, C4D Ratio between the length of the base (distance C3LP-C3LA) and the anterior height (distance C3UA-C3LA) of the body of C3 Ratio between the posterior (distance C3UP-C3LP) and anterior (distance C3UA-C3LA) heights of the body of C3 Ratio between the length of the base (distance C4LP-C4LA) and the anterior height (distance C4UA-C4LA) of the body of C4 Ratio between the posterior (distance C4UP-C4LP) and anterior (distance C4UA-C4LA) heights of the body of C4
C3PAR C4BAR C4PAR
L, Lower; P, posterior; D, deepest; A, anterior; U, upper; Conc, concavity; BAR, base to anterior height ratio; PAR, posterior to anterior heights ratio. Table II.
Classification of CVM stages according to a shape of the body of the second, third, and fourth cervical vertebrae
CVM stage 1 2 3 4 5
Vertebral body shape C3 and C4 flat C3 concavity ⱖ1 mm; C4 flat C2, C3, and C4 concavity ⱖ1 mm; C3 or C4 tapered or horizontal rectangular C3 or C4 square; if C3 or C4 are not square, then horizontal rectangular C3 or C4 vertical rectangular
Flat, Concavity ⬍1 mm; tapered, C3UP-C3LP (or C4UP-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio ⬎1.20; square, C3UP-C3LP (or C4UP-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio (PAR) 0.80 to 1.20, C3LA-C3LP (or C4LA-C4LP) to C3UA-C3LA (or C4UAC4LA) ratio (BAR) 0.85 to 1.15; horizontal rectangular, C3LAC3LP (or C4LA-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio (BAR) ⬎1.15; vertical rectangular, C3LA-C3LP (or C4LA-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio (BAR) ⬍0.85. L, Lower; P, posterior; D, deepest; A, anterior; U, upper; BAR, base to anterior height ratio; PAR, posterior to anterior heights ratio.
Table I. Assignment of a CVM stage to a subject is described in Table II. Statistical analysis
Descriptive statistics (means and standard deviations) were computed for each measurement at T1, T2, and T3. Independent t tests were used to test the differences in measurements between the stable and the relapse groups at T1, T2, and T3. Chi-square tests were used to examine the differences in distributions of sex, extraction treatment alternatives, and CVM stages between the groups.
T tests and Pearson product moment correlation coefficients were calculated to determine the interobserver reproducibility of the II measurements. The association between severe relapse and CVM status was evaluated by calculating the odds ratio (OR) of relapse to stability with logistic regression analysis. Adjustments were made for confounding variables, such as initial and end-of-treatment II, intercanine width change during treatment, treatment alternative, sex, age at T2, and postretention time. The reproducibility of the measurements was assessed by statistically analyzing the difference between double measurements taken 1 week apart on 28 cephalograms selected at random. The error of the method was calculated from the equation: Sx ⫽
冑
⌺ D2 2N
with D representing the difference between the corresponding first and second measurements and N the number of double determinations. The errors for the cephalometric measurements were 0.33, 0.30, and 0.28 mm for C2Conc, C3Conc, and C4Conc, respectively. The errors for calculated ratios were 0.05 for C3PAR, 0.07 for C3BAR and C4BAR, and 0.09 for C4PAR (see Table I for definitions of these terms). The measurement errors were 0.19 mm for intercanine change and 0.38 mm for II. RESULTS
A total of 323 subjects were initially identified (60 patients, 263 controls); 22 were excluded because they were older than 18 at the end of treatment. Thus, there
Fudalej, Rothe, and Bollen 241
American Journal of Orthodontics and Dentofacial Orthopedics Volume 134, Number 2
Table III.
Demographic summary of the groups Stable (n ⫽ 244)
Age at T1 (y) Age at T2 (y) Age at T3 (y) Treatment time (y) Retention time (y) Postretention time (y)
Males Females Angle Class I Angle Class II Angle Class III Extraction treatment plan
Table V.
Stable (n ⫽ 244)
Relapse (n ⫽ 57)
Mean
SD
Mean
SD
P
12.29 15.23 30.44 2.95 2.69 12.56
1.87 1.69 4.33 1.34 2.13 3.77
12.32 15.20 31.73 2.95 2.52 14.05
1.72 1.55 5.11 1.30 1.99 4.96
0.938 0.896 0.052 1.000 0.591 0.013*
n
%
n
%
P
76 168 80 156 8 153
31.1 68.9 32.8 63.9 3.3 62.7
23 34 18 39 0 39
40.4 59.6 31.6 68.4 0 68.4
0.092 0.092 0.985 0.628 0.353 0.510
Interobserver reproducibility for II measure-
ments Investigator 1 (P.F.)
Investigator 2 (L.E.R.)
Time
Mean
SD
Mean
SD
P
R
T1 T2 T3
4.84 1.05 3.25
2.69 0.77 2.29
4.31 1.00 2.90
2.87 0.70 2.21
0.64 0.51 0.79
0.90 0.77 0.97
II was measured on 165 study casts. Differences between the readings were tested by using a t test; additionally, correlation coefficients (R) were calculated.
were 301 subjects: 57 patients (relapse group) and 244 controls (stable group). The cervical vertebrae were traced on 252 subjects (209 stable, 43 relapse subjects); in 49 subjects, either the cephalograms were of inadequate quality or the body of fourth vertebra was blocked by a protective collar on the patient. The measurement of intercanine width change from T1 to T2 was impossible in 40 stable and 8 relapse subjects: 33 stable and 5 relapse subjects had either broken or erupting mandibular canines at T1, and the records of 7 stable and 3 relapse subjects could not be found in the postretention collection. Balanced ratios of sex, Angle Class I and Class II malocclusions, and extraction therapy in both groups are given in Table III. There were 8 Class III subjects in the stable group, but no Class III subjects in the relapse group. Treatment and retention times were similar in both groups. Postretention time was 1.79
Relapse (n ⫽ 57)
Time
Mean
SD
n
Mean
SD
n
P
II T1 II T2 II T3 3-3 T1 3-3 T2
3.81 0.85 1.96 26.03 26.49
3.23 0.58 0.84 1.95 1.63
204 244 244 204 204
5.55 1.49 7.35 24.99 26.23
3.18 0.96 1.18 1.79 1.23
48 57 57 49 49
0.001* ⬍0.001* †
0.004* 0.379
*Statistical significance at the 0.05 level. † T test was not run because case-control status was determined by II at T3. 3-3, Intercanine width.
Table VI. Frequency of CVM stages (%) in groups (chi-square tests determined intergroup differences) CVM stage
T or chi-square tests were used. *Statistical significance at the 0.05 level. Table IV.
