ONLINE ONLY

Stability of unilateral posterior crossbite

correction in the mixed dentition: A randomized
clinical trial with a 3-year follow-up
Sofia Petre
n,a Krister Bjerklin,b and Lars Bondemarkc
Malm€o, Sweden

Introduction: The long-term stability of posterior crossbite correction in the mixed dentition has not been
sufficiently evaluated. Our aim was to compare long-term outcomes in patients with crossbite correction by
using matched controls with normal occlusion. Methods: After 35 patients were treated for crossbite with
a quad-helix or an expansion plate, we used randomized controlled trial methodology to follow them for 3
years posttreatment. All had fulfilled our pretreatment criteria: mixed dentition, unilateral posterior crossbite,
no sucking habits, and no previous orthodontic treatment. Transverse relationships, maxillary and mandibular
widths, overbite, overjet, arch lengths, and midlines were registered on the study models immediately before
and after treatment and at the follow-up 3 years after treatment. The matched control group comprised 20
subjects with normal occlusion and was compared with the first and last registrations for the treated groups.
Results: At follow-up, changes in the treatment groups were equal and stable. The changes were
comparable with the control group. All other changes were minor and had no clinical implications. The long-
term effect of crossbite correction on midline deviation was unpredictable. Conclusions: If crossbite is success-
fully corrected by the quad-helix appliance or the expansion plate, similar long-term stability is achieved. How-
ever, in treated patients, mean maxillary widths never reached those of normal control subjects. (Am J Orthod
Dentofacial Orthop 2011;139:e73-e81)

P
osterior crossbite is a common malocclusion in
the deciduous and mixed dentitions, with preva-
lence rates of 7.5% to 22%.1-7 It has been
claimed that unilateral posterior crossbite should be
treated early to prevent negative long-term effects on
growth and development of the teeth and jaws, such
as disturbance of temporal and masseter muscle activity
in children8-10 and increased risk of craniomandibular
disorders in adolescents.10-17
Two approaches for correction of posterior crossbite
are rapid maxillary expansion and slow expansion, with
a quad-helix appliance or a removable expansion
From the Faculty of Odontology, Malm€ o University, Malm€
o, Sweden.
a
Consultant orthodontist and research fellow, Department of Orthodontics.
b
Head and associate professor, Department of Orthodontics.
c
Dean and professor.
The authors report no commercial, proprietary, or financial interest in the
products or companies described in this article.
Supported by the Swedish Dental Society; Skane County Council, Sweden; and
the Faculty of Odontology, Malm€ o University, Malm€o, Sweden.
Reprint requests to: Sofia Petr
en, Malm€o University, Department of Orthodon-
tics, Faculty of Odontology, S-205 06 Malm€ o, Sweden; e-mail, sofia.petren@
mah.se.
Submitted, March 2010; revised and accepted, June 2010.
0889-5406/$36.00
Copyright Ó 2011 by the American Association of Orthodontists.
doi:10.1016/j.ajodo.2010.06.018
plate.18-22 Two recent systematic reviews concluded
that there is inadequate evidence in support of any
method of crossbite correction in the mixed dentition,
with a need for additional randomized clinical trials
(RCTs).23,24 There is only 1 published RCT comparing
and evaluating the effectiveness of different treatment
strategies for crossbite correction in the mixed
dentition.25 It was a short-term study and clearly dem-
onstrated that the quad-helix appliance was superior
to the expansion plate.
The basic goal of orthodontic treatment is to produce
a normal occlusion that is morphologically stable and
functionally and esthetically well adjusted. However,
treatment outcomes vary, depending on the type and se-
verity of malocclusion, treatment approach and timing,
patient compliance and growth, and adaptability of
the hard and soft tissues. Follow-up studies of treated
patients have shown that, although normal occlusion
and dental alignment have been achieved, there is a ten-
dency for posttreatment relapse.26-29 Crossbite correction
is undertaken in the growing child: hence,
posttreatment changes must be evaluated in the
context of changes associated with normal growth in
children who have had no orthodontic treatment. It
e73
e74 Petr
en, Bjerklin, and Bondemark

Fig 1. Flow chart.

is therefore appropriate and advisable to use control

groups with normal occlusion. Table I. Sexes, mean ages, standard deviations (y) at
There are few published studies of the long-term ef- baseline (T0), after treatment (T1), and at follow-up
fects of crossbite correction22,30,31 and only 1 comparing (T2)
the expansion-plate and the quad-helix methods.22 Quad-helix group Expansion-plate group Normal group
Our aims in this study were, using RCT methodology, (n 5 20) (n 5 15) (n 5 20)
to compare and evaluate the long-term stability in pa- 11 girls, 9 boys 10 girls, 5 boys 9 girls, 11 boys
tients who had crossbite correction with quad-helix ap-
pliances and expansion plates, with a matched control Mean SD Mean SD Mean SD
group with normal occlusion. It was hypothesized that T0 9.0 1.19 8.5 1.02 8.8 0.5
T1 10.5 1.31 10.2 1.11 x x
the 2 treatment methods achieve similar changes of T2 13.8 2.0 13.1 1.29 13.9 1.59
a magnitude comparable with those in subjects with
normal occlusion. x, No registration.

