ORIGINAL ARTICLE

Occlusal changes during a 10-year

posttreatment period and the effect
of fixed retention on anterior tooth
alignment
Ragnar Bjering and Vaska Vandevska-Radunovic
Oslo, Norway

Introduction: The objectives of this research were to evaluate changes in occlusal components in 3 subperiods
during a 10-year posttreatment time span and to examine the long-term effects of fixed retention on maxillary and
mandibular anterior alignment. Methods: Ninety-six patients were examined; the Peer Assessment Rating
Index and Little's Irregularity Index were measured at pretreatment, posttreatment, and 3 (T3), 5 (T5), and
10 (T10) years posttreatment. Unweighted Peer Assessment Rating component scores were analyzed for
differences between all subperiods. The effect of fixed retention on posttreatment changes in Little's Irregularity
Index was analyzed for both jaws with regression analysis. For the maxilla, 2 groups were compared: MX0,
removable retainer until T3 (n 5 52) and MX10, removable retainer until T3 combined with a fixed retainer until
T10 (n 5 23). For the mandible, 3 groups were compared: MD3, fixed retainer until T3 (n 5 19); MD5, fixed
retainer until T5 (n 5 19); and MD10 fixed retainer until T10 (n 5 48). Results: The Peer Assessment Rating
Index percentage of improvement was 79% at T10. A gradual deterioration of occlusal components was
seen, with small insignificant changes in each subperiod. Corrected for pretreatment irregularity, MX10 showed
0.6 mm lower LII than MX0. MD10 had significantly better alignment than MD3 (1.1 mm) and MD5 (0.7 mm).
Conclusions: Gradual occlusal changes of limited clinical importance were seen during a 10-year
posttreatment period. Long-term fixed retention in the maxilla was of minor importance in patients also
wearing removable retainers. In the mandible, a 10-year fixed retention protocol gave moderately lower
alignment scores compared to a 3-year protocol and slightly better alignment compared to a 5-year protocol.
(Am J Orthod Dentofacial Orthop 2018;154:487-94)

