Professional Documents
Culture Documents
568
Ormiston et al 569
570 Ormiston et al
Table I.
Demographics
Unstable (n 41)
Variable
Mean age at T1 (y)
Mean treatment time (y)
Mean retention time (y)
Mean post retention time (y)
Mean T2 MN 3-3 width (mm)
Mean T3 MN 3-3 width (mm)
Angle class (% Class II)
Class II Division 1 (% of all Class II)
Sex (% female)
Extraction (% with extractions)
Stable (n 45)
P value
Mean or %
SD
Minimum
Maximum
Mean or %
SD
Minimum
Maximum
.276
.325
.242
.182
12.6
2.3
2.5
15.1
36.7
36.4
70.7
72.4
48.8
63.4
2.4
.9
1.6
5.0
3.8
3.8
8.8
.9
.5
9.3
24.9
24.6
23.9
5.3
8.4
32.8
42.2
42.0
12.1
2.1
3.0
13.8
36.0
35.6
46.7
76.2
80.0
51.1
2.3
1.1
1.8
3.7
3.9
4.0
7.2
.8
.5
7.4
21.5
21.6
16.2
7.0
10.1
23.3
44.9
44.0
.596
.024
.764
.002
.250
with stability in the univariate analyses were investigated in multiple logistic regression analyses.
RESULTS
Ormiston et al 571
Table II.
Time
P value
ABO score
ABO score
Unweighted PAR
Unweighted PAR
Unweighted PAR
Weighted PAR
Weighted PAR
Weighted PAR
II
II
II
Intercanine width (mm)
Intercanine width (mm)
Intercanine Width (mm)
Intermolar width (mm)
Intermolar width (mm)
Intermolar width (mm)
Arch length (mm)
Arch length (mm)
Arch length (mm)
Overbite (mm)
Overbite (mm)
Overbite (mm)
Overjet (mm)
Overjet (mm)
Overjet (mm)
Mean molar distance from Class I (mm)
Mean molar distance from Class I (mm)
Mandibular plane angle ()
Mandibular plane angle ()
Mandibular plane angle ()
SNA ()
SNA ()
SNA ()
SNB ()
SNB ()
SNB ()
ANB ()
ANB ()
ANB ()
Maxillary Harvold (mm)
Maxillary Harvold (mm)
Maxillary Harvold (mm)
Mandibular Harvold (mm)
Mandibular Harvold (mm)
Mandibular Harvold (mm)
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
T1
T2
T3
.0072
.0000
.0000
.0000
.0000
.0001
.0004
.0000
.0231
.0965
.0000
.8092
.0649
.9503
.3008
.9083
.9899
.0263
.4667
.0721
.1744
.7377
.0036
.0299
.0523
.0000
.9430
.0140
.9548
.6342
.7774
.5236
.5740
.5245
.0451
.0349
.0190
.0325
.0199
.0155
.9162
.9911
.3532
.3470
.4669
.6347
Mean
SD
31.6 12.3
35.9 12.0
18.2
5.3
3.7
2.1
12.6
1.7
35.8
8.9
5.5
4.1
21.6
5.1
5.7
3.7
1.2
1.0
5.3
2.2
25.0
2.4
26.7
3.1
24.6
2.9
41.8
3.3
42.1
2.5
41.8
3.1
57.4
4.8
53.6
5.8
50.3
6.3
4.5
2.3
2.4
1.1
3.6
1.9
6.5
3.7
2.5
0.9
3.93 1.50
0.37 1.31
1.23 1.64
35.7
6.6
36.1
6.8
34.2
7.6
81.6
3.4
80.1
3.4
80.5
4.0
76.0
3.5
76.0
3.6
76.5
3.9
5.6
2.0
4.1
2.3
4.0
2.8
86.4
5.5
87.8
5.6
90.6
6.4
99.4
4.8
104.5
6.3
109.4
8.7
Stable
Minimum
Maximum
10.0
20.0
9.0
1.0
11.0
21.0
1.0
13.0
0.5
0.0
0.0
19.2
17.4
15.1
33.6
37.0
36.5
45.5
45.9
38.0
1.0
0.0
1.5
2.0
0.0
1.00
7.00
5.75
24.1
25.1
21.6
73.0
72.2
70.8
67.9
68.2
69.6
2.1
0.2
1.7
70.1
72.5
74.5
87.7
93.2
95.6
66.0
64.0
29.0
9.0
17.0
58.0
17.0
41.0
15.6
4.6
9.1
29.4
36.1
29.3
48.8
46.2
48.8
68.0
64.7
62.5
10.0
4.0
9.0
13.5
5.0
8.50
1.75
2.75
50.1
52.5
53.8
89.5