MATERIAL AND METHODS
Originally, the study comprised 60 subjects (33 girls,
27 boys): 40 with unilateral posterior crossbite and 20
controls with normal occlusion. All patients were in-
formed of the purpose of the trial and were required to
give written, informed consent before enrollment.
Twenty of the crossbite patients had been treated
with quad-helix appliances and 20 with expansion
plates. Most (n 5 30) of the crossbite patients were
recruited from an RCT study.25 Five patients in the
expansion-plate group failed to complete the study be-
cause of noncompliance and were excluded from the
long-term follow-up. The patient flow is illustrated in
Figure 1. Sample size, sex, and age distribution of the
subjects are given in Table I. Consequently, the follow-
up study comprised 55 patients (30 girls, 25 boys).
All crossbite patients were consecutively recruited
from 2 public dental clinics in the province of Scania,

January 2011
Vol 139
Issue 1 American Journal of Orthodontics and Dentofacial Orthopedics
Petr
en, Bjerklin, and Bondemark
Fig 2. Transversal linear measurements on the study
casts.
Sweden, and from the Department of General Dentistry,
Faculty of Odontology, Malm€ o University, Malm€ o,
Sweden, from 2001 through 2006. The patients met
the following inclusion criteria: mixed dentition (all in-
cisors and first molars in occlusion), unilateral posterior
crossbite, no sucking habits or sucking habit discontin-
ued at least 1 year before the trial, and no previous
orthodontic treatment.
The normal control group was recruited from the In-
stitute for Postgraduate Dental Education, J€ onk€oping,
Sweden. These subjects had normal sagittal occlusion
and no crossbite or other malocclusion traits, and were
matched for age and dental age to the treated subjects.
The patients and their parents were given detailed
oral and written information about the trial by 5 experi-
enced general dental practitioners, who then treated the
patients under the supervision of specialist orthodon-
tists, according to a preset standard concept. The ethics
committee of Lund University, Lund, Sweden, which
American Journal of Orthodontics and Dentofacial Orthopedics
e75
Fig 3. Occlusal view of the quad-helix appliance.
follows the guidelines of the Declaration of Helsinki,
approved the informed consent form and protocol
(LU 399-00).
Study casts were made for the patients at baseline or
pretreatment (T0), posttreatment (T1), and 3 years
posttreatment (T2). For the control subjects with normal
occlusion, 2 registrations were made: at T0 and T2
(at least 4 years after T0).
For the treated patients, the main outcome to be as-
sessed was the success rate of crossbite correction. For
both the treated and normal control subjects, the mea-
surements included maxillary and mandibular interca-
nine and intermolar changes at the shortest linear
distance at the gingival margins and the cusp tips of
the teeth (Fig 2). Overbite, overjet, midline deviation,
and arch length in both jaws were registered. Midline
deviation was defined as the discrepancy between the
maxillary and mandibular midlines.
Intercanine and intermolar measurements were
made to the nearest 0.1 mm, by using a digital sliding
caliper (digital 6, 8M007906, Mauser, Winterthur, Swit-
zerland). One orthodontist (S.P.) undertook all measure-
ments. Overbite and overjet were measured to the
nearest 0.5 mm with a stainless steel ruler. Assessment
of transverse occlusion (crossbite correction) and all
study cast measurements were blinded; ie, the examiner
was unaware of the group to which the patient be-
longed. Furthermore, the T0, T1, and T2 casts were ran-
domized for measurement.
The quad-helix appliance consisted of a standard
stainless steel arch (MIA system, 3M Unitek, Monrovia,
Calif) with stainless steel bands cemented onto the max-
illary first molars with glass ionomer cement (Fig 3). The
quad-helix was activated 10 mm before placement and
then, when necessary, reactivated every 6 weeks until
a normal transverse relationship was achieved: no over-
correction. To prevent or compensate for buccal tipping,
January 2011
Vol 139
Issue 1
e76 Petr
en, Bjerklin, and Bondemark

the appliance was adjusted for buccal root torque. The

treatment result was retained for 6 months.
The expansion plate was made of acrylic, with an
expansion screw and stainless steel clasps on the first
deciduous and permanent molars (Fig 4). The patient ac-
tivated the screw 0.2 mm once a week until a normal
transverse relationship was achieved: no overcorrection.
The patient was instructed clearly to wear the plate night
and day, except for meals and toothbrushing. The treat-
ment result was retained for 6 months.
Twenty randomly selected study casts were measured
on 2 occasions. Paired t tests disclosed no significant
mean differences between the 2 series of records. The
error of the method did not exceed 0.2 mm for any study
Fig 4. Occlusal view of the expansion plate.
variable.32