T
otal stability of the occlusion after orthodontic
treatment seems to be unlikely to achieve. With
time, changes in tooth alignment are inevitable,
and both relapse and growth are contributing factors.1
To counteract unwanted changes, orthodontists
prescribe different types of retainers, sometimes
intended to be used indefinitely.
Both removable and fixed retainers have been found to
be equally effective in controlling relapse up to 2 years
posttreatment.2,3 Part-time wear of thermoplastic and
From the Department of Orthodontics, Faculty of Dentistry, University of Oslo,
Oslo, Norway.
All authors have completed and submitted the ICMJE Form for Disclosure of
Potential Conflicts of Interest, and none were reported.
Address correspondence to: Ragnar Bjering, Department of Orthodontics,
Faculty of Dentistry, University of Oslo, PO Box 1109, Blindern, 0317 Oslo,
Norway; e-mail, ragnar.bjering@odont.uio.no.
Submitted, August 2017; revised and accepted, December 2017.
0889-5406/$36.00
Ó 2018 by the American Association of Orthodontists. All rights reserved.
https://doi.org/10.1016/j.ajodo.2017.12.015
Hawley retainers has proved to be as efficient as full-time
wear.4,5 Methods for controlling the anterior alignment
even without retainers such as interproximal reduction,
sometimes used in combination with overcorrection, have
been shown to be viable in the short term.3,6
It is unclear at what point after debonding the
greatest posttreatment changes occur. Some authors
have found that most changes take place during the
first 2 years after treatment, a period corresponding
well with relapse and settling.7 In contrast, early
stability has been reported for all components of
occlusion 2 years after treatment.8 Others have reported
that most occlusal changes take place during the first
4 years.9 Moreover, long-term studies have concluded
that significant occlusal changes take place even
between 19 and 31 years of age.10 With more
information about posttreatment changes, one could
possibly improve retention strategies.
Prolonged retention can interfere with the natural
reduction in dental arch parameters. Since long-term
487
488
compliance with a removable retainer is expected to be
limited and less practical, a semipermanent or
permanent retainer is often the bonded type. It has
been stated that lifetime permanent retention is the
only way to prevent relapse.11 Al Yami et al7 found a
positive effect of fixed retainers on the Peer Assessment
Raing (PAR) score 11 years posttreatment. Furthermore,
presence of fixed retainers gave better occlusal results
17 years posttreatment in a long-term follow-up.10 In
contrast, some studies have concluded that fixed
retention is not of major importance to the treatment
outcome. Rather than being a protective measure
against long-term changes on an occlusal level, fixed
retainers were found to primarily inhibit changes in
anterior alignment.12 The efficacy of fixed retainers on
mandibular anterior relapse has been reported.13
Nonetheless, satisfactory alignment has also been found
at 10 years postretention, even after a short retention
protocol.14 In the maxilla, fixed retainers appear to
have less influence on the stability of alignment
compared with the mandible.13,15 There seems to be
uncertainty about how much a fixed retainer will
improve the alignment in the long term.
For the time being, the preferred type and duration
of retention have not been established.16,17 Use of
retention appliances varies between countries and
largely depends on personal preferences.18,19 It is
therefore important to improve our knowledge about
the effect of different retention protocols, types, and
durations on long-term treatment outcomes.
The aims of the study were to evaluate changes in
occlusal components in 3 stages during a 10-year
posttreatment period and to examine the effect of type
and duration of retention on maxillary and mandibular
anterior alignment.
MATERIAL AND METHODS
The Department of Orthodontics at the University of
Oslo in Norway routinely summons patients for
checkups at 3, 5, and 10 years posttreatment. Included
in this retention archive are nonsurgical patients aged
20 years or younger at the beginning of treatment,
without agenesis, trauma, or autotransplantations to
the anterior regions. To detect a minimum difference
of 10 PAR score points between pretreatment and
10 years posttreatment with a standard deviation of
11, a sample size of 12 patients was required to provide
80% statistical power with an alpha of 0.05. Attendance
at the 10-year follow-up appointment per March 22,
2017 served as inclusion criteria for this longitudinal
analytical study. Approval was granted by the Regional
Committee for Medical and Health Research Ethics and
the Norwegian Data Protection Official for Research.
October 2018
Vol 154
Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Bjering and Vandevska-Radunovic
Table I. Dental cast measurements with definitions
Measurement Definition
PAR index Measured according to the conventions
described by Richmond et al20
LII The sum of the linear displacements of the
anatomic contact points from canine to
canine, according to Little21
Overjet Distance parallel to the occlusal plane
from the buccal surface of the most
protruding maxillary incisor to the
buccal surface of the corresponding
mandibular incisor
Overbite Maximum distance of the mandibular
incisors overlapped by the maxillary
central incisors
Canine relationship* Distance from the cusp tip of the maxillary
canine to the distal contact point of the
mandibular canine
Molar relationship* Deviation from a neutral occlusion,
defined as occlusion of the mesiobuccal
cusp of the maxillary first molar within