86.0
87.4
82.0
81.9
83.7
11.2
11.8
12.0
95.0
97.6
100.9
110.0
121.6
132.2
Mean
SD
25.2
9.0
17.8
6.2
13.3
5.0
2.0
1.3
1.5
0.7
27.6 10.2
2.9
2.3
2.5
1.9
3.9
3.6
0.9
0.7
1.7
0.8
24.7
2.2
25.6
2.5
24.6
3.0
42.5
2.6
42.2
2.8
41.8
3.2
59.7
4.8
54.5
6.2
52.6
5.8
3.9
2.0
2.3
1.1
2.7
0.9
4.9
3.5
2.1
0.7
2.42 0.74
0.39 1.17
0.34 0.62
35.6
5.1
35.5
5.4
33.8
5.9
82.2
4.3
80.6
4.3
81.0
4.5
77.6
3.8
77.7
3.7
78.4
3.6
4.6
2.6
2.9
2.5
2.6
2.4
86.2
5.2
87.8
4.7
89.4
5.7
100.5
6.5
105.5
6.3
108.6
7.1
Minimum
Maximum
10.0
5.0
2.0
0.0
0.0
3.0
0.0
0.0
0.0
0.0
0.0
16.5
15.1
14.3
37.3
36.7
37.2
47.4
44.0
43.8
0.5
1.0
1.0
2.0
0.0
1.5
7.0
4.0
23.9
22.0
21.7
74.2
72.6
72.9
69.7
70.0
71.6
1.6
2.7
1.8
75.2
79.2
80.4
88.9
95.1
96.8
51.0
34.0
27.0
6.0
2.0
46.0
9.0
8.0
14.7
2.7
3.4
30.0
30.4
32.3
48.6
49.1
48.9
69.9
64.0
61.9
13.5
5.0
4.0
12.0
4.0
5.0
1.5
1.8
45.2
44.3
45.2
91.0
91.1
95.2
87.3
88.8
88.5
9.9
9.1
9.0
99.7
99.8
101.0
118.0
122.6
126.2
572 Ormiston et al
Table III.
Category
T2 point range
T2 mean
T3 mean
P value
10-20
20-30
30-40
over 40
15.1
24.1
34.1
47.9
19.8
24.8
26.8
40.6
.0679
.8778
.0012
.1248
19
32
23
12
1
2
3
4
would have approximately half the risk of being unstable (0.885 0.53). This variable intuitively makes
sense, because the greater the initial arch length, the
more likely that a patient has adequate or excess space.
If we expect a return to the original condition, these
patients are less likely to have crowding in the long
term.
Birkeland et al15 and Simons and Joondeph12 also
found that patients with more extreme malocclusions
before treatment tended to have greater relapses postretention. Shields et al,9 in contrast to our study, found
no predictors of stability. The difference between our
study and that of Shields et al might be the differences
in sample selection and scoring discussed earlier. Fidler
et al17 were also unable to detect any predictors of
stability. This might be because few subjects experienced significant amounts of relapse, which might have
led to reduced statistical power and the failure to detect
any predictors of relapse. The II at T1 was also
predictive of stability, having a crude odds ratio that
was similar to that of the PAR score at T1. Kahl-Nieke
et al29 also found that some pretreatment variables were
associated with postretention crowding and increases in
incisor irregularity. Their study involved a population
that was primarily treated with removable appliances,
and, therefore, the results might not completely apply to
a population treated primarily with fixed edgewise
appliances.
These results indicate that there is a strong correlation between occlusal index scores at T1 and postretention stability. This information suggests that we
should use long-term retention for patients with more
severe initial malocclusions.