Statistical analysis
The sample size for each group had previously been
calculated and was based on a significance level of
a 5 0.05 and a power (1-b) of 90%, to detect a mean in-
tergroup difference in expansion of 2 mm (61.5 mm).25
The sample size calculation showed that each group
needed a minimum of 12 subjects. To increase the power
further and to compensate for possible attrition, it was
decided to add 10 more subjects to the follow-up study.
The data were statistically analyzed by using SPSS
software (version 16.0, SPSS, Chicago, Ill). The Fisher ex-
act test was used to calculate the success rates of cross-
bite correction and midline deviation. For numeric
variables, the arithmetic means and standard deviations
were calculated. Analysis of variance (ANOVA) with the
Tukey post-hoc test was used to compare changes
within and between groups. Differences with probabili-
ties of less than 5% (P \0.05) were considered statisti-
cally significant.
RESULTS
There were no significant differences at T0 between
the treated groups with respect to age or crossbite
side. Since no significant sex differences were found
for any of the study variables, the data for the sexes
were pooled for analysis.
The T0 measurements of the jaws are summarized in
Table II. The crossbite subjects had significantly smaller
values for all maxillary variables. With respect to the
mandibular variables—overbite, overjet, and arch
length—no intergroup differences were found, except
that the arch lengths of quadrants 2 and 4 in the control
group were greater than in the crossbite groups. The
prevalences of midline deviation at T0 were 19 of 20
in the quad-helix group, 14 of 15 in the expansion-
plate group, and 10 of 20 in the normal group.
The changes during the treatment period (T0-T1) are
summarized in Table III. The maxillary intermolar and
intercanine distances increased significantly in both
treatment groups. Mandibular intermolar expansion
was significantly greater in the expansion-plate group
than in the quad-helix group. There were no intergroup
differences with respect to overbite or overjet. The arch
length in quadrant 3 was greater in the expansion-
plate group than in the quad-helix group. After treat-
ment, a correct midline was achieved in more than half
the crossbite patients, but there was no significant
inter-group difference.
At 3 years posttreatment, all 15 patients in the
expansion-plate group and 19 of the 20 patients in
the quad-helix group had normal transverse relation-
ships; ie, the treatment outcome was longitudinally
stable. However, during the posttreatment period,
significant decreases in maxillary and mandibular trans-
verse dimensions occurred in the treatment groups.
There were no significant intergroup differences, except
for mandibular intermolar distance at the gingival mar-
gin, which decreased more in the expansion-plate group
(Table IV). The corrected midlines in the treatment
groups relapsed in some patients, but there were also
a few spontaneous corrections. Moreover, during the
observation period, both spontaneous midline correc-
tions and deviations occurred in the control group.
The overall changes during the 4-year study period
(T0-T2) are shown in Table V. Significant increases in
maxillary transverse distances were found in all groups
(P\0.01), except for intercanine distance at the gingival
margin. There was no significant intergroup difference
in expansion, except for maxillary intermolar cusp tip
distance, which was greater in the treatment groups
than in the control group. In the mandible, there were
small, albeit significant, differences in the groups, but

January 2011
Vol 139
Issue 1 American Journal of Orthodontics and Dentofacial Orthopedics
Petr
en, Bjerklin, and Bondemark e77

Table II. Baseline measurements (mm) for the quad-helix (A), expansion-plate (B), and normal (C) groups
Quad-helix (A) 95% CI Expansion plate (B) 95% CI Normal group (C) 95% CI
n 5 20 for mean n 5 15 for mean n 5 20 for mean