the buccal groove of the mandibular
first molar
Intercanine distance Distance between the cusp tips of fully
erupted teeth
*Distal occlusions were recorded as positive values; mesial occlusions
as negative values.
One hundred twenty-five patients met the inclusion
criteria. Exclusions were made according to the
following criteria: missing or damaged dental cast
(pretreatment, posttreatment, or 10-year follow-up)
(n 5 18), retreatment (n 5 5), single-arch treatment
(n 5 4), and extractions of incisors (n 5 2). The final
study sample included 96 patients (43 male, 53 female)
treated with full fixed appliances. Study casts were
available for all patients at pretreatment (T0),
posttreatment (T1), and 10 years posttreatment (T10),
as well as for 70 patients at 3 years posttreatment (T3)
and 86 patients at 5 years posttreatment (T5). Fifty-six
patients were treated without extractions, 28 patients
were treated with extraction of 4 premolars, 8 patients
had extraction of 2 maxillary premolars, and 4 patients
had extraction of 2 mandibular premolars. Mean
pretreatment age was 12.2 years (61.4). Mean treatment
duration was 2.6 years (60.9). Average follow-up
periods were 3.1 years (60.4) at T3, 5.3 years (60.6)
at T5, and 10.1 years (60.8) at T10.
The PAR index20 was used to assess the occlusion at
all time stages. Anterior tooth alignment was scored
using Little's Irregularity Index (LII).21 In addition,
several dental cast measurements were registered
(Table I). All variables were measured by the same
examiner (R.B.) to the closest 0.1 mm using a digital
caliper (Digital 6; Mauser, Oberndorf, Germany), except
Bjering and Vandevska-Radunovic
Table II. Number of patients and distribution among
the retention protocols for the total sample and sub-
groups in the retention analysis
Maxilla
Total patient sample
No fixed retention (removable 52
retainer in maxilla)
Fixed retainer until T3 6 19
Fixed retainer until T5 7 19
Fixed retainer until T10 31
Subgroups used in retention analysis
MX0 Removable retainer until T3 52
MX10 Removable retainer until T3 23
combined with fixed retainer
until T10
MD3 Fixed retainer until T3
MD5 Fixed retainer until T5
MD10 Fixed retainer until T10
A subgroup was created only if a sufficient number of patients could
be assigned to the group. Time points used are 3 (T3), 5 (T5), and
10 years posttreatment (T10).
for overjet and overbite, which were measured to the
nearest 0.5 mm using a ruler. All patients had
received a removable retainer in the maxilla with
instructions for initial full-time wear followed by a
gradual reduction until cessation at around the
3-year checkup (T3). Duration of fixed retainer wear
varied for both arches. To examine the effect of
duration of retention on anterior tooth alignment,
retention subgroups were created based on the applied
retention protocol (Table II). For the maxilla, 2 groups
of patients were compared: MX0, removable retainer
until T3 (n 5 52); and MX10, removable retainer
until T3 combined with fixed retainer until
T10 (n 5 23). Only patients with a retainer bonded
to the four incisors (n 5 12) or all six anterior teeth
(n 5 11) were included in the MX10 group. For the
mandible, 3 groups were compared: MD3, fixed
retainer until T3 (n 5 19); MD5, fixed retainer until
T5 (n 5 19); and MD10, fixed retainer until T10
(n 5 48).
Intraexaminer reliability for the dental cast
measurements was determined by scoring 30 randomly
selected sets of models twice, 4 weeks apart; intraclass
correlation coefficients were between 0.94 and 0.99.
Skeletal characteristics and incisor positions were
measured before and after treatment with cephalometric
analysis of lateral radiographs (Fig 1; Table III).
Reliability for the cephalometric analyses was tested by
retracing 30 cephalograms after 3 weeks; intraclass
correlation coefficient values were between 0.93 and
0.99.
American Journal of Orthodontics and Dentofacial Orthopedics
489
Mandible
10
48
19
19
48
Fig 1. Landmarks used in the pretreatment and
posttreatment cephalometric analyses of skeletal
characteristics and incisor protrusion and inclination.
The tracings were oriented horizontally 7 down from
the sella-nasion line. S, sella; N, nasion; A, point A;
B, point B; Isa, incision superius apicalis; Isb, incision
superius buccalis; Ii, incision inferius; Is, incision
superius; Iib, incision inferius buccalis; Iia, incision
inferius apicalis; iGo, gonion inferius; Me, menton.
Statistical analysis
For the time periods T0-T1, T1-T3, T3-T5, T5-T10,
T0-T10, and T1-T10, continuous variables were
analyzed for differences using a paired t test. The
preventive effect of different retention protocols on
posttreatment alignment deterioration was compared
using a forced-entry linear regression analysis: with
the retention subgroups coded as dummy variables 1
regression model was needed to compare the 2 maxilla
groups, and 2 models were needed to compare the 3
mandible groups. A number of other variables, including
dental cast measurements, pretreatment skeletal
characteristics, and posttreatment dental protrusion
and inclination, were tested for confounding
and included in the regression models if required
(Table IV). When performing the regression analysis for
the maxilla, treatment changes in LII were found to be
a confounding variable; the variable was consequently
included as a covariate in the adjusted regression model.
No confounding variables were detected related to the
regression analysis for the mandible. There was no
violation of the assumptions of normality, linearity,
and homoscedasticity. A post hoc power analysis was
conducted for each regression model using the software
October 2018
Vol 154
Issue 4
490 Bjering and Vandevska-Radunovic