Mean II (in mm) measured on dental casts
1 2 3 4 5 Undetermined
Stable group (n ⫽ 244)
Relapse group (n ⫽ 57)
P
3.7 4.1 48.0 25.8 4.1 14.3
3.5 3.5 45.6 19.3 3.5 24.6
0.74 0.86 0.86 0.39 0.86 0.09
years longer in the relapse group, and the difference was statistically significant (P ⫽ 0.013). Evaluation of the reproducibility of the II measurements (Table IV) showed no significant interobserver difference (P ⬎0.05). II measured at T1 and T3 showed high correlation (R ⫽ 0.90 and 0.97, respectively). Interobserver reproducibility of T2 measurements had a weaker correlation (R ⫽ 0.77). However, the paired t test did not detect a significant difference at T2 (P ⫽ 0.51). Table V summarizes the II and intercanine width measurements at all time points. II at T1 was 1.74 mm more in the relapse group, and the difference was significant (P ⫽ 0.001). II at T2 was also higher in the relapse group, and the difference was significant (P ⬍0.001). Intercanine width before treatment was greater in the stable subjects at 1 mm than in the relapse subjects, and the difference was significant (P ⫽ 0.004). At T2, the groups did not differ in intercanine width (P ⫽ 0.379). Intercanine width was expanded during treatment (T1-T2) by 1.24 mm in the relapse group. Expansion of this width in the stable group was 0.46 mm, and the difference between groups was significant (P ⫽ 0.003). There was no difference in the distribution of CVM stages (Table VI) between the stable and the relapse groups (P ⬎0.05). In 86% of the stable and relapse
242 Fudalej, Rothe, and Bollen
American Journal of Orthodontics and Dentofacial Orthopedics August 2008
Table VII.
CVM model evaluating the association between relapse status and cervical vertebral maturation stage at T2
CVM stage at T2 II T1 II T2 3-3 T1 to T2 Treatment alternative Sex Age at T2 Postretention time
OR
95% CI
P
0.961 1.179 0.934 1.253 2.249 1.658 0.855 1.112
0.545-1.697 1.046-1.329 0.548-1.593 0.957-1.642 0.768-6.584 0.602-4.567 0.645-1.135 1.008-1.226
0.89 0.007* 0.80 0.10 0.14 0.33 0.28 0.034*
Logistic regression model adjusted for potential confounding variables: II at T1 and T2, 3-3 change (T1-T2), extraction vs nonextraction treatment, sex, age at T2, and postretention time. OR is the odds ratio of unstable (II ⬎6 mm) to stable (II ⬍3.5 mm) incisor alignment. *Statistical significance at the 0.05 level. 3-3, Intercanine width.
subjects, the growth spurt occurred no later than 2 year before the end of treatment (CVM stages 3 and 4). Only 9% of the subjects in both groups had not started the growth spurt at the end of orthodontic treatment (CVM stages 1 and 2). Relatively few patients completed treatment not later than 2 years after the growth spurt (CVM stage 5). Table VII shows the result of the logistic regression analysis evaluating the association between relapse of incisor alignment and CVM status at T2. An OR greater than 1 implies that a higher value of the predictor is associated with an increased risk of relapse. If the OR is less than 1, a higher value of the predictor is associated with a decreased risk of relapse. The results show that CVM status at T2 is not associated with relapse of incisor alignment (P ⫽ 0.89). Only II before treatment and postretention time were found to be related to the outcome. The OR at T1 suggests that a subject with 1 mm greater II before treatment has 1.18 times the chance of relapse at T3 (P ⫽ 0.007). The OR of 1.11 for postretention time suggests that a person with 1 year longer postretention time has 1.11 times the odds of relapse at T3 (P ⫽ 0.03). Intercanine width change during treatment (T1-T2), II at T2, age at T2 , extraction vs nonextraction treatment, and sex were unrelated to long-term relapse of incisor alignment. DISCUSSION
The effectiveness of dentofacial orthopedics often depends on circumpubertal acceleration of growth. Because maturational changes of the cervical vertebrae are correlated with periods of acceleration and decel-
eration of craniofacial growth, the CVM method was created.32 To facilitate identification of a period of intensive facial growth, 6 CVM stages were originally proposed.34-37 A recent modification limited the number of CVM stages to 5: from stage 1, indicating that the growth peak will occur not earlier than 1 year after this stage, to stage 5, indicating that the growth spurt occurred about 2 years before this stage.34 A growth peak occurs between the second and third stages in 95% of subjects. Although a focus of the CVM method is on prediction of the pubertal growth spurt, the index gives an indirect answer about a patient’s growth potential. The higher the maturational stage, the less future growth. According to Hassel and Farman,35 65% to 85% of adolescent growth is expected during CVM stage 1, as opposed to 5% to 10% of adolescent growth in a person in stage 4. A patient in stage 5 should have little or no craniofacial growth. Our results cannot confirm an association between craniofacial growth as measured with the CVM method and the relapse of incisor alignment. Logistic regression analysis failed to detect any effect of CVM status at T2 on relapse even after controlling for possible confounding variables. Orthodontic treatment in most subjects (⬎90%) from both groups was finished after the adolescent growth spurt (CVM stages 3-5). Comparison of the frequency of CVM stages at T2 between the groups with chi-square tests showed no intergroup difference. These results should, however, be interpreted with caution, since the CVM method might not be ideal in detecting postadolescent craniofacial growth. The data on mandibular length changes (CoGn) by Baccetti et al34 demonstrate that, during a growth peak (CVM stages 2 to 3), the mandible elongates by about 6 mm, as opposed to 2 mm later between CVM stages 4 and 5. One can infer that, once a subject attains CVM stage 5, little mandibular growth can be expected. If these data are contrasted with data from the same growth study and presented by Riolo et al,38 it is clear that mandibular and possibly craniofacial growth continues in subjects classified as in CVM stage 5. For exmple, Baccetti et al34 reported that mandibular length (Co-Gn) in CVM stage 5 subjects (mean age, 13.8 years; SD, 1.2) was 121.77 mm (SD, 5.93), whereas Riolo et al38 demonstrated that 16-yearold subjects (also in CVM stage 5) had a Co-Gn distance about 6 mm longer. This implies that the CVM method might not be sensitive enough to distinguish between subjects with considerable postadolescent craniofacial growth and those with little growth left, but all classified as CVM stage 5. A large II before treatment was found to predict the relapse of incisor alignment. An odds ratio of 1.18
Fudalej, Rothe, and Bollen 243
American Journal of Orthodontics and Dentofacial Orthopedics Volume 134, Number 2