Group
Baseline records Mean SD Lower Upper Mean SD Lower Upper Mean SD Lower Upper difference
Maxilla
Intermolar distance, 29.7 2.44 28.5 30.8 29.9 2.3 28.6 31.2 33.6 2.72 32.3 34.9 C.A,Bz
gingival margin
Intermolar distance, 47.1 2.62 45.9 48.3 46.7 2.39 45.3 48 50.6 2.49 49.4 51.7 C.A,By
mesiobuccal cusp tips
Intercanine distance, 22.8 1.29 22.2 23.4 22.8 1.78 21.8 23.7 25 2.35 23.9 26.1 C.Ay; C.By
gingival margin
Intercanine distance, 29.4 1.82 28.5 30.2 30.1 2.13 29 31.3 32.5 2.31 31.4 33.6 C.Az; C.By
buccal cusp tips
Mandible
Intermolar distance, 33.4 2.39 32.3 34.5 33.2 2.49 31.8 34.6 33.1 1.54 32.4 33.8 NS
gingival margin
Intermolar distance, 45.6 2.9 44.2 46.9 45.2 2.88 43.7 46.8 44.9 2.15 43.9 45.9 NS
mesiobuccal cusp tips
Intercanine distance, 20.7 1.6 20 21.5 20.7 1.38 19.8 21.5 21 1.5 20.3 21.7 NS
gingival margin
Intercanine distance, 26 2 25.1 27 26.3 1.3 25.5 27.1 26.9 1.5 26.1 27.6 NS
buccal cusp tips
Overjet 2.9 1.33 2.3 3.5 3.2 1.5 2.3 4 4 2.4 2.9 5.1 NS
Overbite 2.4 1.31 1.8 3 2.1 1 1.5 2.6 2.3 2.1 1.4 3.4 NS
Arch length quadrant 1 36.9 1.95 36 37.8 37 1.25 36.4 37.7 38 2.66 36.8 39.3 NS
Arch length quadrant 2 37.3 2.1 36.3 38.3 36.7 1.16 36 37.3 38.6 1.62 37.8 39.3 C.By
Arch length quadrant 3 35.2 1.49 34.5 35.9 35 1.26 34.3 35.7 35.3 1.64 34.5 36.1 NS
Arch length quadrant 4 34.3 1.94 33.5 35.3 35.3 1.24 34.6 35.9 35.6 1.36 35 36.3 C.A*
*P \0.05; yP \0.01; zP \0.001; NS, not significant (ANOVA).

no intergroup differences except for intermolar distance
at the gingival margin, which was greater in the control
group than in the treated groups.
For overbite, overjet, and arch length, there were no
differences within or between groups. In all 3 groups, the
mandibular midline changed in most subjects, some
during treatment and some during the posttreatment
period. The changes fluctuated in both directions—
from correct to incorrect and vice versa. No significant
differences were found between the groups.
At T2, the transverse distances for all maxillary vari-
ables remained significantly smaller in both treatment
groups than in the control group (Table VI). In the man-
dible, there were no differences between the groups.
Arch length in both jaws on the right side (quadrants 1
and 4) was greater at T2 in the control group than in
the treatment groups.
DISCUSSION
The results of this RCT study confirm the hypotheses
that at follow-up the changes in the treated groups were
similar and of a magnitude comparable with those in the
control group. Thus, this study clearly confirms that if
crossbite is successfully corrected by the quad-helix ap-
pliance or the expansion plate, similar long-term stabil-
ity is achieved, and the prognosis is favorable. However,
at the end of the follow-up period, the maxillary trans-
verse distances were still significantly smaller in both
treatment groups compared with the control group.
Thus, despite active transverse expansion, the width of
the maxilla in a former crossbite patient group never
reaches the mean maxillary width of a normal group.
No overcorrection was carried out in any of the sub-
jects; ie, active treatment ceased when normal transverse
relationships were achieved. Thus, overcorrection might
be unnecessary, since crossbite correction without over-
expansion still was found to be stable in a long-term
perspective.
Furthermore, it can be stated that the effects of
crossbite correction on mandibular width, overbite,
overjet, and arch length were only minor, with no clinical
implications.
Midline deviation is frequently seen in patients with
posterior crossbite because of a forced mandibular shift.
In this study, maxillary expansion enhanced the midline
correction, but the long-term effects were unpredict-
able. One explanation is that midlines can vary over

American Journal of Orthodontics and Dentofacial Orthopedics January 2011
Vol 139
Issue 1
e78 Petr
en, Bjerklin, and Bondemark

Table III. Differences in intermolar distance, intercanine distance, overjet, overbite, and arch length (mm) within and
between groups
Quad5helix (A) 95% CI Expansion plate (B) 95% CI
n 5 20 for mean n 5 15 for mean

Group
Difference records T1-T0 Mean SD Lower Upper Mean SD Lower Upper difference
Maxilla
Intermolar distance, gingival margin 3.7z 1.58 2.9 4.4 3.2z 1.24 2.5 3.9 NS
Intermolar distance, mesiobuccal cusp tips 4.1z 1.45 3.5 4.8 3.8z 1.62 2.9 4.7 NS
Intercanine distance, gingival margin 1.5y 1.64 0.7 2.3 2.4z 1.44 1.5 3.3 NS
Intercanine distance, buccal cusp tips 2.7z 1.57 1.9 3.4 2.6z 1.58 1.7 3.6 NS
Mandible
Intermolar distance, gingival margin 0.4 0.82 0.7 0 0.4* 0.67 0 0.8 B.Ay
Intermolar distance, mesiobuccal cusp tips 0.1 1.4 0.8 0.5 1.2z 0.7 0.8 1.5 B.Ay
Intercanine distance, gingival margin 0.2 1.05 0.3 0.8 0.6 1.63 0.6 1.7 NS
Intercanine distance, buccal cusp tips 0.5 1.21 1.1 0.1 0.5 1.42 0.5 1.5 NS
Overjet 0.1 0.83 0.5 0.3 0 1.32 1.2 0.3 NS
Overbite 0.2 0.92 0.3 0.6 0.6 1.03 0 1.1 NS
Arch length quadrant 1 0.5 1.05 0 1 0.6 1.32 0.1 1.3 NS
Arch length quadrant 2 0 1.91 0.9 0.9 1y 0.87 0.5 1.5 NS
Arch length quadrant 3 0.7* 1.24 0.1 0.1 0.4 0.93 0.1 0.9 B.Ay
Arch length quadrant 4 0.3 1.76 1.1 0.5 0.2 0.91 0.7 0.3 NS
*P \0.05; yP \0.01; zP \0.001; NS, not significant (ANOVA).

