ORIGINAL ARTICLE

Long- and short-term effects of headgear traction

with and without the maxillary second molars
Nir Shpack,a Tamar Brosh,b Yoav Mazor,c Yoav Shapinko,d Moshe Davidovitch,e Rachel Sarig,e
Susanne Reimann,f Christoph Bourauel,g and Alexander D. Vardimonh
Tel Aviv, Israel, and Bonn, Germany

Introduction: A quantitative assessment of maxillary rst molar distalization with and without the maxillary sec-
ond molar (M2) was carried out. Methods: Fifty-six cervical headgear patients undergoing xed appliance ortho-
dontic treatment were divided into 2 groups: before (G M2) and after (G 1 M2) eruption of the maxillary second
molars (ages, 11.87 6 1.20, and 13.05 6 1.55 years, respectively). The tightness of the dental contact point
(TDCP) and the space between the second premolar and the maxillary rst molar were measured at 6 levels
of headgear force (0-15 N) at 3 intervals 6 months apart (T0, T1, T2). Results: Relationships were found be-
tween space and TDCP, time, and presence or absence of the maxillary second molar at T1 and T2
(P \0.001). The TDCP decreased and space increased with increase in initial headgear force. An increase in
initial force beyond 6 to 9 N did not signicantly increase the initial maxillary rst molar distalization. The
G M2 TDCP and space measurements were similar to those of G 1 M2 at T2 with the eruption of the maxillary
second molar. From T0 to T1, maxillary rst molar distalization was greater in G M2. In comparison with our
previous headgear-alone study, initial distalization with a fully bonded appliance was reduced by 4-fold.
Conclusions: Headgear therapy is more effective before the eruption of the maxillary second molar. Once it
erupts, the distalization pace of the maxillary rst molar is reduced, but it can nevertheless be pursued at a slower
pace when the maxillary second molar is present. (Am J Orthod Dentofacial Orthop 2014;146:467-76)