implies that 1 mm of II before treatment increases the chance of relapse by 18%. This finding is consistent with results of Årtun et al2 and Kahl-Nieke and Schwarze,39 who reported that a large amount of pretreatment mandibular incisor crowding is directly correlated with relapse. Ormiston et al12 used primarily the peer assessment rating (PAR) index to distinguish stable from relapse subjects in their retrospective casecontrol study and demonstrated an association between II at T1 and long-term relapse. Their OR (1.14) is similar to the value found in our study. Postretention time was found to be associated with instability of the mandibular incisors. The OR of 1.11 suggests that, for each additional year out of retention, the chance of relapse increases by 11%. This finding agrees with the results of other studies.16,17,27,40,41 Sinclair and Little28 examined 65 untreated subjects in mixed dentition (9-10 years of age), early permanent dentition (12-13 years), and early adulthood (19-20 years) and found an increase in mandibular II of 0.7 mm during the postadolescent years (13-20 years). Buschang and Shulman16 analyzed cross-sectional data of 9044 untreated subjects from the Third National Health and Nutrition Examination Survey and concluded that, although the greatest increase of crowding occurs during early adulthood (20-25 years), incisor irregularity increases with time. Driscoll-Gilliland et al17 investigated the association between growth and stability in treated and untreated subjects from 14.5 to 25 years of age. Even though the II was less in the treated group at the start of observation, both groups showed similar—1 mm—increases in incisor crowding. Little et al13 studied changes from 10 to 20 years postretention and concluded that a trend toward increased II continues in the second decade after retention, but the rate of change diminishes. Other researchers who examined untreated subjects arrived at opposite conclusions.41,42 Richardson and Gormley42 examined changes from 18 to 28 years of age and found few dental changes. Bondevik41 reported similar findings in his slightly older sample (23-34 years). The possible explanation for the difference is that, instead of II, “arch space condition” was used to measure changes in the mandibular arch. Sex was not found to be associated with the long-term stability of incisor alignment. This finding agrees with the results of Rothe et al.11 However, the sample selected for our study was based on that sample. Therefore, it was expected that some findings would be shared by both investigations. The results of other studies conflict with ours. Ormiston et al12 studied the University of Washington sample of 86 stable and unstable subjects and found males to be 4.4 times more
prone to relapse than females. The smaller number of subjects in their sample might explain the difference. Bishara et al26 examined dentofacial changes between 26 and 45 years of age and found that anterior “toothsize-arch-length-deficiency” increased more in men than women. Small sample size (15 men,15 women) makes that result questionable. Also, Buschang and Shulman16 concluded that males have greater II (3.92 mm) than females (3.40 mm). The large number of subjects (9044) in their study increased the power of the study so that subtle intersex differences could be detected. On the other hand, Sinclair and Little,27 who investigated long-term changes in untreated persons with normal occlusions, demonstrated that II increased more in females than in males. The difference equaled 0.30 mm and was likely clinically insignificant. This investigation had some limitations that might affect the results. Treatment notes were unavailable to the investigators, so the subjects in both groups might have differed in the details of therapy. These potential confounding variables were not controlled for. Circumferential supracrestal fiberotomy, believed to enhance stability, might have been performed in some subjects. However, this procedure was rare at the time of collection of the sample from the University of Washington. The case-control design of this study involved selection of 2 groups that differed in postretention II index: a stable group with II ⬍3.5 mm and a relapse group with II ⬎6 mm. The subjects with postretention II ⱖ3.5 mm and ⱕ6 mm were eliminated from the sample. Therefore, this design enabled us to assess the odds of relapse vs stable dentition, not just the odds of relapse. CONCLUSIONS
These results demonstrate no association between skeletal maturation at the end of treatment assessed with the CVM method and long-term stability of incisor alignment. Pretreatment II and postretention time were found to be correlated with increased relapse. Sex was not related to incisor relapse. REFERENCES 1. Årtun J, Krogstad O, Little RM. Stability of mandibular incisors following excessive proclination: a study in adults with surgically treated mandibular prognathism. Angle Orthod 1990;60:99106. 2. Årtun J, Garol JD, Little RM. Long-term stability of mandibular incisors following successful treatment of Angle Class II, Division 1 malocclusions. Angle Orthod 1996;66:229-38. 3. Edwards JG. A long-term prospective evaluation of the circumferential supracrestal fiberotomy in alleviating orthodontic relapse. Am J Orthod Dentofacial Orthop 1988;93:380-7.
244 Fudalej, Rothe, and Bollen
4. Glenn G, Sinclair PM, Alexander RG. Nonextraction orthodontic therapy: posttreatment dental and skeletal stability. Am J Orthod Dentofacial Orthop 1987;92:321-8. 5. Kahl-Nieke B, Fischbach H, Schwarze CW. Post-retention crowding and incisor irregularity: a long-term follow-up evaluation of stability and relapse. Br J Orthod 1995;22:249-57. 6. Little RM, Wallen TR, Riedel RA. Stability and relapse of mandibular anterior alignment-first premolar extraction cases treated by traditional edgewise orthodontics. Am J Orthod 1981;80:349-65. 7. Little RM, Riedel RA. Postretention evaluation of stability and relapse—mandibular arches with generalized spacing. Am J Orthod Dentofacial Orthop 1989;95:37-41. 8. Luppanapornlarp S, Johnston LE Jr. The effects of premolarextraction: a long-term comparison of outcomes in “clear-cut” extraction and nonextraction Class II patients. Angle Orthod 1993;63:257-72. 9. McReynolds DC, Little RM. Mandibular second premolar extraction postretention evaluation of stability and relapse. Angle Orthod 1990;61:133-44. 10. Paquette DE, Beattie JR, Johnston LE. A long-term comparison of nonextraction and premolar extraction edgewise therapy in “borderline” Class II patients. Am J Orthod Dentofacial Orthop 1992;102:1-14. 11. Rothe LE, Bollen AM, Little RM, Herring SW, Chaison JB, Chen CSK, et al. Trabecular and cortical bone as risk factors for orthodontic relapse. Am J Orthod Dentofacial Orthop 2006;130: 476-84. 12. Ormiston JP, Huang GJ, Little RM, Decker JD, Seuke GD. Retrospective analysis of long-term stable and unstable orthodontic treatment outcomes. Am J Orthod Dentofacial Orthop 2005;128:568-74. 13. Little RM, Riedel RA, Årtun J. An evaluation of changes in mandibular anterior alignment from 10 to 20 years postretention. Am J Orthod Dentofacial Orthop 1988;93:423-8. 14. Blake M, Bibby K. Retention and stability: a review of the literature. Am J Orthod Dentofacial Orthop 1998;114:299-306. 15. Shah AA. Postretention changes in mandibular crowding: a review of the literature. Am J Orthod Dentofacial Orthop 2003;124:298-308. 16. Buschang PH, Shulman JD. Incisor crowding in untreated persons 15-50 years of age: United States, 1988-1994. Angle Orthod 2003;73:502-8. 17. Driscoll-Gilliland J, Buschang PH, Behrents RG. An evaluation of growth and stability in untreated and treated subjects. Am J Orthod Dentofacial Orthop 2001;120:588-97. 18. Richardson ME. Late lower arch crowding: the role of facial morphology. Angle Orthod 1986;56:244-54. 19. Perera PSG. Rotational growth and incisor compensation. Angle Orthod 1987;57:39-49. 20. Issacson R, Zapfel R, Worms F, Erdmand A. Effects of rotational jaw growth on the occlusion and profile. Am J Orthod 1977;72: 276-86. 21. Behrents RG. A treatise on the continuum of growth in the aging craniofacial skeleton [thesis]. Ann Arbor: University of Michigan; 1984.