Table IV. Differences in intermolar distance, intercanine distance, overjet, overbite, and arch length (mm) within and
between groups
Quad-helix (A) 95% CI Expansion plate (B) 95% CI
n 5 20 for mean n 5 15 for mean

Group
Difference records T2-T1 Mean SD Lower Upper Mean SD Lower Upper difference
Maxilla
Intermolar distance, gingival margin 0.8* 1.48 1.5 1.5 0.6 1.14 1.2 0 NS
Intermolar distance, mesiobuccal cusp tips 0.8 1.7 1.5 0 0.4 1.33 1.1 0.4 NS
Intercanine distance, gingival margin 1.2z 1.19 1.8 0.6 1.4* 1.54 2.4 0.3 NS
Intercanine distance, buccal cusp tips 0.4 1.67 0.4 1.3 0.2 1.09 0.5 0.9 NS
Mandible
Intermolar distance, gingival margin 0.2 0.92 0.7 0.2 1y 1.15 1.6 0.4 B.A*
Intermolar distance, mesiobuccal cusp tips 0.5 1.2 1.04 0 1.3y 1.16 1.9 0.6 NS
Intercanine distance, gingival margin 1y 1.1 1.6 0.4 1.8* 1.4 2.8 0.9 NS
Intercanine distance, buccal cusp tips 0.1 1.29 0.6 0.8 0.7 1.35 1.6 0.2 NS
Overjet 0.4 1 0 0.9 0.3 0.7 0.1 0.7 NS
Overbite 0.5 1.18 0 1.1 0.5 1.1 0.1 1.1 NS
Arch length quadrant 1 0.6 1.19 1.2 0.1 0.8 2.15 2 0.4 NS
Arch length quadrant 2 0.1 1.28 0.7 0.5 0.3 1.04 0.9 0.3 NS
Arch length quadrant 3 1.2 1.62 1.9 0.4 1.7 1.42 2.5 1 NS
Arch length quadrant 4 1.2 1.32 1.8 0.6 1.5 1.13 2.1 0.9 NS
*P \0.05; yP \0.01; zP \0.001; NS, not significant (ANOVA).

time during periods of growth and tooth eruption. This
was also found in the normal group.
Despite several studies of crossbite correction23-25 in
recent years, to date only 1 compared the long-term ef-
fects of the quad-helix appliance and the expansion plate
in the mixed dentition.22 Our study confirmed earlier
findings by Bjerklin22 that the maxillary width of the
treated crossbite patients never reached the same width
as the normal control subjects. On the other hand, we
found no differences in the success rates between the
treatment groups, whereas Bjerklin reported that treat-
ment with the expansion plate was somewhat more stable.

January 2011
Vol 139
Issue 1 American Journal of Orthodontics and Dentofacial Orthopedics
Petr
en, Bjerklin, and Bondemark e79

Table V. Differences in intermolar distance, intercanine distance, overjet, overbite, and arch length (mm) within and
between groups
Quad-helix (A) 95% CI Expansion plate (B) 95% CI Normal group 95% CI
n 5 20 for mean n 5 15 for mean n 5 20 for mean