I
t is an accepted treatment modality to direct extra-
oral forces through the maxillary permanent rst
molar to cause its distalization via headgear (HG).
However, this effect can vary considerably under
a
Chair, Department of Orthodontics, the Maurice and Gabriela Goldschleger
School of Dental Medicine, Tel Aviv University, Tel Aviv, Israel.
b
Chair, Department of Oral Biology, the Maurice and Gabriela Goldschleger
School of Dental Medicine, Tel Aviv University, Tel Aviv, Israel.
c
Private practice, Tel Aviv, Israel.
d
Instructor, Department of Orthodontics, the Maurice and Gabriela Goldschleger
School of Dental Medicine, Tel Aviv University, Tel Aviv, Israel.
e
Lecturer, Department of Orthodontics, the Maurice and Gabriela Goldschleger
School of Dental Medicine, Tel Aviv University, Tel Aviv, Israel.
f
Senior researcher, Department of Oral Technology, Center of Dento-Maxillo-
Facial Medicine, Faculty of Medicine, University of Bonn, Bonn, Germany.
g
Endowed professor, Center of Dento-Maxillo-Facial Medicine, Faculty of Medi-
cine, University of Bonn, Bonn, Germany.
h
Professor, Department of Orthodontics, the Maurice and Gabriela Goldschleger
School of Dental Medicine, Tel Aviv University, Tel Aviv, Israel.
Nir Shpack and Tamar Brosh are joint rst authors and contributed equally to this
work.
All authors have completed and submitted the ICMJE Form for Disclosure of
Potential Conicts of Interest, and none were reported.
Supported in part by a grant from the GIF (German-Israeli Foundation) (grant no.
01910911).
Address correspondence to: Nir Shpack, Department of Orthodontics, the
Maurice and Gabriela Goldschleger School of Dental Medicine, Tel Aviv Univer-
sity, Tel Aviv, Israel; e-mail, nshpack@post.tau.ac.il.
Submitted, September 2013; revised and accepted, June 2014.
0889-5406/$36.00
Copyright 2014 by the American Association of Orthodontists.
http://dx.doi.org/10.1016/j.ajodo.2014.06.015
different HG loading conditions as well as in the pres-
ence or absence of teeth distal to the maxillary perma-
nent rst molar.1 The presence of the second molar
(M2) and third molar in different developmental and
eruptive stages can also act as a physical impediment
to the extent and direction of maxillary permanent rst
molar distalization.2-5
Kloehn6 advocated early HG treatment to distalize the
maxillary teeth into correct occlusion with the mandibular
dentition. It has also been proposed that treatment of
Class II malocclusions should be performed before the
eruption of the maxillary permanent second molars and
canines, since the latter might affect treatment ef-
ciency.7,8 Previous studies with noncompliance
appliances have reported that before maxillary second
molar eruption, the maxillary rst molar can be
distalized by 1 to 2 mm per month with little anchorage
lossie, forward displacement of the anterior
segment.2,9 It has been shown when using the
pendulum appliance that once the maxillary second
molars have erupted, distal movement of the maxillary
rst molar is slowed, and anchorage loss is more
common,4 producing mesial movement of the anterior
teeth.10 Ten Hoeve2 and Jeckel and Rakosi3 concluded
that distalization of the maxillary permanent rst molar
is restrained by the maxillary second molar; consequently,
467
468
they recommended distalization before second molar
eruption. Similar ndings have been reported for several
intraoral molar distalization appliances.11,12
On the other hand, Muse et al,13 using a Wilson rapid
molar distalization appliance, and Ghosh and Nanda,14
using the pendulum appliance, found that the presence
of the maxillary second molar did not signicantly affect
the rate of rst molar movement. Additional studies with
repelling magnets15 and the pendulum appliance10,16
support the latter claims of the minor effect of the
maxillary second molar on rst molar distalization.
This diversity could be related to the interrupted force
regimen produced by HG compared with the
continuous force of noncompliance appliances.
In a previous study, we demonstrated that an incre-
mental increase in cervical HG force, applied per se,
caused differences in the gap created between the maxil-
lary rst molar and the second premolar, and that rst
molar distalization was inuenced by the absence or
presence of the maxillary second molar.2 In this study,
we investigated whether similar effects on molar distal-
ization can be expected when HG traction is applied
during full-arch xed appliance therapy and over a
longer period of time. Furthermore, we investigated
whether HG therapy is affected by short-term (initial
periodontal ligament [PDL] reaction to HG placement)
and long-term (bone remodeling) changes, and we
determined the range for the initial HG force.
With respect to the latter, few previous studies have
examined this issue, mostly describing the orthopedic
effects of diverse HG forces. For example, Zentner
et al17 studied orthopedic forces of 5.6 N (Newton 5
101.97 gram-force 5 3.597 oz) vs orthodontic forces
of 3.5 N on a macerated human skull using holographic
interferometry. They reported that lower HG forces pro-
duced signicantly greater deformations than did higher
forces; this indicates that there might be an upper limit
to the optimal force level.
Dental changes caused by HG therapy have been
reported in the literature. Andreasen and Johnson18
applied diverse forces in each side and found that over
a 12-week period the molar receiving 4 N of HG force
distalized 2.5 times more than did the other molar, which
received 2 N of force. After 8 weeks in the 2-N side, no
further increase in molar distalization was found,
whereas the 4-N side demonstrated continuation of
molar distalization at the same pace to end of the study
(12 weeks). This suggests that there might be a lower
limit to the optimal force level.
The objectives of this study were to evaluate HG
distalization efcacy concomitant with edgewise xed
appliance treatment in relation to 4 variables: the absence
or presence of the maxillary second molar, the amount of