Comparison of the two retention subgroups in

Table III. Mean cephalometric values for the patient
maxilla showed that adjusted for treatment correction
sample at T0 and T1
in LII, presence of a fixed retainer for 10 years
T0 T1 posttreatment in addition to a removable retainer
Mean SD Mean SD
lowered the posttreatment increase in LII by
SNA ( ) 81.1 3.4 80.3 3.8 0.6 millimeters compared to patients who only had
SNB ( ) 77.0 3.4 77.3 3.7 worn a removable retainer (Table V). Statistical power
ANB ( ) 4.1 2.6 3.1 2.2 for the maxillary regression analysis with 2 independent
NL/NSL 7.5 3.5 7.8 3.7 variables was above 98% at a large effect size
ML/NSL 32.9 5.3 32.6 5.7
(R2 5 0.26) and above 79% at a moderate effect size
ML/NL 25.4 5.2 24.8 5.3
Isb-NA 5.3 2.2 4.3 1.6 (R2 5 0.13), with a sample size of 75 and an alpha level
lib-NB 4.9 2.6 5.1 2.0 of 0.05.
ILs/NA 22.6 9.3 23.3 5.7 The increase in the mandibular LII from posttreatment
ILi/NB 25.0 7.1 27.7 6.4 to 10 years posttreatment was 1.1 mm lower for patients
ILi/ML 95.1 7.1 97.8 7.9 wearing a fixed retainer for 10 years compared with
ILs/lLi 128.3 13.3 126.0 8.7
3 years, and 0.7 mm lower compared with 5 years. There
NL/NSL, Maxillary plane angle; ML/NSL, mandibular plane angle; were no significant differences in posttreatment changes
ML/NL, intermaxillary angle; Isb-NA, maxillary incisor protrusion
to NA line; Iib-NB, mandibular incisor protrusion to NB line; ILs/
in LII between the groups with fixed retention for 3 and
NA, maxillary incisor inclination to NA line; ILi/NB, mandibular 5 years (Table VI). Statistical power for the mandibular
incisor inclination to NB line; ILi/ML, mandibular incisor inclination analyses with 2 independent variables was above 99%
to mandibular plane; ILs/ILi, interincisor angle. at a large effect size (R2 5 0.26) and above 85% at a
moderate effect size (R2 5 0.13) , with a sample size of
package G*Power (version 3.1.9.2; Franz Faul,
86 and an alpha level of 0.05.
Universit€at Kiel, Kiel, Germany). The statistical analyses
were performed with SPSS software (version 24; IBM, DISCUSSION
Armonk, NY), using a significance level of 0.05.
This study shows that changes in occlusal parameters
such as overjet, overbite, incisor alignment, and
RESULTS
posterior occlusion underwent gradual deterioration
The mean weighted PAR score at pretreatment was during a 10-year period after orthodontic treatment. In
24.0 (69.8). The posttreatment score was 2.6 (63.4), each subperiod, the changes were insignificant. When
which increased to 3.6 (63.7) at T3, 4.5 (64.8) at T5, viewed over a 10-year perspective, the changes were
and 5.1 (64.9) at T10. The PAR index percentage statistically significant, although of limited clinical
improvements were 89.2% at T1, 85.0% at T3, 81.3% importance. For patients wearing a removable maxillary
at T5, and 78.8% at T10. Total weighted PAR scores retainer, the additional benefit of having a fixed retainer
showed statistically significant changes between all for 10 years was significant, but of minor clinical
studied time periods (Table V). The unweighted PAR importance. In the mandible, fixed retention for 10 years
component scores showed a significant reduction from gave moderately better alignment compared to a 3-year
T0 to T1 and, with the exception of the centerline protocol, and slightly better alignment compared to a
component, increased some from T1 to T10. At 10 years 5-year protocol.
posttreatment, all scores were still significantly lower The results indicate that dual retention in the maxilla
compared with pretreatment. For the stages T1-T3, is unnecessary for the average patient. When adjusted
T3-T5, and T5-T10, the main trend was small for corrections made during treatment, the effect of
insignificant changes in the different parts of the long-term wear of maxillary fixed retention in addition
occlusion (Fig 2). Only the anterior components showed to a removable retainer was small in terms of improved
significant, albeit minor, changes from T3 to T5 and stability. A removable retainer will probably be sufficient
T5 to T10. for patients with low to moderate pretreatment
The measurements for LII, overjet, overbite, irregularity. However, since posttreatment deterioration
canine/molar relationship, and maxillary/mandibular in the maxilla correlated with the degree of alignment
canine distances showed the same pattern as the PAR correction, a fixed retainer could be considered for
component scores. Apart from significant but small patients with greater pretreatment irregularity, and
changes in anterior alignment in all posttreatment certain patients might need the combination with a
stages, there were essentially no changes in these removable retainer. Previous long-term studies
variables. found a marked association between treatment and

October 2018
Vol 154
Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Bjering and Vandevska-Radunovic 491