Group
Difference records T2-T0 Mean SD Lower Upper Mean SD Lower Upper Mean SD Lower Upper difference
Maxilla
Intermolar distance, 2.8z 1.71 2 3.6 2.6z 1.19 1.9 3.3 2z 1.66 1.2 2.8 NS
gingival margin
Intermolar distance, 3.4z 1.38 2.7 4 3.5z 1.19 2.8 4.1 1.9z 1.58 1.2 2.6 A,B.Cy
mesiobuccal cusp tips
Intercanine distance, 0.2 1.88 0.7 1.1 0.6 2.25 0.8 2.1 0.2 2.12 1.2 0.8 NS
gingival margin
Intercanine distance, buccal 3.2z 2.28 2.1 4.3 2.5z 1.68 1.5 3.5 1.6y 2.38 0.5 2.7 NS
cusp tips
Mandible
Intermolar distance, 0.6* 1 1.1 0.1 0.6 1.2 1.2 0 0.5* 1.03 0 1 C.A,By
gingival margin
Intermolar distance, 0.6 1.33 1.2 0 0.1 1.2 0.8 0.6 0 1.32 0.6 0.7 NS
mesiobuccal cusp tips
Intercanine distance, 1y 1.16 1.6 0.4 1.3y 1.19 2 0.5 1.6z 1.16 2.1 1.1 NS
gingival margin
Intercanine distance, buccal 0.7 1.6 1.5 0.2 0.4 1.12 1.1 0.2 0.4 1.64 1.1 0.4 NS
cusp tips
Overjet 0.4 1.41 0.3 1 0.2 1.27 0.9 0.5 0.6 2.16 1.6 0.5 NS
Overbite 0.7 1.38 0 1.3 1.1 1.16 0.4 1.7 0.8 1.34 0.1 1.4 NS
Arch length quadrant 1 0.2 1.7 1 0.6 0.2 1.5 1.1 0.6 0.7 2.29 0.3 1.8 NS
Arch length quadrant 2 0 2.03 1 0.9 0.7 1.42 0.1 1.5 0.1 1.68 0.9 0.7 NS
Arch length quadrant 3 1.9 1.62 2.6 1.1 1.4 1.5 2.2 0.5 1.4 1.49 2.1 0.7 NS
Arch length quadrant 4 1.5 1.89 2.4 0.6 1.7 1.13 2.3 1.1 1.3 1.57 2 0.5 NS
*P \0.05; yP \0.01; zP \0.001; NS, not significant (ANOVA).

In a recent study, McNamara et al33 reported that
rapid maxillary expansion has a favorable effect on the
occlusal relationship. This referred to measurements of
molar relationships on cephalometric tracings; however,
overjet and overbite showed no differences between the
treated patients and the controls. Thus, these findings
agreed with our results in this study, since the effect of
crossbite correction on overjet and overbite in our study
were small and of minor clinical significance.
This follow-up study originated from a prospective
RCT. An RCT design has several advantages. The ran-
domization process diminishes the risk of error from
such factors as selection bias, the clinician’s preferred
treatment method, and the differences in the skills of
the general practitioners with respect to the various
treatment methods. Furthermore, random allocation
of subjects reduces bias and confounding variables
by ensuring that both known and unknown determi-
nants of outcome are evenly distributed among the
subjects. The prospective design also ensures that the
baseline characteristics, treatment progression, and
side effects can be strictly controlled and accurately
observed.
This study was preceded by a power analysis to
achieve a valid sample size. This ensured that the data
were not biased by loss of data. The power analysis
was also used to achieve a sufficient sample size and,
considering ethical aspects, not to incorporate more pa-
tients than necessary in the study. Moreover, to reduce
the risk of bias, measurement of the study models was
blinded; the examiner was unaware of the patients’
groups. Thus, the design and methodology ensured
good external validity of the results.
It is claimed that rapid maxillary expansion increases
dental space in the jaws.18,34 We found no increase in
maxillary arch length in the treatment groups. Thus,
the study does not support the assumption that
crossbite correction by quad-helix or expansion-plate
treatment will increase the available tooth space.
This study concerned slow maxillary expansion. A fu-
ture study approach on evaluation of slow vs rapid
expansion would be of great interest and is underway.
In this study, former crossbite patients were com-
pared with normal control subjects. Ideally, the study
should have included an untreated control group of sub-
jects with posterior crossbites to evaluate the potential

American Journal of Orthodontics and Dentofacial Orthopedics January 2011
Vol 139
Issue 1
e80 Petr
en, Bjerklin, and Bondemark

Table VI. Mean values and standard deviations (mm) of the 3 groups at T2
Quad-helix (A) 95% CI Expansion plate (B) 95% CI Normal group (C) 95% CI
n 5 20 for mean n 5 15 for mean n 5 20 for mean