American Journal of Orthodontics and Dentofacial Orthopedics August 2008
22. Formby WA, Nanda RS, Currier GF. Longitudinal changes in the adult facial profile. Am J Orthod Dentofacial Orthop 1994;105: 464-76. 23. Bondevik O. Growth changes in the cranial base and the face: a longitudinal cephalometric study of linear and angular changes in adult Norwegians. Eur J Orthod 1995;17:525-32. 24. Foley TF, Mamandras AH. Facial growth in females 14 to 20 years of age. Am J Orthod Dentofacial Orthop 1992;101:248-54. 25. Pollard LE, Mamandras AH. Male postpubertal facial growth in Class II malocclusions. Am J Orthod Dentofacial Orthop 1995; 108:62-8. 26. Bishara SE, Treder JE, Jakobsen JR. Facial and dental changes in adulthood. Am J Orthod Dentofacial Orthop 1994;106:175-86. 27. Sinclair PM, Little RM. Maturation of untreated normal occlusion. Am J Orthod 1983;83:114-23. 28. Fudalej P, Årtun J. Mandibular growth rotation effects on postretention stability of mandibular incisor alignment. Angle Orthod 2007;77:199-205. 29. Williams S, Andersen CE. Incisor stability in patients with anterior rotational mandibular growth. Angle Orthod 1995;65: 431-42. 30. Nanda RS, Nanda SK. Considerations of dentofacial growth in long-term retention and stability: is active retention needed? Am J Orthod Dentofacial Orthop 1992;101:297-302. 31. Hägg U, Taranger J. Maturation indicators and the pubertal growth spurt. Am J Orthod 1982;82:299-309. 32. Lamparski D. Skeletal age assessment utilizing cervical vertebrae [thesis]. Pittsburgh: University of Pittsburgh; 1972. 33. Little RM. The irregularity index: a quantitative score of mandibular anterior alignment. Am J Orthod 1975;68:554-63. 34. Baccetti T, Franchi L, McNamara JA Jr. An improved version of the cervical vertebral maturation (CVM) method for the assessment of mandibular growth. Angle Orthod 2002;72:316-23. 35. Hassel B, Farman AG. Skeletal maturation evaluation using cervical vertebrae. Am J Orthod Dentofacial Orthop 1995;107: 58-66. 36. Franchi L, Baccetti T, McNamara JA Jr. Mandibular growth as related to cervical vertebral maturation and body height. Am J Orthod Dentofacial Orthop 2000;118:335-40. 37. San Roman P, Palma JC, Oteo MD, Nevado E. Skeletal maturation determined by cervical vertebrae development. Eur J Orthod 2002;24:303-11. 38. Riolo ML, Moyers RE, McNamara JA, Hunter WS. An atlas of craniofacial growth: cephalometric standards from the University School Growth Study. Monograph 2. Craniofacial Growth Series. Ann Arbor: Center for Human Growth and Development; University of Michigan; 1974. 39. Kahl-Nieke B, Schwarze CW. Post-retention crowding and incisor irregularity: a long-term follow-up evaluation of stability and relapse. Br J Orthod 1995;22:249-57. 40. Little RM. Stability and relapse of dental arch alignment. Br J Orthod 1990;17:235-41. 41. Bondevik O. Changes in occlusion between 23 and 34 years of age. Angle Orthod 1998;68:75-80. 42. Richardson ME, Gormley JS. Lower arch crowding in the third decade. Eur J Orthod 1998;20:597-607.
Effects of posttreatment skeletal maturity measured with the cervical vertebral maturation method on incisor alignment relapse Piotr Fudalej,a Laura E. Rothe,b and Anne-Marie Bollenc Warsaw, Poland, Spokane, and Seattle, Wash Introduction: Our aim was to test the hypothesis that relapse of incisor alignment is associated with skeletal maturity at the end of treatment, as assessed with the cervical vertebral maturation (CVM) method. Methods: This was a case-control study with information from the postretention database at the University of Washington. Mandibular incisor irregularity (II) at least 10 years out of retention (T3) was used to define the subjects (II ⬎6 mm, relapse group) and the controls (II ⬍3.5 mm, stable group). The following model measurements were made: II at pretreatment (T1), II at posttreatment (T2), and intercanine width at T1 and T2. On cephalograms taken T2, the CVM status was determined. Logistic regression analyses were used to determine the association between relapse and CVM status after treatment. The models were adjusted for potentially confounding variables (II at pretreatment and posttreatment, intercanine width change during treatment, sex, age at T2, and treatment alternatives). Results: No association between CVM stage at T2 and relapse was found (P ⫽ 0.89). Both groups had similar distributions of the CVM stages (P ⬎0.05). Pretreatment II and postretention time were found to be correlated with long-term incisor stability (P ⫽ 0.007 and 0.034, respectively). Sex was not related to relapse (P ⫽ 0.33). Conclusions: Maturity of craniofacial structures at the end of treatment evaluated with the CVM method is not associated with long-term stability of incisor alignment. (Am J Orthod Dentofacial Orthop 2008;134:238-44)
L
ong-term stability of orthodontically aligned teeth frequently poses a serious challenge. Many studies1-12 have demonstrated that relapse, particularly of mandibular incisor alignment, affects more than half of the patients 10 years after discontinuation of retention.6 Changes from 10 to 20 years out of retention further increase incisor irregularity (II) so that less than 30% of patients have satisfactory dental alignment 20 years after discontinuing retention.13 Among the potential causes of incisor relapse discussed in recent literature reviews, the growth of dentofacial structures merits attention.14,15 When a typical orthodontic treatment is completed, variable skeletal maturity of craniofacial structures remains. Since II, in both orthodontically treated and untreated subjects, has the highest rate of increase during the late a
Assistant professor, Center for Craniofacial Disorders, National Research Institute for Mother and Child, Warsaw, Poland. b Private practice, Spokane, Wash. c Professor, Department of Orthodontics; University of Washington, Seattle. Reprint requests to: Piotr Fudalej, Center for Craniofacial Disorders, National Research Institute for Mother and Child, Kasprzaka Str. 17a, Warsaw, Poland; e-mail, [email protected]. Submitted, July 2006; revised and accepted, September 2006. 0889-5406/$34.00 Copyright © 2008 by the American Association of Orthodontists. doi:10.1016/j.ajodo.2006.09.060
238
teens and early twenties,13,16 circumpubertal growth is suspected to be related to crowding.17-20 A finding that craniofacial growth continues, although at a limited rate, after maturity aroused additional interest in the search for an association between craniofacial growth and long-term stability of incisor alignment.21-26 Driscoll-Gilliland et al17 followed a sample of treated and untreated subjects from 14 to 29 years of age and found a similar increase of incisor irregularity in both groups. They demonstrated that the development of crowding was correlated with vertical facial growth and eruption of the incisors. Richardson18 examined skeletal and dental changes from 13 to 18 years of age and concluded that mandibular growth rotation contributed to the development of crowding. Perera19 investigated craniofacial changes in untreated subjects from 11 to 20 years of age and found that mandibular rotation and mandibular incisor crowding were associated, and the correlation coefficient equaled 0.51. Ormiston et al12 compared 45 subjects with stable incisor alignment and 41 subjects with unstable incisor alignment and found that both male sex and sustained period of growth were associated with increased instability. Opposite conclusions were arrived at by Sinclair and Little,27 who examined a sample of untreated normal subjects at the ages of 9 to 10, 12 to 13, and 19 to 20 but found no cephalometric variable associated with II. Fudalej and
American Journal of Orthodontics and Dentofacial Orthopedics Volume 134, Number 2
Årtun28 evaluated an association between relapse and 3 facial types (short, average, and long) considered to have different types of mandibular rotation. No correlation was found. Also, Williams and Andersen,29 who examined craniofacial changes 4 years after discontinuation of retention, concluded that mandibular rotation was not correlated with crowding. Nanda and Nanda,30 in the summary of their research regarding facial growth, stated that the onset of puberty is different in children with different growth patterns. A child with a skeletal deep bite has the growth spurt on average 1.5 to 2 years later than one with an open bite. They recommended that deep-bite patients have longer retention periods than skeletal open-bite subjects. Their conclusion was that duration of retention should depend on the patient’s maturation status: ie, anticipated craniofacial growth. Chronologic and dental ages, although easy to determine, are poorly correlated with craniofacial growth.31 Hand-wrist radiographic evaluation is accurate but requires an additional x-ray.31 The cervical vertebral maturation (CVM) method combines ease and accuracy.32 It was proposed as a synthetic way to assess skeletal maturity that is directly correlated with growth potential. The method is based on the morphologic alterations of the cervical vertebrae during growth. The consistent sequence of morphologic changes of the vertebral bodies allows us to distinguish the CVM stages that correspond to a patient’s skeletal maturity. CVM stage 1 implies that the pubertal growth spurt is likely within 2 years after this stage, whereas a person in stage 5 had the pubertal growth spurt approximately 2 years before. Determination of CVM stage at the end of orthodontic treatment permits assessment of skeletal maturity and expected craniofacial growth. Therefore, our aim in this study was to test the hypothesis that relapse of incisor alignment is associated with skeletal maturity assessed with the CVM method. MATERIAL AND METHODS
We used a sample from another study on long-term incisor stability, and a detailed description of the selection process is reported elsewhere.11 The postretention collection in the Department of Orthodontics at the University of Washington was screened, and Little’s II index33 was measured on all models made at least 10 years out of retention (T3). Inclusion criteria were (1) patients or relapse group: II ⬎6 mm; and (2) controls or stable group: II ⬍3.5 mm. Subjects older than 18 at the end of treatment (T2) were excluded.