Table IV. Variables investigated for possible confounding when performing linear regression analyses to predict the
influence of fixed retention on the outcome of maxillary and mandibular posttreatment change in LII at T10
Obtained from Variable Description MX0 MX10 MD3 MD5 MD10
Dental casts Treatment change in LII –3.9 –6.0 –3.0 –3.0 –4.4
Treatment change in overjet –2.8 –3.3 –3.1 –3.1 –2.6
Treatment change in overbite –2.0 –1.5 –1.2 –2.0 –1.8
Treatment change in intercanine distance 0.2 0.5 –0.6 0.2 0.2
Treatment change in canine relation –2.3 –1.5 –2.0 –2.9 –1.5
Lateral cephalogram Mesial basal jaw relationship at T0 ANB \1 6 1 3 3 4
Distal basal jaw relationship at T0 ANB .4 26 13 10 9 27
Low mandibular plane angle at T0 ML/NSL \29 11 7 4 7 9
High mandibular plane angle at T0 ML/NSL .37 7 7 4 0 12
Maxillary incisor protrusion at T1 Isb-NA .2 mm above norm 16 9 9 4 16
Maxillary incisor retrusion at T1 Isb-NA .2 mm below norm 2 0 0 1 1
Maxillary incisor proclination at T1 ILs/NA .3 above norm 23 12 11 9 22
Maxillary incisor retroclination at T1 ILs/NA .3 below norm 6 3 3 3 5
Mandibular incisor protrusion at T1 Iib-NB .2 mm above norm 8 7 4 5 11
Mandibular incisor retrusion at T1 Iib-NB .2 mm below norm 3 2 1 1 4
Mandibular incisor proclination at T1 ILi/NB .3 above norm 14 8 8 4 14
Mandibular incisor retroclination at T1 ILi/NB .3 below norm 3 3 1 2 6
Written records Age at T1 (y) 14.5 14.8 14.7 15.1 14.7
Follow-up time at T10 (y) 10.1 10.0 10.1 10.3 10.2
Sex (M/F) 26/26 7/16 9/10 12/7 18/30
Extraction/nonextraction 13/39 13/10 6/13 7/12 18/30

ANB, Relative protrusion of the mandible to the maxilla; ML/NSL, mandibular plane angle; Isb-NA, maxillary incisor protrusion to NA line; ILs/NA,
maxillary incisor inclination to NA line; Iib-NB, mandibular incisor protrusion to NB line; ILiNB, mandibular incisor inclination to NB line.
For each retention subgroup (MX0, MX10, MD3, MD5 and MD10) dental cast mean values in millimeters and frequencies of cephalometric values
are shown. Treatment change was defined as T1 value minus T0 value. Normal values for maxillary and mandibular incisor protrusion and incli-
nation were corrected for Steiner's acceptable compromises.

Table V. Paired sample t test of mean changes in unweighted PAR component scores, total weighted PAR scores, and
other dental cast variables from T0 to T1, T1 to T3, T3 to T5, T5 to T10, T0 to T10, and T1 to T10
T0-T1 T1-T3 T3-T5 T5-T10 T0-T10 T1-T10

(96) (70) (66) (86) (96) (96)

Anterior component- maxilla –4.1* 0.0 0.1y 0.2y –3.8* 0.3z
Anterior component-mandible –2.0* 0.0 0.1y 0.2z –1.7* 0.3*
Posterior component –2.1* 0.1 0.1 0.1 –1.8* 0.3y
Overjet component –1.7* 0.1 0.1 0.1 –1.5* 0.2*
Overbite component –1.0* 0.1 0.1 0.1 –0.7* 0.3*
Centerline component –0.3* 0.0 0.1 0.0 –0.3* 0.0
Total weighted PAR score –21.4* 0.7y 1.1y 0.9§ –18.8* 2.5*
LII-maxilla –4.6* 0.4* 0.2§ 0.3* –3.8* 0.8*
LII-mandible –3.6* 0.3* 0.3§ 0.4* –2.7* 0.9*
Overjet –2.8* 0.2 0.1 0.1 –2.5* 0.3z
Overbite –1.7* 0.1 0.2 0.1 –1.3* 0.4z
Canine relationship –1.9* 0.0 0.0 0.0 –1.9* 0.0
Molar relationship –1.9* –0.1 0.0 0.0 –1.9* 0.0
Maxillary intercanine distance 0.3 0.2 –0.1 –0.2§ 0.2 –0.1
Mandibular intercanine distance 0.0 0.1 –0.1 –0.1 0.0 0.1

LII, Molar/canine relationships, intercanine distances measured in millimeters.

Total patient sample (n 5 96) was tested. Values in parentheses are number of subjects, for some periods reduced by limited availability of study
casts at T3 and T5.
*P #0.0001; yP #0.05; zP #0.001; §P #0.01.

posttreatment alignment changes in the maxilla.14 alignment. These suggested retention strategies do not
Nonetheless, maxillary alignment tends to be quite sta- apply to special clinical circumstances such as trauma
ble in the long term, more so than mandibular to the anterior teeth, large diastemas, autotransplants,

American Journal of Orthodontics and Dentofacial Orthopedics October 2018
Vol 154
Issue 4
492 Bjering and Vandevska-Radunovic

Fig 2. Overview of mean unweighted PAR component scores at pretreatment (T0), posttreatment (T1),
3 years posttreatment (T3), 5 years posttreatment (T5), and 10 years posttreatment (T10) for the total
patient sample (n 5 96). U, upper; L, lower.