October 2014  Vol 146  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Shpack et al
distal movement of the maxillary rst molar, the magni-
tude of the HG force, and maxillary second molar eruption.
The null hypotheses were that at all stages of
treatment, the amount of distalization of the maxillary
rst molar is not affected by the presence or absence
of the maxillary second molar, and the initial force
magnitude has no effect on initial maxillary rst molar
distal displacement.
MATERIAL AND METHODS
The study sample comprised 56 patients presenting for
treatment at the Department of Orthodontics, University
of Tel Aviv in Israel, diagnosed as having an Angle Class
II malocclusion and treated without extractions with
cervical HG for at least 1 year for a minimum of 12 hours
per day. Subjects were included in the study when part of
their malocclusion was related to maxillary dentoskeletal
protrusion (SNA, .83 ; mean, 85.9 6 2.03 ). Syndromic
patients (eg, cleft lip) were excluded, as were those with
early mesial drift of the maxillary rst molar (eg,
congenitally missing second premolar) and adults
(age, .16.5 year). The Class II severity ranged from a
half-step to a full-step molar relationship. All treatments
included the use of fully bonded edgewise appliances
without intermaxillary or intramaxillary elastics. Data
were gathered at 3 time points: T0, start of HG therapy;
T1, 6 months after T0; and T2, 12 months after T0.
The sample was divided into 2 groups: (1) G M2:
both maxillary second molars were unerupted at T0 or
showed both clinically and radiographically no contact
point between the rst and second molars (21 subjects:
9 boys, 12 girls; mean age, 11.87 6 1.20 years), and (2)
G 1 M2: at T0, both maxillary second molars were
present in the oral cavity, with both clinically and radio-
graphically an interproximal contact point between the
rst and second molars (35 subjects: 18 boys, 17 girls;
mean age, 13.05 6 1.55 years).
In addition to determining the presence or absence of
the second molars, we examined the Nolla19 stage of
maxillary second molar development radiographically.
A 0.022 3 0.028-in preadjusted appliance was
bonded to the maxillary dentition, into which was
inserted an uncinched nickel-titanium archwire. The
archwires were changed according to the progress in
leveling, from 0.014 to 0.018 in. A Kloehn type of cervical
HG was delivered to the bands on the permanent rst
molars. The HG consisted of a medium outer bow, with
U loops (3M Unitek, Monrovia, Calif); the outer arm
was positioned horizontally (with no adjusted upward
tilt). The HG force was increased gradually from 0 to
15 N in 3-N increments (0, 3, 6, 9, 12, and 15 N). Zero re-
ected the measurement when the facebow was inserted
in the mouth without attaching the (activation) neck
Shpack et al
Fig 1. Force gauge used to measure the applied HG force.
strap. Upon activation, the force created by the neck strap
was measured by a force gauge (weighing scale, spring
type WS072; Narang Medical, New Delhi, India) (Fig 1).
Displacement of the maxillary right rst molar was
measured by 2 direct methods at T0, T1, and T2. The rst
method was as a function of the change in tightness of
the contact point between the second premolar and the
rst molar using the tightness of dental contact point
(TDCP) device, according to the method of Vardimon
et al20-23 (Fig 2, A). Briey, the device measures the
amount of force required to insert a metal strip
(0.15 mm) between 2 adjacent dental contact points in
TDCP units (Fig 2, B). The TDCP was measured on the
right side and could be measured as long as the maxillary
rst molar displacement was smaller than 0.15 mm.
In the second method, custom-made metal gauges
(leaves) were inserted at the contact point of the second
premolar and the rst molar in serial fashion. The gauges
(made at workshop of the Faculty of Physics, Tel Aviv
University) differed in thickness from 0.1 to 0.6 mm, in
0.05-mm increments (totaling 11 gauges) (Fig 3). Dis-
placements less than 0.1 mm were considered to be no
displacement. Each of the 2 methods was measured while
the subject was seated in a dental chair reclined to 135 ,
and the HG force was increased gradually. The total time
for TDCP and space measurements for the 6 force mag-
nitudes was about 12 minutes. The measurements were
taken continuously. All measurements were taken by
the same investigator (Y.M.). The average HG force
worn by the patient during the 12 hours of wear was 4 N.
Statistical analysis
The random errors (calculated using the Dahlberg
formula24) and the systematic error (calculated using a
paired t test) of the 2 instruments were examined on 5
subjects from each group; the 2 sets of measurements
were taken at 1-week intervals.
The Dahlberg formula is the following.
q2
PN di
D5 i5I 2N
American Journal of Orthodontics and Dentofacial Orthopedics
469
D is the random error, di is the difference between the
rst and the second measurements, and N is the sample
size that was remeasured.
For repeated measures of the TDCP, the random error
was 0.209 N (13%); for the space measurements, it was
0.012 mm (8.6%). The larger error calculated in the
former was caused by stretching of the PDL during the
initial TDCP measurements; this affected subsequent
measurements. Therefore, only the rst TDCP measure-
ments were considered. No signicant differences were
found in the paired t tests between the rst and second
measurements, conrming the absence of systematic
error.
Analysis of variance (ANOVA) with repeated measures
with the post hoc Tukey test was used to analyze each
parameter (TDCP and space), where the HG force (0-
15 N) and time (T0, T1, T2) were within factors, and
maxillary second molar presence or absence (G 1 M2,
G M2) was the between factor. Since the TDCP values
did not show a normal distribution, the formula 1/(7
TDCP) was used to normalize the results: 7 reects the
maximum TDCP value (in Newtons) that can be regis-
tered by the TDCP device.
Pearson correlation analysis was used to analyze the
correlations between TDCP and space for each time