Fudalej, Rothe, and Bollen 239
Fig. Schematic representation of landmarks identified and traced on the second, third, and fourth cervical vertebral bodies.
For the subjects included in this study, II was measured on the models taken at the initial orthodontic examination (T1) and at T2. Patient information and treatment history for all subjects was obtained from the database. Age at the start of treatment and sex were recorded. Treatment time was determined by subtracting the date of the T1 records from the date of the T2 records. Retention time was estimated by subtracting the date of the T2 records from the patient-reported end-of-retention date. The postretention time was estimated by subtracting the reported end-of-retention date from the T3 record date. II was measured as the sum of the linear displacements of the anatomic contact points of each mandibular incisor from the adjacent tooth anatomic point on all study models by one investigator (L.E.R.). II was remeasured on 165 study models by another investigator (P.F.). Intercanine width was measured as the distance between the cusp tips of the mandibular canines at T1 and T2 on all models. The measurements were made to the nearest 0.01 mm with a digital caliper (Fred V. Fowler, Newton, Mass). Lateral cephalograms taken at T2 were used to determine CVM status.34 On each cephalogram, the second, third, and fourth cervical vertebrae (C2, C3, and C4) were identified. The traced landmarks and performed measurements are shown in the Figure and
240 Fudalej, Rothe, and Bollen
Table I.
American Journal of Orthodontics and Dentofacial Orthopedics August 2008
Landmarks and measurements
Landmark or measurement
Description
C2LP, C2D, C2LA C3UP, C3UA C3LP, C3D, C3LA C4UP, C4UA C4LP, C4D, C4LA C2Conc C3Conc C4Conc C3BAR
The most posterior, the deepest, and the most anterior points on the lower border of the body of C2 The most superior points of the posterior and anterior borders of the body of C3 The most posterior, the deepest, and the most anterior points on the lower border of the body of C3 The most superior points of the posterior and anterior borders of the body of C4; The most posterior, the deepest, and the most anterior points on the lower border of the body of C4 Distance from the line connecting C2LP and C2LA to the deepest point on the lower border of the vertebra, C2D Distance from the line connecting C3LP and C3LA to the deepest point on the lower border of the vertebra, C3D Distance from the line connecting C4LP and C4LA to the deepest point on the lower border of the vertebra, C4D Ratio between the length of the base (distance C3LP-C3LA) and the anterior height (distance C3UA-C3LA) of the body of C3 Ratio between the posterior (distance C3UP-C3LP) and anterior (distance C3UA-C3LA) heights of the body of C3 Ratio between the length of the base (distance C4LP-C4LA) and the anterior height (distance C4UA-C4LA) of the body of C4 Ratio between the posterior (distance C4UP-C4LP) and anterior (distance C4UA-C4LA) heights of the body of C4
C3PAR C4BAR C4PAR
L, Lower; P, posterior; D, deepest; A, anterior; U, upper; Conc, concavity; BAR, base to anterior height ratio; PAR, posterior to anterior heights ratio. Table II.
Classification of CVM stages according to a shape of the body of the second, third, and fourth cervical vertebrae
CVM stage 1 2 3 4 5
Vertebral body shape C3 and C4 flat C3 concavity ⱖ1 mm; C4 flat C2, C3, and C4 concavity ⱖ1 mm; C3 or C4 tapered or horizontal rectangular C3 or C4 square; if C3 or C4 are not square, then horizontal rectangular C3 or C4 vertical rectangular
Flat, Concavity ⬍1 mm; tapered, C3UP-C3LP (or C4UP-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio ⬎1.20; square, C3UP-C3LP (or C4UP-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio (PAR) 0.80 to 1.20, C3LA-C3LP (or C4LA-C4LP) to C3UA-C3LA (or C4UAC4LA) ratio (BAR) 0.85 to 1.15; horizontal rectangular, C3LAC3LP (or C4LA-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio (BAR) ⬎1.15; vertical rectangular, C3LA-C3LP (or C4LA-C4LP) to C3UA-C3LA (or C4UA-C4LA) ratio (BAR) ⬍0.85. L, Lower; P, posterior; D, deepest; A, anterior; U, upper; BAR, base to anterior height ratio; PAR, posterior to anterior heights ratio.
Table I. Assignment of a CVM stage to a subject is described in Table II. Statistical analysis
Descriptive statistics (means and standard deviations) were computed for each measurement at T1, T2, and T3. Independent t tests were used to test the differences in measurements between the stable and the relapse groups at T1, T2, and T3. Chi-square tests were used to examine the differences in distributions of sex, extraction treatment alternatives, and CVM stages between the groups.