Absent from the final multivariate predictive formula are factors related to the posttreatment occlusion
(eg, PAR and ABO at T2). Although the posttreatment
PAR, II, and ABO scores were all lower for the stable
group, these differences were not large (less than 2
unweighted PAR points, less than 0.5 mm of irregularity, and less than 7 ABO points). This suggests that
patients from both groups were finished to comparable
levels, and that the posttreatment condition was not the
most influential factor in long-term stability. Although
some believe that the perfection of the occlusion
reduces the likelihood of subsequent relapse,18-20 the
analysis of the ABO scores at T2 shows a regression to
the mean, in which subjects with low T2 scores tend to
increase (deteriorate) at T3, and those with high T2
scores tend to decrease (improve) at T3.
Several studies contradict our results, finding that
finish quality had greater effects on stability. KahlNieke et al29 concluded that Class II or III molar
relationships that remained at the end of treatment were
Ormiston et al 573
associated with postretention changes of incisor alignment. The limitations of this study were discussed
previously. Birkeland et al15 found that PAR scores at
T2 were associated with long-term stability, but the
strength of this association was questionable (R .31).
The ABO was much more critical in scoring occlusions
at T2; this might explain the difference in conclusions
when the PAR and ABO indexes are used. Parkinson et
al30 found that patients with the greatest amount of
occlusal contact at T2 had the least amount of overbite
increase between T2 and T3, but they used some
individual measurements in their study. The use of a
single comprehensive occlusal index such as the PAR
or the ABO might have made comparisons to our result
more meaningful.
Several recent studies support our result. Otuyemi
and Jones21 found that PAR scores in well-treated
patients tended to deteriorate over time. Only 38% of
the patients had maintained their treatment results at T3
as measured by the PAR index. In relating T2 occlusal
discrepancies with T3 irregularity index changes, Weiland31 noted that occlusal slides (shifts) decreased from
T2 to T3, suggesting that some problems at T2 will
diminish in severity by T3. The scores obtained with
the ABO index reflected a similar pattern; the ABO
scores tended to regress toward the mean. Little et al16
also showed a similar tendency for changes in occlusal
relationships. Kashner24 (using the British weighting of
the PAR) and Fernandes23 found no correlation between level of treatment quality and long-term stability.
Our results suggest that finishing occlusal relationships to perfection might not ensure postretention
stability. If this is true, other factors might play more
important roles in stability, such as treatment and
retention methods, compliance, the tendency to return
to the pretreatment condition, growth, and other factors
that we might be unaware of. Additionally, there was
considerable individual variation in stability after
orthodontic treatment.
Our study had some limitations. Information about
treatment and retention was limited to what was available in the patient records. Various cephalometric
machines had been used, and quantifying magnification
between radiographs was problematic for linear measurements. The number of patients with restorations
generally increased over time; this might alter or mask
marginal ridge discrepancies and true alignment of
teeth. No attempt was made to control for this problem.
Information about supracrestal fiberotomy was not
available from most patient charts, but this procedure
was not commonly used when most of these patients
were treated. In addition, by selecting the most stable
and the most unstable adolescent subjects for the
574 Ormiston et al
REFERENCES
1. Blake M, Bibby K. Retention and stability: a review of the
literature. Am J Orthod Dentofacial Orthop 1998;114:299-306.
2. Isaacson RJ, Zapfel RJ, Worms FW, Erdman AG. Effects of
rotational jaw growth on the occlusion and profile. Am J Orthod
1977;72:276-86.
3. Schudy GF. Posttreatment craniofacial growth: its implications
in orthodontic treatment. Am J Orthod 1974;65:39-57.
4. Siatkowski RE. Incisor uprighting: mechanism for late secondary
crowding in the anterior segments of the dental arches. Am J
Orthod 1974;66:398-410.
5. Sampson WJ. Current controversies in late incisor crowding.
Ann Acad Med Singapore 1995;24:129-37.
6. McReynolds DC, Little RM. Mandibular second premolar extractionpostretention evaluation of stability and relapse. Angle
Orthod 1991;61:133-44.
7. 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.
8. Sinclair PM, Little RM. Dentofacial maturation of untreated
normals. Am J Orthod 1985;88:146-56.
9. Shields TE, Little RM, Chapko MK. Stability and relapse of
mandibular anterior alignment: a cephalometric appraisal of
first-premolar-extraction cases treated by traditional edgewise
orthodontics. Am J Orthod 1985;87:27-38.
10. Harris EH, Gardner RZ, Vaden JL. A longitudinal cephalometric
study of postorthodontic craniofacial changes. Am J Orthod
Dentofacial Orthop 1999;115:77-82.