point at all HG forces.
RESULTS
Nolla stages of maxillary second molar development
were 7.6 6 0.5 (1/3 to 2/3 of the root was completed) for
G M2 and 9.5 6 0.5 (root completed with open or
closed apex) for G 1 M2. These values correspond to
previous studies.25,26
At T0, there was no signicant difference in TDCP be-
tween G 1 M2 and G M2. At no HG load force (0 N),
the 2 groups had similar contact point tightness. In both
groups, a nonsignicant decrease in contact point tight-
ness was found as the HG force was increased; however,
the decrease was greater in subjects without the second
molar (G M2) (Fig 4).
October 2014  Vol 146  Issue 4
470
Fig 2. A, The TDCP device for measuring interproximal
forces comprises a handle and an electronic unit; BJ,
bow jig; MS, metal strip; B, the handle includes a beam
(B) with 2 integrated strain gauges (SG), and the bow
jig with a 0.15-mm metal strip is connected to the beam;
C, intraoral TDCP (blue arrow) measurement at the con-
tact point of the second premolar and the rst molar, while
the inner arch of the HG at a given force is connected to
the rst molar.
At T1, minor to moderate corrections in the Class II
relationship were found in both groups. At no force
(0 N), the initial TDCP values were signicantly lower
than the initial values at T0 (P \0.001). The decrease
was signicantly more evident in G M2 (P \0.01)
(Fig 4).
At T2, in all subjects, the molar relationships were
improved; Class I molar relationships were almost
completed, meaning that the range of distalization
was 3 to 6 mm. The TDCP values for G M2 were signif-
icantly higher than those obtained at T1 (P\0.001). The
October 2014  Vol 146  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Shpack et al
initial TDCP value at no load decreased signicantly as
the force load increased (P \0.001) (Fig 4).
The ANOVA with repeated measures demonstrated a
signicant (P \0.001) relationship among TDCP, space,
and time in both groups (presence or absence of maxil-
lary second molars) (Table I).
At T0, at 0 N, only a small space width was measured
in both groups. The increase in space was more evident in
G M2 as the force was increased, but the overall change
was not signicant for either group (P 5 0.10) (Fig 5).
HG force elevation was effective up to 9 N for both
groups. Further increases in force did not cause an in-
crease in space.
At T1, at 0 N, the space measurement was signi-
cantly (P \0.001) higher in the G M2 than in the
G 1 M2; the increase in space width was similar in
both groups as the HG force was increased (P \0.01)
(Fig 5). The increase in space width was signicant in
both groups up to 3 N; after this, the force elevation
was not followed by a signicant space increase.
At T2, with 0 N, the space width was signicantly
smaller in G M2 than in G 1 M2 (P\0.001). However,
the space increased signicantly more at the rst force
loading (3 N) in the G M2 group. Any additional force
loading produced a nonsignicant increase in space
width in both groups (Fig 5).
The ANOVA with repeated measures demonstrated
relationships among space, time, and the presence or
absence of the maxillary second molar (P \0.001)
(Table I). An interaction was found between HG force
and the presence or absence of the maxillary second
molar and space measurements (P\0.001). The interac-
tion was well noted at T2 (Fig 5), since the effect of the
force was exhibited mainly at the low force load for the
G M2 group.
In both groups, at all time points (T0, T1, T2), the in-
crease in HG force was followed by a decrease in the
TDCP and an increase in space opening (Figs 6 and 7).
The changes in both types of measurements were similar
but inverse as the HG force increased from 0 to 15 N. In
both groups, the major change occurred at 3 N for T1
and T2 (Figs 6 and 7). At T0, in the G M2, the space
and TDCP measurements substantially increased or
decreased progressively from 0 to 9 N, respectively
(Fig 6). In contrast, in the G 1 M2, the space and
TDCP measurements moderately increased or decreased
progressively from 0 to 15 N, respectively (Fig 7). After
9 N, a plateau was found for both parameters (Fig 6).
That is, for G M2 at T0, increasing the HG force
from 9 to 15 N showed no increase or decrease in the
space and TDCP measurements, respectively (Fig 6).
Pearson correlations (Table II) demonstrated signi-
cant correlations between space and TDCP for all time
Shpack et al
Fig 3. A, Devices for measuring interproximal spaces
comprise 11 metal leaves in different thicknesses from
0.1 to 0.6 mm in 0.05-mm increments; B, space measure-
ment with a metal leaf (red arrow) at the contact point of
the second premolar and the rst molar, while the inner
arch of the HG at a given force is connected to the rst
molar.
points (T0, T1, T2) and all HG loadings (0-15 N), besides
T0 at 0 and 3 N. Negative correlations between the TDCP
and space were found as the HG force was elevated; ie,
TDCP decreased, and space increased (Table II; Figs 6
and 7). At T0, the correlation became signicant up to
6 N. Highly signicant correlations were found at T1
and T2 for all HG force magnitudes.
DISCUSSION
In this study, we investigated the effect of the pres-
ence or absence of the maxillary second molar on the
distalization of the maxillary rst molar using HG trac-
tion during edgewise xed appliance therapy, in both
the short term (PDL effect) and the long term (osseous
effect). This was measured as a function of the space
created between the maxillary rst molar and the second
premolar and evaluated using the TDCP device and
metal leaves to measure contact point tightness and
the space at the contact point of the second premolar
and the rst molar. Although the second premolar simul-
taneously drifted distally with the rst molar during HG
American Journal of Orthodontics and Dentofacial Orthopedics
471
treatment27 by the transseptal bers linkage, its drift is
less than that of the rst molar27 because of its transsep-
tal ber connection to the rst premolar. Since second