Table VI. Overview of linear regression models showing the effect of different retention protocols on changes in LII
from T1 to T10 (Little_T1T10)
Regression model Coefficients b value 95% CI P value
Maxilla*
Model P value: 0.009y (constant) 0.367 (–0.118, 0.852) 0.135
R2: 0.12 MX10* –0.643 (–1.205, –0.080) 0.026z
Reg equation: Little_T1T10 5 0.37 - 0.64*MX10 - 0.14*Little_T0T1 Little_T0T1 –0.143 (–0.243, –0.044) 0.005y

Mandible§
Model P value: \0.001k (constant) 1.611 (1.194, 2.027) \0.001k
R2: 0.21 MD5§ –0.468 (–1.058, 0.121) 0.118
Reg equation: Little_T1T10 5 1.61 - 0.47*MD5 - 1.12*MD10 MD10§ –1.115 (–1.607, –0.623) \0.001k

Mandible{
Model P value: \0.001k (constant) 1.142 (0.726, 1.559) \0.001k
R2: 0.21 MD3{ 0.468 (–0.121, 1.058) 0.118
Reg equation: Little_T1T10 5 1.14 1 0.47*MD3 - 0.65*MD10 MD10{ –0.646 (–1.138, –0.154) 0.011z
Reg, regression.
Durations of retention until T3, T5, and T10 were examined. For the maxilla, 2 groups were compared: MX0 and MX10; the model was adjusted for
treatment change in LII (Little_T0T1). For the mandible, 3 groups were compared: MD3, MD5, and MD10. The retention subgroups were coded as
dummy variables (0 as absence and 1 as presence of the retention protocol) and compared with the reference group, requiring 1 regression model for
the maxilla and 2 for the mandible.
*MX0 as reference variable; yP #0.01; zP #0.05; §MD3 as reference variable; kP #0.0001; {MD5 as reference variable.

anterior tongue thrust, periodontitis, extensive anterior
space closures, and so on.
For the mandible, a fixed retainer for 10 years gave
significant but small to moderate improvements in
alignment compared with 3-year and 5-year retainer
wear. The small difference in posttreatment irregularity
between patients with a 3-year and a 5-year retention
protocol was not statistically significant. Consequently,
our results remain inconclusive. Nevertheless, the overall
results indicate that a time factor is present regarding
duration of mandibular retention. With larger sample
sizes in subgroups MD3 and MD5, the differences
observed between the 2 groups would have been, most
probably, statistically significant. Previous research