T tests and Pearson product moment correlation coefficients were calculated to determine the interobserver reproducibility of the II measurements. The association between severe relapse and CVM status was evaluated by calculating the odds ratio (OR) of relapse to stability with logistic regression analysis. Adjustments were made for confounding variables, such as initial and end-of-treatment II, intercanine width change during treatment, treatment alternative, sex, age at T2, and postretention time. The reproducibility of the measurements was assessed by statistically analyzing the difference between double measurements taken 1 week apart on 28 cephalograms selected at random. The error of the method was calculated from the equation: Sx ⫽
冑
⌺ D2 2N
with D representing the difference between the corresponding first and second measurements and N the number of double determinations. The errors for the cephalometric measurements were 0.33, 0.30, and 0.28 mm for C2Conc, C3Conc, and C4Conc, respectively. The errors for calculated ratios were 0.05 for C3PAR, 0.07 for C3BAR and C4BAR, and 0.09 for C4PAR (see Table I for definitions of these terms). The measurement errors were 0.19 mm for intercanine change and 0.38 mm for II. RESULTS
A total of 323 subjects were initially identified (60 patients, 263 controls); 22 were excluded because they were older than 18 at the end of treatment. Thus, there
Fudalej, Rothe, and Bollen 241
American Journal of Orthodontics and Dentofacial Orthopedics Volume 134, Number 2
Table III.
Demographic summary of the groups Stable (n ⫽ 244)
Age at T1 (y) Age at T2 (y) Age at T3 (y) Treatment time (y) Retention time (y) Postretention time (y)
Males Females Angle Class I Angle Class II Angle Class III Extraction treatment plan
Table V.
Stable (n ⫽ 244)
Relapse (n ⫽ 57)
Mean
SD
Mean
SD
P
12.29 15.23 30.44 2.95 2.69 12.56
1.87 1.69 4.33 1.34 2.13 3.77
12.32 15.20 31.73 2.95 2.52 14.05
1.72 1.55 5.11 1.30 1.99 4.96
0.938 0.896 0.052 1.000 0.591 0.013*
n
%
n
%
P
76 168 80 156 8 153
31.1 68.9 32.8 63.9 3.3 62.7
23 34 18 39 0 39
40.4 59.6 31.6 68.4 0 68.4
0.092 0.092 0.985 0.628 0.353 0.510
Interobserver reproducibility for II measure-
ments Investigator 1 (P.F.)
Investigator 2 (L.E.R.)
Time
Mean
SD
Mean
SD
P
R
T1 T2 T3
4.84 1.05 3.25
2.69 0.77 2.29
4.31 1.00 2.90
2.87 0.70 2.21
0.64 0.51 0.79
0.90 0.77 0.97
II was measured on 165 study casts. Differences between the readings were tested by using a t test; additionally, correlation coefficients (R) were calculated.
were 301 subjects: 57 patients (relapse group) and 244 controls (stable group). The cervical vertebrae were traced on 252 subjects (209 stable, 43 relapse subjects); in 49 subjects, either the cephalograms were of inadequate quality or the body of fourth vertebra was blocked by a protective collar on the patient. The measurement of intercanine width change from T1 to T2 was impossible in 40 stable and 8 relapse subjects: 33 stable and 5 relapse subjects had either broken or erupting mandibular canines at T1, and the records of 7 stable and 3 relapse subjects could not be found in the postretention collection. Balanced ratios of sex, Angle Class I and Class II malocclusions, and extraction therapy in both groups are given in Table III. There were 8 Class III subjects in the stable group, but no Class III subjects in the relapse group. Treatment and retention times were similar in both groups. Postretention time was 1.79
Relapse (n ⫽ 57)
Time
Mean
SD
n
Mean
SD
n
P
II T1 II T2 II T3 3-3 T1 3-3 T2
3.81 0.85 1.96 26.03 26.49
3.23 0.58 0.84 1.95 1.63
204 244 244 204 204
5.55 1.49 7.35 24.99 26.23
3.18 0.96 1.18 1.79 1.23
48 57 57 49 49
0.001* ⬍0.001* †
0.004* 0.379
*Statistical significance at the 0.05 level. † T test was not run because case-control status was determined by II at T3. 3-3, Intercanine width.
Table VI. Frequency of CVM stages (%) in groups (chi-square tests determined intergroup differences) CVM stage
T or chi-square tests were used. *Statistical significance at the 0.05 level. Table IV.
Mean II (in mm) measured on dental casts
1 2 3 4 5 Undetermined
Stable group (n ⫽ 244)
Relapse group (n ⫽ 57)
P
3.7 4.1 48.0 25.8 4.1 14.3
3.5 3.5 45.6 19.3 3.5 24.6
0.74 0.86 0.86 0.39 0.86 0.09
years longer in the relapse group, and the difference was statistically significant (P ⫽ 0.013). Evaluation of the reproducibility of the II measurements (Table IV) showed no significant interobserver difference (P ⬎0.05). II measured at T1 and T3 showed high correlation (R ⫽ 0.90 and 0.97, respectively). Interobserver reproducibility of T2 measurements had a weaker correlation (R ⫽ 0.77). However, the paired t test did not detect a significant difference at T2 (P ⫽ 0.51). Table V summarizes the II and intercanine width measurements at all time points. II at T1 was 1.74 mm more in the relapse group, and the difference was significant (P ⫽ 0.001). II at T2 was also higher in the relapse group, and the difference was significant (P ⬍0.001). Intercanine width before treatment was greater in the stable subjects at 1 mm than in the relapse subjects, and the difference was significant (P ⫽ 0.004). At T2, the groups did not differ in intercanine width (P ⫽ 0.379). Intercanine width was expanded during treatment (T1-T2) by 1.24 mm in the relapse group. Expansion of this width in the stable group was 0.46 mm, and the difference between groups was significant (P ⫽ 0.003). There was no difference in the distribution of CVM stages (Table VI) between the stable and the relapse groups (P ⬎0.05). In 86% of the stable and relapse
242 Fudalej, Rothe, and Bollen
American Journal of Orthodontics and Dentofacial Orthopedics August 2008
Table VII.
CVM model evaluating the association between relapse status and cervical vertebral maturation stage at T2
CVM stage at T2 II T1 II T2 3-3 T1 to T2 Treatment alternative Sex Age at T2 Postretention time
OR
95% CI
P
0.961 1.179 0.934 1.253 2.249 1.658 0.855 1.112
0.545-1.697 1.046-1.329 0.548-1.593 0.957-1.642 0.768-6.584 0.602-4.567 0.645-1.135 1.008-1.226
0.89 0.007* 0.80 0.10 0.14 0.33 0.28 0.034*
Logistic regression model adjusted for potential confounding variables: II at T1 and T2, 3-3 change (T1-T2), extraction vs nonextraction treatment, sex, age at T2, and postretention time. OR is the odds ratio of unstable (II ⬎6 mm) to stable (II ⬍3.5 mm) incisor alignment. *Statistical significance at the 0.05 level. 3-3, Intercanine width.