premolar movement is a by-product of rst molar move-
ment and was present in all compared groups, the dis-
cussion is interpreted with respect to the reactions of
only the rst and second molars.
The short-term effect was measured at T0, with no dif-
ferences found between groups. As HG force was increased
from 0 to 15 N, there were a nonsignicant 13% decrease
in TDCP value and a 15% increase in space width in the
G 1 M2. A similar effect but of signicantly greater magni-
tude was found in the G M2, where TDCP decreased by
25% and space increased by 31%. These ndings suggest
that the presence of the maxillary second molar has a sig-
nicant effect on rst molar distalization and cause rejec-
tion of the rst null hypothesis of this study.
In both groups, rst molar initial distalization
occurred in the PDL space per se, since the gain in space
never exceeded the maximum PDL space (150-
380 mm).28 The displacement of the rst molar in the
PDL space resulted at all levels of HG force and all 3
time points. These ndings support those by Kinzinger
et al,4 who illustrated the role of the PDL in rst molar
distalization with a pendulum appliance during different
stages of eruption of the second and third molars. That
no signicant changes were noted in the TDCP and
space values as the HG load continued to increase at
T0 can be explained by the anatomy of the PDL space.
Once this space is occluded, no further rst molar move-
ment is expected without ensuing structural changes.
This is responsible for the previously described lag
period in orthodontic tooth movement.29-32
In our previous study, we found that after initiation
of HG treatment, the maxillary rst molar was distalized
by 0.13 and 0.23 mm for HG treatment per se in G 1 M2
and G M2, respectively, compared with 0.03 and
0.05 mm for the combined HG with edgewise appliances
in the present study.5 The same amount of force was
applied in both studies; however, in the previous study
it was calculated per molar (force changes, 0-7.5 N),
and in this study it was calculated for both molars (force
changes, 0-15 N).
This implies that the initial distalization was 4 and 5
times more effective for HG treatment per se than the
combined edgewise and HG regimens. This can be ex-
plained by molar distalization associated with distal
crown tip (up to 11.7 ).10,14 Basically, every initial
movement within the PDL without guiding measures
(eg, bracket system) is characterized by an initial sharp
tipping movement.32 However, this effect can be
reduced by guiding the molar along an archwire15 or
by producing a counteracting moment.33 Moreover,
October 2014  Vol 146  Issue 4
472
Treatment me (months)
0 0
6.00 T0
6 months
5.00
P2-M1 TDCP (N)
4.00
3.00
2.00
1.00
0.00
0 3 6 9 12 15
Fig 4. Change in TDCP at the contact point of the second premolar and the rst molar at 6 HG forces
(0-15 N) applied at 6-month intervals at 3 time points (T0, T1, T2) in the 2 groups, G M2 and G 1 M2
(patients without and with second molars).
friction developed within the molar tube increases with
increasing crown tip and archwire size.34 This might
explain the decrease in the HG effect in our study.
Thus, HG treatment combined with a preadjusted
bracket system might control the bodily distal movement
of the rst molar but profoundly reduce the distalization
ability.
At T1, both short-term and long-term HG effects
were present. The long-term effect was evident after
the rst molar accomplished some distal movement dur-
ing the rst 6-month period. This was observed by the
decline in TDCP at the HG force of 0 N from T0 to T1
(15% and 50% for G 1 M2 and G M2, respectively).
Thus, after 6 months of HG treatment, more initial space
was obtained in G M2, which was measured as a
3-fold greater TDCP decline than in G 1 M2. That is,
the initial space at the contact point of the second pre-
molar and the rst molar signicantly increased after
6 months of HG traction from T0 to T1 at no force
load (0 N) by 70.5% in G M2 and by 35.7% in G 1 M2.
Comparing the long-term effect of HG with xed ap-
pliances (present study) vs HG alone (previous study)
demonstrated that initial distalizations of 0.2 mm for
G M2 and 0.13 mm for G 1 M2 were accomplished
after 6 months of treatment in the former and 3 months
in the latter.5 That is, a 2-fold extension in time is
needed for the combined HG and brackets treatment
to accomplish the same initial distalization as with
October 2014  Vol 146  Issue 4
Shpack et al
6 6 12 12
T1 T2
6 months
G+M
M TDCP
G-M
M TDCP
0 3 6 9 12 15 0 3 6 9 12 15
Head gear force (N)
HG only. Because this study was conducted over a
3-month period, we cannot compare HG treatment per
se with 1 year of treatment of HG with brackets (T2).5
Most likely, in spite of the low force magnitude that light
nickel-titanium archwires exert, the counterclockwise
moment acting by these archwires on the rst molar
(counteracting the clockwise moment: ie, distal tip of
the HG) was effectual in decreasing the initial pace of
distalization, probably due to the time factor (24-hour
archwire vs 12-hour HG). Additionally, the friction pro-
duced by the archwire is not present when HG is used.
At T1, the short-term effect was found to be similar
in the 2 groups. The peak change occurred at 3 N of HG
initial force for both TDCP and space. The pattern of
change in TDCP and space with the increase in HG force
was the same for both groups; ie, a decrease in TDCP
was highly correlated with an increase in space (R,
approximately 0.7). The fact that TDCP values were
signicantly lower and space values higher in G M2
in comparison with G 1 M2 is a product of the long-
term effect: ie, the already accomplished rst molar
distal movement during the 6-month period that is
characterized by bone remodeling and PDL widening,
which was more distinct in G M2.29-31 Although it
was not measured, most likely the distal relocation of
the rst molar is not a pure translation; in the absence
of the second molar, more tipping of the rst molar
was expected.4
American Journal of Orthodontics and Dentofacial Orthopedics
Shpack et al 473