October 2018
Vol 154
Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Bjering and Vandevska-Radunovic
found that mandibular alignment is best preserved with
a fixed retainer, and the longer the duration, the better
the alignment.13,22 But adequate long-term stability of
mandibular alignment has also been reported after
only 1.8 years of fixed retention.14 Our findings for the
mandible show that the longest retention time gave
the best alignment scores, but the advantage over a
shorter retention time was small to moderate in a
10-year perspective. This situation may of course change
as time progresses. Little et al23 regarded an irregularity
score of less than 3.5 mm to be clinically acceptable.
Authors of other studies comparing different retention
methods have considered a mean difference of 1.4 mm
in mandibular alignment irregularity to be of no clinical
importance.24 The discrepancies in LII found separating
the retention subgroups in this study are within these
values. However, to what extent a deviation will be
tolerated will vary from person to person, and the
orthodontist should strive to determine when minor
alignment changes could become disturbing to the
patient.
The PAR score improvement of 79% at 10 years
posttreatment demonstrates that overall orthodontic
treatment can be stable to a great extent, even in the
long term. Previous reports on treatment stability 10
to 17 years posttreatment have shown improvements
in the PAR index ranging from 49% to 73%.7,8,10,14,25
Compared with these, our study showed favorable
long-term outcomes. We found a gradual change in
the total weighted PAR index and most of the
unweighted occlusion parameters during the 10-year
follow-up. This is in contrast to the findings of Al
Yami et al7 and Otuyemi and Jones,8 who found that
the largest changes in occlusion took place in the first
2 and 4 years posttreatment, respectively. On a
component level, the centerline did not change,
complementing previous findings of good long-term
midline stability.7,8 The remaining components showed
minor adjustments, and the anterior alignment stood
out as the component most prone to changes.
Altogether, the observed changes resembled growth
changes seen in untreated persons.26
There are some limitations to this study. As a
retrospective study, it does not have the scientific impact
of a randomized controlled trial. However, a randomized
controlled trial with a follow-up period of 10 years can
be difficult to complete because of dropouts. Also, the
regression models for the maxilla and the mandible
differ in terms of patient samples and the number of
groups. They were based on slightly different patient
samples according to the disparity in duration of fixed
retainer wear. Therefore, the results from them are not
directly comparable with each other.
American Journal of Orthodontics and Dentofacial Orthopedics
493
Furthermore, the low R2 values imply that the
significant regression variables lack some precision in
their estimates. The confidence intervals for these
variables were approximately 1 mm wide, demonstrating
some uncertainty in the expected differences in retentive
effects between the retention protocols. This shows that
the variability in posttreatment alignment deterioration
was not broadly explained by differences in retainer
wear alone, indicating that other factors also contribute
to the process. It seems reasonable to expect that
complex processes such as growth and maturation
account for some variations in the patient sample. Since
cephalometric measurements were not available for T10,
we do not know how growth and incisor positions
changed during the posttreatment period. However,
decisions regarding retention in clinical practice are
made at the time of debonding at the latest. Any
information to significantly influence the choice of
retention strategy should be present at that time. Growth
prediction has proved to be difficult.27 For the examined
patients, neither pretreatment sagittal/vertical jaw
relationship nor posttreatment incisor position outside
the normal range were significantly associated with the
dependent and independent variables. This contrasts
with the findings of Franklin et al,28 who reported
cephalometric values such as vertical dimension and
mandibular incisor to the A-Pg plane to be important
predictive factors for long-term alignment stability.
However, in our study, many patients still wore retainers
at T10, whereas the study of Franklin et al was done at
least 5 years postretention. From a clinical standpoint,
it seems reasonable that the advantages of prolonged
retention must outweigh the potential disadvantages.
Therefore, all aspects of long-term effects of retention
appliances must be studied to establish a preferred
duration of retention.
CONCLUSIONS
1. Stability of occlusion measured by the PAR index
percentage improvement method was 79% at
10 years posttreatment.
2. Components of the occlusion underwent gradual
and minor deterioration dispersed evenly
throughout the 10-year posttreatment period.
3. Posttreatment changes in maxillary alignment were
significantly negatively correlated with the induced
treatment changes.
4. In patients wearing a removable maxillary retainer
for 3 years, the additional benefit of wearing a fixed
retainer for 10 years was minor regarding stability of