Group
Records at T2 Mean SD Lower Upper Mean SD Lower Upper Mean SD Lower Upper difference
Maxilla
Intermolar distance, 32.5 2.67 31.2 33.7 32.5 2.2 31.3 33.7 35.6 2.6 34.4 36.8 C.Az;C.By
gingival margin
Intermolar distance, 50.5 2.89 49.1 51.8 50.1 2.61 48.7 51.6 52.5 2.23 51.4 53.5 C.A,B*
mesiobuccal cusp tips
Intercanine distance, 23 1.83 22.1 23.9 23.5 2.12 22.1 24.8 24.8 2.02 23.8 25.7 C.A*
gingival margin
Intercanine distance, 32.5 2.52 31.3 33.7 32.7 1.65 31.7 33.7 34.2 1.99 33.2 35 C.A*
buccal cusp tips
Mandible
Intermolar distance, 32.8 2.72 31.6 34.1 32.6 2.5 31.2 34 33.6 2.02 32.7 34.6 NS
gingival margin
Intermolar distance, 45 3.17 43.5 46.4 45.1 2.79 43.6 46.6 45 2.69 43.7 46.2 NS
mesiobuccal cusp tips
Intercanine distance, 19.6 1.77 18.7 20.5 19.2 1.86 18.1 20.3 19.4 1.27 18.8 20 NS
gingival margin
Intercanine distance, 25.3 2.21 24.1 26.4 25.8 1.42 24.9 26.6 26.4 1.65 25.7 27.3 NS
buccal cusp tips
Overjet 3.3 1.21 2.7 3.8 3 0.98 2.5 3.5 3.5 1.6 2.7 4.2 NS
Overbite 3.1 1.3 2.5 3.7 3.1 1.45 2.3 3.9 3.1 1.6 2.4 3.9 NS
Arch length quadrant 1 36.8 1.8 35.9 37.6 36.8 1.6 35.9 37.7 38.8 1.63 38 39.5 C.Az;C.By
Arch length quadrant 2 37.3 1.95 36.3 38.2 37.4 1.91 36.3 38.4 38.5 1.85 37.6 39.3 NS
Arch length quadrant 3 33.3 1.8 32.5 34.2 33.6 1.69 32.7 34.5 33.9 1.37 33.3 34.6 NS
Arch length quadrant 4 32.9 1.63 32.1 33.6 33.6 1.55 32.7 34.4 34.4 1.58 33.6 35.1 C.A*
*P \0.05; yP \0.01; zP \0.001; NS, not significant (ANOVA).

impact of the condition on long-term growth. However, 3. There was no clinically significant correlation be-
postponement of a needed intervention for 4 years is tween maxillary expansion and changes in overbite,
ethically unacceptable. overjet, or arch length.
In our opinion, the follow-up period of 3 years was 4. Maxillary expansion might enhance midline
adequate for long-term conclusions, because, at T2, correction, but the long-term effects are unpre-
a complete permanent dentition was established in al- dictable.
most all subjects. On the other hand, these promising re-
sults raise further questions. Having established that the We thank the staff of the public dental clinics of Oxie
2 treatment strategies are equally effective with respect and Trelleborg, Sweden, for their valuable assistance
to clinical outcomes, other aspects now warrant investi- with the clinical procedures.
gation. A comparative study of the cost-effectiveness of
the 2 methods is currently in progress. REFERENCES

1. Kutin G, Hawes RR. Posterior cross-bites in the deciduous and

CONCLUSIONS
1. The long-term stability of crossbite correction in the
mixed dentition is favorable, and the quad-helix
and expansion-plate methods achieve similar
results.
2. Despite active transverse expansion, the average
posttreatment width of the maxilla in the former
crossbite patients never reached that of the normal
control subjects.
mixed dentitions. Am J Orthod 1969;56:491-504.
2. Thilander B, Myrberg N. The prevalence of malocclusion in Swed-
ish schoolchildren. Scand J Dent Res 1973;81:12-20.
3. Heikenheimo K, Salmi K. Need for orthodontic intervention in five-
year-old Finnish children. Proc Finn Dent Soc 1987;83:165-9.
4. B€assler-Zeltmann S, Kretschmer I, G€oz G. Malocclusion and the
need for orthodontic treatment in 9-year-old children. J Orofac
Orthop 1998;59:193-201.
5. Keski-Nisula K, Lehto R, Lusa V, Keski-Nisula L, Varella J. Occur-
rence of malocclusion and need of orthodontic treatment in early
mixed dentition. Am J Orthod Dentofacial Orthop 2003;124:
631-8.