Table I. TDCP and space (mm) measurements with progressive increases in HG force (015 N) at 3 times and post hoc
analysis (Tukey)
Group Time 0 3 6 9 12 15
TDCP G 1 M2 T0 5.66 6 0.18 5.55 6 0.19 5.40 6 0.19 5.32 6 0.20 5.20 6 0.20 4.97 6 0.19
T1 4.78 6 0.27 3.95 6 0.24 3.72 6 0.25 3.43 6 0.23 3.31 6 0.23 3.17 6 0.22
T2 4.43 6 0.27 3.76 6 0.25 3.50 6 0.24 3.24 6 0.23 3.11 6 0.23 2.84 6 0.20
Tukey T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2
signicance P \0.001 P \0.001 P \0.001 P \0.001 P \0.001 P \0.001
G M2 T0 5.47 6 0.35 5.24 6 0.36 4.82 6 0.36 4.36 6 0.34 4.28 6 0.32 4.21 6 0.32
T1 2.69 6 0.53 1.98 6 0.42 1.83 6 0.40 1.77 6 0.38 1.67 6 0.37 1.65 6 0.36
T2 5.19 6 0.24 4.01 6 0.21 3.66 6 0.18 3.49 6 0.17 3.45 6 0.17 3.03 6 0.17
Tukey T0, T2s T1 T0 s T1 s T2 T0 s T1 s T2 T0, T2 s T1 T0, T2 s T1 T0 s T1 s T2
signicance P 5 0.001 P 5 0.02 P 5 0.03 P 5 0.002 P 5 0.001 P 5 0.008
Space (mm) G 1 M2 T0 0.14 6 0.01 0.15 6 0.01 0.16 6 0.01 0.16 6 0.01 0.16 6 0.01 0.17 6 0.01
T1 0.19 6 0.01 0.24 6 0.01 0.25 6 0.01 0.26 6 0.01 0.26 6 0.01 0.27 6 0.01
T2 0.23 6 0.01 0.27 6 0.01 0.30 6 0.01 0.30 6 0.01 0.29 6 0.01 0.30 6 0.01
Tukey T0 s T1s T2 T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2
signicance P \0.001 P \0.001 P \0.001 P \0.001 P \0.001 P \0.001
G M2 T0 0.17 6 0.01 0.18 6 0.01 0.20 6 0.02 0.21 6 0.01 0.21 6 0.01 0.22 6 0.01
T1 0.29 6 0.02 0.35 6 0.00 0.36 6 0.02 0.36 6 0.02 0.36 6 0.02 0.36 6 0.02
T2 0.13 6 0.01 0.23 6 0.01 0.26 6 0.01 0.26 6 0.01 0.26 6 0.01 0.28 6 0.01
Tukey T0 s T1 s T2 T0 s T1 s T2 T0 s T1 s T2 T0, T2 s T1 T0, T2 s T1 T0 s T1 s T2
signicance P 5 0.035 P 5 0.017 P 5 0.018 P 5 0.001 P \0.001 P 5 0.03