maxillary alignment.
October 2018
Vol 154
Issue 4
494
5. In the mandible fixed retention for 10 years gave
moderately better alignment compared to a 3-year
protocol and slightly better alignment compared
to a 5-year protocol.
ACKNOWLEDGMENTS
We thank Andrej Grijbovski at the Norwegian
Institute of Public Health for statistical guidance and
Tanya Franzen for editing.
REFERENCES
1. Ormiston JP, Huang GJ, Little RM, Decker JD, Seuk GD. Retrospec-
tive analysis of long-term stable and unstable orthodontic treatment
outcomes. Am J Orthod Dentofacial Orthop 2005;128:568-74.
2. Hoybjerg AJ, Currier GF, Kadioglu O. Evaluation of 3 retention
protocols using the American Board of Orthodontics cast and
radiograph evaluation. Am J Orthod Dentofacial Orthop 2013;
144:16-22.
3. Edman Tynelius G, Bondemark L, Lilja-Karlander E. A randomized
controlled trial of three orthodontic retention methods in Class I
four premolar extraction cases—stability after 2 years in retention.
Orthod Craniofac Res 2013;16:105-15.
4. Gill DS, Naini FB, Jones A, Tredwin CJ. Part-time versus full-time
retainer wear following fixed appliance therapy: a randomized
prospective controlled trial. World J Orthod 2007;8:300-6.
5. Shawesh M, Bhatti B, Usmani T, Mandall N. Hawley retainers full-
or part-time? A randomized clinical trial. Eur J Orthod 2010;32:
165-70.
6. Aasen TO, Espeland L. An approach to maintain orthodontic align-
ment of lower incisors without the use of retainers. Eur J Orthod
2005;27:209-14.
7. Al Yami EA, Kuijpers-Jagtman AM, van 't Hof MA. Stability of or-
thodontic treatment outcome: follow-up until 10 years postreten-
tion. Am J Orthod Dentofacial Orthop 1999;115:300-4.
8. Otuyemi OD, Jones SP. Long-term evaluation of treated class II di-
vision 1 malocclusions utilizing the PAR index. Br J Orthod 1995;
22:171-8.
9. Greco PM, English JD, Briss BS, Jamieson SA, Kastrop MC,
Castelein PT, et al. Posttreatment tooth movement: for better or
for worse. Am J Orthod Dentofacial Orthop 2010;138:552-8.
10. Lagerstrom L, Fornell AC, Stenvik A. Outcome of a scheme for
specialist orthodontic care, a follow-up study in 31-year-olds.
Swed Dent J 2011;35:41-7.
11. Little RM. Clinical implications of the University of Washington
post-retention studies. J Clin Orthod 2009;43:645-51.
12. Steinnes J, Johnsen G, Kerosuo H. Stability of orthodontic treat-
ment outcome in relation to retention status: an 8-year follow-
up. Am J Orthod Dentofacial Orthop 2017;151:1027-33.
October 2018
Vol 154
Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Bjering and Vandevska-Radunovic
13. Bjering R, Birkeland K, Vandevska-Radunovic V. Anterior tooth
alignment: a comparison of orthodontic retention regimens 5
years posttreatment. Angle Orthod 2015;85:353-9.
14. Bjering R, Sandvik L, Midtbo M, Vandevska-Radunovic V. Stability
of anterior tooth alignment 10 years out of retention. J Orofac Or-
thop 2017;78:275-83.
15. Andren A, Naraghi S, Mohlin BO, Kjellberg H. Pattern and amount
of change after orthodontic correction of upper front teeth 7 years
postretention. Angle Orthod 2010;80:432-7.
16. Littlewood SJ, Millett DT, Doubleday B, Bearn DR,
Worthington HV. Retention procedures for stabilising tooth posi-
tion after treatment with orthodontic braces. Cochrane Database
Syst Rev 2016;CD002283.
17. Al-Moghrabi D, Pandis N, Fleming PS. The effects of fixed and
removable orthodontic retainers: a systematic review. Prog Orthod
2016;17:24.
18. Renkema AM, Sips ET, Bronkhorst E, Kuijpers-Jagtman AM. A sur-
vey on orthodontic retention procedures in The Netherlands. Eur J
Orthod 2009;31:432-7.
19. Vandevska-Radunovic V, Espeland L, Stenvik A. Retention: type,
duration and need for common guidelines. A survey of Norwegian
orthodontists. Orthodontics (Chic.) 2013;14:e110-7.
20. Richmond S, Shaw WC, O'Brien KD, Buchanan IB, Jones R,
Stephens CD, et al. The development of the PAR Index (Peer
Assessment Rating): reliability and validity. Eur J Orthod 1992;
14:125-39.
21. Little RM. The irregularity index: a quantitative score of mandib-
ular anterior alignment. Am J Orthod 1975;68:554-63.
22. de Bernabe PG, Montiel-Company JM, Paredes-Gallardo V,
Gandia-Franco JL, Bellot-Arcis C. Orthodontic treatment stability
predictors: a retrospective longitudinal study. Angle Orthod
2017;87:223-9.
23. Little RM, Wallen TR, Riedel RA. Stability and relapse of mandib-
ular anterior alignment-first premolar extraction cases treated by
traditional edgewise orthodontics. Am J Orthod 1981;80:
349-65.
24. Edman Tynelius G, Petren S, Bondemark L, Lilja-Karlander E. Five-
year postretention outcomes of three retention methods–a ran-
domized controlled trial. Eur J Orthod 2015;37:345-53.
25. Woods M, Lee D, Crawford E. Finishing occlusion, degree of stabil-
ity and the PAR index. Aust Orthod J 2000;16:9-15.
26. Eslambolchi S, Woodside DG, Rossouw PE. A descriptive study of
mandibular incisor alignment in untreated subjects. Am J Orthod
Dentofacial Orthop 2008;133:343-53.
27. Leslie LR, Southard TE, Southard KA, Casko JS, Jacobsen JR,
Tolley EA, et al. Prediction of mandibular growth rotation: assess-
ment of the Skieller, Bjork, and Linde-Hansen method. Am J Or-
thod Dentofacial Orthop 1998;114:659-67.
28. Franklin S, Rossouw PE, Woodside DG, Boley JC. Searching for
predictors of long-term stability. Semin Orthod 2013;19:
279-92.