subjects, the growth spurt occurred no later than 2 year before the end of treatment (CVM stages 3 and 4). Only 9% of the subjects in both groups had not started the growth spurt at the end of orthodontic treatment (CVM stages 1 and 2). Relatively few patients completed treatment not later than 2 years after the growth spurt (CVM stage 5). Table VII shows the result of the logistic regression analysis evaluating the association between relapse of incisor alignment and CVM status at T2. An OR greater than 1 implies that a higher value of the predictor is associated with an increased risk of relapse. If the OR is less than 1, a higher value of the predictor is associated with a decreased risk of relapse. The results show that CVM status at T2 is not associated with relapse of incisor alignment (P ⫽ 0.89). Only II before treatment and postretention time were found to be related to the outcome. The OR at T1 suggests that a subject with 1 mm greater II before treatment has 1.18 times the chance of relapse at T3 (P ⫽ 0.007). The OR of 1.11 for postretention time suggests that a person with 1 year longer postretention time has 1.11 times the odds of relapse at T3 (P ⫽ 0.03). Intercanine width change during treatment (T1-T2), II at T2, age at T2 , extraction vs nonextraction treatment, and sex were unrelated to long-term relapse of incisor alignment. DISCUSSION
The effectiveness of dentofacial orthopedics often depends on circumpubertal acceleration of growth. Because maturational changes of the cervical vertebrae are correlated with periods of acceleration and decel-
eration of craniofacial growth, the CVM method was created.32 To facilitate identification of a period of intensive facial growth, 6 CVM stages were originally proposed.34-37 A recent modification limited the number of CVM stages to 5: from stage 1, indicating that the growth peak will occur not earlier than 1 year after this stage, to stage 5, indicating that the growth spurt occurred about 2 years before this stage.34 A growth peak occurs between the second and third stages in 95% of subjects. Although a focus of the CVM method is on prediction of the pubertal growth spurt, the index gives an indirect answer about a patient’s growth potential. The higher the maturational stage, the less future growth. According to Hassel and Farman,35 65% to 85% of adolescent growth is expected during CVM stage 1, as opposed to 5% to 10% of adolescent growth in a person in stage 4. A patient in stage 5 should have little or no craniofacial growth. Our results cannot confirm an association between craniofacial growth as measured with the CVM method and the relapse of incisor alignment. Logistic regression analysis failed to detect any effect of CVM status at T2 on relapse even after controlling for possible confounding variables. Orthodontic treatment in most subjects (⬎90%) from both groups was finished after the adolescent growth spurt (CVM stages 3-5). Comparison of the frequency of CVM stages at T2 between the groups with chi-square tests showed no intergroup difference. These results should, however, be interpreted with caution, since the CVM method might not be ideal in detecting postadolescent craniofacial growth. The data on mandibular length changes (CoGn) by Baccetti et al34 demonstrate that, during a growth peak (CVM stages 2 to 3), the mandible elongates by about 6 mm, as opposed to 2 mm later between CVM stages 4 and 5. One can infer that, once a subject attains CVM stage 5, little mandibular growth can be expected. If these data are contrasted with data from the same growth study and presented by Riolo et al,38 it is clear that mandibular and possibly craniofacial growth continues in subjects classified as in CVM stage 5. For exmple, Baccetti et al34 reported that mandibular length (Co-Gn) in CVM stage 5 subjects (mean age, 13.8 years; SD, 1.2) was 121.77 mm (SD, 5.93), whereas Riolo et al38 demonstrated that 16-yearold subjects (also in CVM stage 5) had a Co-Gn distance about 6 mm longer. This implies that the CVM method might not be sensitive enough to distinguish between subjects with considerable postadolescent craniofacial growth and those with little growth left, but all classified as CVM stage 5. A large II before treatment was found to predict the relapse of incisor alignment. An odds ratio of 1.18
Fudalej, Rothe, and Bollen 243
American Journal of Orthodontics and Dentofacial Orthopedics Volume 134, Number 2
implies that 1 mm of II before treatment increases the chance of relapse by 18%. This finding is consistent with results of Årtun et al2 and Kahl-Nieke and Schwarze,39 who reported that a large amount of pretreatment mandibular incisor crowding is directly correlated with relapse. Ormiston et al12 used primarily the peer assessment rating (PAR) index to distinguish stable from relapse subjects in their retrospective casecontrol study and demonstrated an association between II at T1 and long-term relapse. Their OR (1.14) is similar to the value found in our study. Postretention time was found to be associated with instability of the mandibular incisors. The OR of 1.11 suggests that, for each additional year out of retention, the chance of relapse increases by 11%. This finding agrees with the results of other studies.16,17,27,40,41 Sinclair and Little28 examined 65 untreated subjects in mixed dentition (9-10 years of age), early permanent dentition (12-13 years), and early adulthood (19-20 years) and found an increase in mandibular II of 0.7 mm during the postadolescent years (13-20 years). Buschang and Shulman16 analyzed cross-sectional data of 9044 untreated subjects from the Third National Health and Nutrition Examination Survey and concluded that, although the greatest increase of crowding occurs during early adulthood (20-25 years), incisor irregularity increases with time. Driscoll-Gilliland et al17 investigated the association between growth and stability in treated and untreated subjects from 14.5 to 25 years of age. Even though the II was less in the treated group at the start of observation, both groups showed similar—1 mm—increases in incisor crowding. Little et al13 studied changes from 10 to 20 years postretention and concluded that a trend toward increased II continues in the second decade after retention, but the rate of change diminishes. Other researchers who examined untreated subjects arrived at opposite conclusions.41,42 Richardson and Gormley42 examined changes from 18 to 28 years of age and found few dental changes. Bondevik41 reported similar findings in his slightly older sample (23-34 years). The possible explanation for the difference is that, instead of II, “arch space condition” was used to measure changes in the mandibular arch. Sex was not found to be associated with the long-term stability of incisor alignment. This finding agrees with the results of Rothe et al.11 However, the sample selected for our study was based on that sample. Therefore, it was expected that some findings would be shared by both investigations. The results of other studies conflict with ours. Ormiston et al12 studied the University of Washington sample of 86 stable and unstable subjects and found males to be 4.4 times more
prone to relapse than females. The smaller number of subjects in their sample might explain the difference. Bishara et al26 examined dentofacial changes between 26 and 45 years of age and found that anterior “toothsize-arch-length-deficiency” increased more in men than women. Small sample size (15 men,15 women) makes that result questionable. Also, Buschang and Shulman16 concluded that males have greater II (3.92 mm) than females (3.40 mm). The large number of subjects (9044) in their study increased the power of the study so that subtle intersex differences could be detected. On the other hand, Sinclair and Little,27 who investigated long-term changes in untreated persons with normal occlusions, demonstrated that II increased more in females than in males. The difference equaled 0.30 mm and was likely clinically insignificant. This investigation had some limitations that might affect the results. Treatment notes were unavailable to the investigators, so the subjects in both groups might have differed in the details of therapy. These potential confounding variables were not controlled for. Circumferential supracrestal fiberotomy, believed to enhance stability, might have been performed in some subjects. However, this procedure was rare at the time of collection of the sample from the University of Washington. The case-control design of this study involved selection of 2 groups that differed in postretention II index: a stable group with II ⬍3.5 mm and a relapse group with II ⬎6 mm. The subjects with postretention II ⱖ3.5 mm and ⱕ6 mm were eliminated from the sample. Therefore, this design enabled us to assess the odds of relapse vs stable dentition, not just the odds of relapse. CONCLUSIONS
These results demonstrate no association between skeletal maturation at the end of treatment assessed with the CVM method and long-term stability of incisor alignment. Pretreatment II and postretention time were found to be correlated with increased relapse. Sex was not related to incisor relapse. REFERENCES 1. Årtun J, Krogstad O, Little RM. Stability of mandibular incisors following excessive proclination: a study in adults with surgically treated mandibular prognathism. Angle Orthod 1990;60:99106. 2. Årtun J, Garol JD, Little RM. Long-term stability of mandibular incisors following successful treatment of Angle Class II, Division 1 malocclusions. Angle Orthod 1996;66:229-38. 3. Edwards JG. A long-term prospective evaluation of the circumferential supracrestal fiberotomy in alleviating orthodontic relapse. Am J Orthod Dentofacial Orthop 1988;93:380-7.