Treatment me (months)
0 0 6 6 12 12
0.45 T0 T1 T2
6 months
6 months

0.40
0.35
P2-M1 space (mm)

0.30
0.25
0.20
0.15
0.10 G+M
M space
0.05 M space
G-M
0.00
0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15
Head gear force (N)

Fig 5. Change in space at the contact point of the second premolar and the rst molar at 6 HG forces
(0-15 N) applied at 6-month intervals at 3 time points (T0, T1, T2) in the 2 groups, G M2 and G 1 M2
(patients without and with second molars).

At T2, because the mean dental age of G M2
reached that of G 1 M2 at T1, the presence of the erupted
second molar in G M2 was evident by the signicant in-
crease in TDCP at 0 N of HG force from T1 to T2. This pro-
duced a resistance to distalization in G M2 as reected
by a sharp reduction in space at G M2 at 0 N of HG force
from T1 to T2 (61%). The similarity between the 2 groups
at T2 (erupted second molar) was demonstrated by a par-
allel marked decrease in the TDCP reading and an increase
in space readings with HG force elevation from 3 to 15 N.
Besides the eruption of the second molar in G M2, at
T2, this tooth most likely reached the Nolla tooth devel-
opment stage of G 1 M2 at T0 (root completed with
open or closed apex). Consequently, it is suggested that

American Journal of Orthodontics and Dentofacial Orthopedics October 2014  Vol 146  Issue 4
474 Shpack et al

Treatment me (months)
0 0 6 6 12 12
6.00 0.50
T1 T2

6 months
T0

6 months
0.45
5.00
0.40

P2-M1 space (mm)

0.35
4.00
P2-M1 TDCP (N)

0.30
3.00 0.25
0.20
2.00
0.15
0.10
1.00 G-M
M TTDCP
0.05
G-M
M space
0.00 0.00
0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15
Head gear force (N)

Fig 6. TDCP vs space at the contact point of the second premolar and the rst molar at 6 HG forces
(0-15 N) applied in 6-month intervals at 3 time points (T0, T1, T2) in the G M2 group (patients
without second molars).

Treatment me (months)
0 0 6 6 12 12
6.00 0.50
T0 T1 T2
6 months

0.45
6 months

5.00
0.40

P2-M1 space (mm)

0.35
P2-M1 TDCP (N)

4.00
0.30
3.00 0.25
0.20
2.00
0.15

G+M
M TDCP 0.10
1.00
G+M
M space 0.05
0.00 0.00
0 3 6 9 12 15 0 3 6 9 12 15 0 3 6 9 12 15
Head gear force (N)

Fig 7. TDCP vs space at the contact point of the second premolar and the rst molar at 6 HG forces
(0-15 N) applied in 6-month intervals at 3 time points (T0, T1, T2) in the G 1 M2 group (patients with
second molars).

the enlargement in second molar root length contributed
to the increase in resistance to distalization. In spite of the
increased resistance to initial distalization of the rst
molar, molar distalization continued but at a slower
pace. This nding corresponds with other studies that
found on average a decrease in rst molar distalization
by second and third molar buds.2-4,10
However, our study demonstrates that the initial dis-
talization potential of the rst molar for G 1 M2 was the
same at T1 and T2; this is supported by the ndings of
Abed and Brin,35 who reported on the eruption path of
the second molar during HG treatment, and can be
explained by the same stage of crown development of
the third molar at this age (13-14 years).36 Therefore,

October 2014  Vol 146  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Shpack et al 475

Table II. Pearson correlations between TDCP and space at 3 times for all HG forces (015 N)
Time Pearson correlation and P value 0 3 6 9 12 15
T0 r 0.216 0.258 0.472 0.491 0.507 0.532
P 0.109 0.054 0.001 0.001 0.001 0.001
T1 r 0.739 0.787 0.795 0.742 0.759 0.866
P 0.001 0.001 0.001 0.001 0.001 0.001
T2 r 0.570 0.659 0.662 0.643 0.643 0.773
P 0.001 0.001 0.001 0.001 0.001 0.001