January 2011
Vol 139
Issue 1 American Journal of Orthodontics and Dentofacial Orthopedics
Petr
en, Bjerklin, and Bondemark
6. Tausche E, Luck O, Harzer W. Prevalence of malocclusions in the
early mixed dentition and orthodontic treatment need. Eur J
Orthod 2004;26:237-44.
7. da Silva Filho OG, Santamaria M Jr, Capelozza Filho L. Epidemiol-
ogy of posterior crossbite in the primary dentition. J Clin Pediatr
Dent 2007;32:73-8.
8. Troelstrup B, M€ oller E. Electromyography of the temporalis and
masseter muscles in children with unilateral cross-bite. Scand J
Dent Res 1970;8:425-30.
9. Ingervall B, Thilander B. Activity of temporal and masseter muscles
in children with a lateral forced bite. Angle Orthod 1975;45:
249-58.
10. Thilander B, Rubio G, Pena L, de Mayorga C. Prevalence of tempo-
romandibular dysfunction and its association with malocclusion in
children and adolescents: an epidemiologic study related to spec-
ified stages of dental development. Angle Orthod 2002;72:
146-54.
11. Egermark-Eriksson I, Carlsson GE, Magnusson T, Thilander B.
A longitudinal study on malocclusion in relation to signs and
symptoms of cranio-mandibular disorders in children and adoles-
cents. Eur J Orthod 1990;12:399-407.
12. Egermark I, Magnusson T, Carlsson G. A 20-year follow-up of
signs and symptoms of temporomandibular disorders and maloc-
clusions in subjects with and without orthodontic treatment in
childhood. Angle Orthod 2003;73:109-15.
13. Mohlin B, Thilander B. The importance of the relationship between
malocclusion and mandibular dysfunction and some clinical appli-
cations in adults. Eur J Orthod 1984;6:192-204.
14. Riolo ML, Brandt D, TenHave TR. Associations between occlusal
characteristics and signs and symptoms of TMJ dysfunction in
children and young adults. Am J Orthod Dentofacial Orthop
1987;92:467-77.
15. Pullinger AG, Seligman DA, Gornbein JA. A multiple logistic re-
gression analysis of the risk and relative odds of temporomandib-
ular disorders as a function of common occlusal factors. J Dent Res
1993;72:968-79.
16. Lambourne C, Lampasso J, Buchanan WC, Dunford R, McCall W.
Malocclusion as a risk factor in the etiology of headaches in chil-
dren and adolescents. Am J Orthod Dentofacial Orthop 2007;132:
754-61.
17. Sonnessen L, Bakke M, Solow B. Malocclusion traits and symptoms
and signs of temporomandibular disorders in children with severe
malocclusion. Eur J Orthod 1998;20:543-59.
18. Proffit WR. Treatment of orthodontic problems in preadolescent
children (section VI). In: Proffit WR, editor. Contemporary ortho-
dontics. 3rd ed. St Louis: Mosby; 2000. p. 435-9.
19. Boysen B, La Cour K, Athanasiou AT, Gjessing PE. Three-dimen-
sional evaluation of dentoskeletal changes by quad-helix or
removable appliances. Br J Orthod 1992;19:97-107.
American Journal of Orthodontics and Dentofacial Orthopedics
e81
20. Sandikcioglu M, Hazar S. Skeletal and dental changes after maxil-
lary expansion in the mixed dentition. Am J Orthod Dentofacial
Orthop 1997;111:321-7.
21. Erdinc A, Ugur T, Erbay E. A comparison of different treatment
techniques for posterior crossbite in the mixed dentition. Am J
Orthod Dentofacial Orthop 1999;116:287-300.
22. Bjerklin K. Follow-up control of patients with unilateral posterior
cross-bite treated with expansion plates or the quad-helix appli-
ance. J Orofac Orthop 2000;61:112-24.
23. Harrison JE, Ashby D. Orthodontic treatment for posterior cross-
bites. Cochrane Database Systematic Review 2001;1:CD000979.
DOI: 10.1002/14651858.CD000979. Available at: http://www.
cochrane.org/reviews/en/ab000979.html. Date of most recent
amendment: November 16, 2000. Accessed on February 1, 2010.
24. Petr
en S, S€
oderfeldt B, Bondemark L. A systematic review concern-
ing early orthodontic treatment of unilateral posterior crossbite.
Angle Orthod 2003;73:588-96.
25. Petr
en S, Bondemark L. Correction of unilateral posterior crossbite
in the mixed dentition: a randomized controlled trial. Am J Orthod
Dentofacial Orthop 2008;133:790.e7-13.
26. McNamara JA Jr, Baccetti T, Franchi L, Herberger TA. Rapid max-
illary expansion followed by fixed appliances: a long-term evalua-
tion of changes in arch dimensions. Angle Orthod 2003;73:
344-53.
27. Little RM, Riedel RA, 
Artun J. An evaluation of changes in mandib-
ular anterior alignment from 10 to 20 years postretention. Am J
Orthod Dentofacial Orthop 1988;93:423-8.
28. McReynolds DC, Little RM. Mandibular second premolar extrac-
tion—postretention evaluation of stability and relapse. Angle
Orthod 1991;61:133-44.
29. 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.
30. Bartzela T, Jonas I. Long-term stability of unilateral posterior
crossbite correction. Am J Orthod Dentofacial Orthop 2007;77:
237-43.
31. Huynh T, Kennedy D, Joondeph D, Bollen AM. Treatment response
and stability of slow maxillary expansion using Haas, hyrax, and
quad-helix appliances: a retrospective study. Am J Orthod Dento-
facial Orthop 2009;136:331-9.
32. Dahlberg G. Statistical methods for medical and biological
students. London, United Kingdom: Allen and Unwin; 1940.
p. 122-32.
33. McNamara J, Sigler LM, Franchi L, Guest S, Baccetti T. Changes in
occlusal relationships in mixed dentition patients treated with
rapid maxillary expansion. Angle Orthod 2010;80:230-8.
34. McNamara JA. Early intervention in the transverse dimension: is it
worth the effort? Am J Orthod Dentofacial Orthop 2002;121:
572-4.
January 2011
Vol 139
Issue 1