244 Fudalej, Rothe, and Bollen
4. Glenn G, Sinclair PM, Alexander RG. Nonextraction orthodontic therapy: posttreatment dental and skeletal stability. Am J Orthod Dentofacial Orthop 1987;92:321-8. 5. Kahl-Nieke B, Fischbach H, Schwarze CW. Post-retention crowding and incisor irregularity: a long-term follow-up evaluation of stability and relapse. Br J Orthod 1995;22:249-57. 6. Little RM, Wallen TR, Riedel RA. Stability and relapse of mandibular anterior alignment-first premolar extraction cases treated by traditional edgewise orthodontics. Am J Orthod 1981;80:349-65. 7. Little RM, Riedel RA. Postretention evaluation of stability and relapse—mandibular arches with generalized spacing. Am J Orthod Dentofacial Orthop 1989;95:37-41. 8. Luppanapornlarp S, Johnston LE Jr. The effects of premolarextraction: a long-term comparison of outcomes in “clear-cut” extraction and nonextraction Class II patients. Angle Orthod 1993;63:257-72. 9. McReynolds DC, Little RM. Mandibular second premolar extraction postretention evaluation of stability and relapse. Angle Orthod 1990;61:133-44. 10. Paquette DE, Beattie JR, Johnston LE. A long-term comparison of nonextraction and premolar extraction edgewise therapy in “borderline” Class II patients. Am J Orthod Dentofacial Orthop 1992;102:1-14. 11. Rothe LE, Bollen AM, Little RM, Herring SW, Chaison JB, Chen CSK, et al. Trabecular and cortical bone as risk factors for orthodontic relapse. Am J Orthod Dentofacial Orthop 2006;130: 476-84. 12. Ormiston JP, Huang GJ, Little RM, Decker JD, Seuke GD. Retrospective analysis of long-term stable and unstable orthodontic treatment outcomes. Am J Orthod Dentofacial Orthop 2005;128:568-74. 13. Little RM, Riedel RA, Årtun J. An evaluation of changes in mandibular anterior alignment from 10 to 20 years postretention. Am J Orthod Dentofacial Orthop 1988;93:423-8. 14. Blake M, Bibby K. Retention and stability: a review of the literature. Am J Orthod Dentofacial Orthop 1998;114:299-306. 15. Shah AA. Postretention changes in mandibular crowding: a review of the literature. Am J Orthod Dentofacial Orthop 2003;124:298-308. 16. Buschang PH, Shulman JD. Incisor crowding in untreated persons 15-50 years of age: United States, 1988-1994. Angle Orthod 2003;73:502-8. 17. Driscoll-Gilliland J, Buschang PH, Behrents RG. An evaluation of growth and stability in untreated and treated subjects. Am J Orthod Dentofacial Orthop 2001;120:588-97. 18. Richardson ME. Late lower arch crowding: the role of facial morphology. Angle Orthod 1986;56:244-54. 19. Perera PSG. Rotational growth and incisor compensation. Angle Orthod 1987;57:39-49. 20. Issacson R, Zapfel R, Worms F, Erdmand A. Effects of rotational jaw growth on the occlusion and profile. Am J Orthod 1977;72: 276-86. 21. Behrents RG. A treatise on the continuum of growth in the aging craniofacial skeleton [thesis]. Ann Arbor: University of Michigan; 1984.
American Journal of Orthodontics and Dentofacial Orthopedics August 2008
22. Formby WA, Nanda RS, Currier GF. Longitudinal changes in the adult facial profile. Am J Orthod Dentofacial Orthop 1994;105: 464-76. 23. Bondevik O. Growth changes in the cranial base and the face: a longitudinal cephalometric study of linear and angular changes in adult Norwegians. Eur J Orthod 1995;17:525-32. 24. Foley TF, Mamandras AH. Facial growth in females 14 to 20 years of age. Am J Orthod Dentofacial Orthop 1992;101:248-54. 25. Pollard LE, Mamandras AH. Male postpubertal facial growth in Class II malocclusions. Am J Orthod Dentofacial Orthop 1995; 108:62-8. 26. Bishara SE, Treder JE, Jakobsen JR. Facial and dental changes in adulthood. Am J Orthod Dentofacial Orthop 1994;106:175-86. 27. Sinclair PM, Little RM. Maturation of untreated normal occlusion. Am J Orthod 1983;83:114-23. 28. Fudalej P, Årtun J. Mandibular growth rotation effects on postretention stability of mandibular incisor alignment. Angle Orthod 2007;77:199-205. 29. Williams S, Andersen CE. Incisor stability in patients with anterior rotational mandibular growth. Angle Orthod 1995;65: 431-42. 30. Nanda RS, Nanda SK. Considerations of dentofacial growth in long-term retention and stability: is active retention needed? Am J Orthod Dentofacial Orthop 1992;101:297-302. 31. Hägg U, Taranger J. Maturation indicators and the pubertal growth spurt. Am J Orthod 1982;82:299-309. 32. Lamparski D. Skeletal age assessment utilizing cervical vertebrae [thesis]. Pittsburgh: University of Pittsburgh; 1972. 33. Little RM. The irregularity index: a quantitative score of mandibular anterior alignment. Am J Orthod 1975;68:554-63. 34. Baccetti T, Franchi L, McNamara JA Jr. An improved version of the cervical vertebral maturation (CVM) method for the assessment of mandibular growth. Angle Orthod 2002;72:316-23. 35. Hassel B, Farman AG. Skeletal maturation evaluation using cervical vertebrae. Am J Orthod Dentofacial Orthop 1995;107: 58-66. 36. Franchi L, Baccetti T, McNamara JA Jr. Mandibular growth as related to cervical vertebral maturation and body height. Am J Orthod Dentofacial Orthop 2000;118:335-40. 37. San Roman P, Palma JC, Oteo MD, Nevado E. Skeletal maturation determined by cervical vertebrae development. Eur J Orthod 2002;24:303-11. 38. Riolo ML, Moyers RE, McNamara JA, Hunter WS. An atlas of craniofacial growth: cephalometric standards from the University School Growth Study. Monograph 2. Craniofacial Growth Series. Ann Arbor: Center for Human Growth and Development; University of Michigan; 1974. 39. Kahl-Nieke B, Schwarze CW. Post-retention crowding and incisor irregularity: a long-term follow-up evaluation of stability and relapse. Br J Orthod 1995;22:249-57. 40. Little RM. Stability and relapse of dental arch alignment. Br J Orthod 1990;17:235-41. 41. Bondevik O. Changes in occlusion between 23 and 34 years of age. Angle Orthod 1998;68:75-80. 42. Richardson ME, Gormley JS. Lower arch crowding in the third decade. Eur J Orthod 1998;20:597-607.