our study does not support the recommendations of
Kinzinger et al4 of prior germectomy of the third molar
when a simultaneous distalization of the rst and second
molars is required.
Regarding the effective initial HG force, at T0 (treat-
ment initiation), G M2 had nonsignicant changes in
the TDCP and space measurements at 6 to 9 N. In
G 1 M2, only a small change was observed for all force
levels. That is, initial force elevation was effective up to
6 N for G M2, whereas for G 1 M2 the effect was not
signicant. Our T0 data support those of Andreasen and
Johnson,18 who reported decreased effectiveness below
3 N of HG force.
At T1, the most signicant effect was observed at 3 N,
and nonsignicant changes were observed beyond 6 N
for both groups. Also at T2, the most signicant effect
was observed at 3 N, especially in G M2. The initial
HG force load was most effective at 3 N, and only small
changes were noted with HG forces greater than 6 N.
Hence, regarding the effective initial force, there was a
positive correlation between the initial HG force and
the initial distalization up to a measured force of 6 to
9 N, after which any elevation of the initial force was
not followed by a signicant increase in distalization
in either group. Most likely, additional factors such as
mobility of the rst molar37 and enlargement of the
PDL space over time38 affect the magnitude of the effec-
tive initial force. This might be the reason that at T1 and
T2 the greatest change in initial distalization (space)
occurred from 0 to 3 N, whereas 9 N was required at
T0 to reach the most efcient force level. This suggests
that our ndings cause rejection of the second null hy-
pothesis.
CONCLUSIONS
1. It is recommended to start HG treatment before
eruption of the maxillary second molar.
2. Maxillary rst molar distalization can be pursued
even after eruption of the maxillary second molars.
3. Once the maxillary second molar erupts, maxillary
rst molar distalization continues at a slower pace
similar to treatment initiated in the presence of
the maxillary second molar.
4. The initial effective HG force level is 3 to 6 N and
should not exceed 9 N.
5. HG per se is 4 times more effective than HG com-
bined with an edgewise xed appliance.
ACKNOWLEDGMENTS
We thank Ilana Gelernter from the Department of
Statistics, Tel Aviv University, for her assistance.
REFERENCES
1. Yoshida N, Jost-Brinkmann PG, Yamada Y. Initial tooth movement
under extraoral force and considerations for controlled molar
movement. Angle Orthod 1995;65:199-208.
2. Ten Hoeve A. Palatal bar and lip bumper in nonextraction treat-
ment. J Clin Orthod 1985;19:272-91.
3. Jeckel N, Rakosi T. Molar distalization by intra-oral force applica-
tion. Eur J Orthod 1991;3:43-6.
4. Kinzinger GSM, Fritz UB, Sander FG, Diedrich PR. Efciency of a
pendulum appliance for molar distalization related to second
and third molar eruption stage. Am J Orthod Dentofacial Orthop
2004;125:8-23.
5. Reimann S, Keilig L, Jager A, Brosh T, Shpinko Y, Vardimon AD,
et al. Numerical and clinical study of the biomechanical behaviour
of teeth under orthodontic loading using a headgear appliance.
Med Eng Phys 2009;31:539-46.
6. Kloehn SJ. Guiding alveolar growth and eruption of teeth to
reduce treatment time and produce a more balanced denture
and face. Angle Orthod 1947;17:10-33.
7. Dewel BF. Objectives of mixed dentition treatment in orthodontics.
Am J Orthod 1964;50:504-19.
8. Ricketts RM. The inuence of orthodontic treatment on facial
growth and development. Angle Orthod 1960;30:103-33.
9. Locatelly R, Bednar J, Dietz V. Molar distalization with superelastic
NiTi wire. J Clin Orthod 1992;16:277-9.
10. Bussick TJ, McNamara JA. Dentoalveolar and skeletal changes
associated with the pendulum appliance. Am J Orthod Dentofacial
Orthop 2000;117:333-43.
11. Hilgers JJ. The pendulum appliance for Class II non-compliance
therapy. J Clin Orthod 1992;26:706-14.
12. Karlsson I, Bondemark L. Intraoral maxillary molar distalization
movement before and after eruption of second molars. Angle Or-
thod 2006;76:923-9.
13. Muse DS, Fillman MJ, Emmerson WJ, Mitchell RD. Molar and
incisor changes with Wilson rapid molar distalization. Am J Orthod
Dentofacial Orthop 1993;104:556-65.
14. Ghosh J, Nanda RS. Evaluation of an intraoral maxillary molar dis-
talization technique. Am J Orthod Dentofacial Orthop 1996;110:
639-46.

American Journal of Orthodontics and Dentofacial Orthopedics October 2014  Vol 146  Issue 4
476
15. Bondemark L, Kurol J, Bernhold M. Repelling magnets versus supere-
lastic nickel-titanium coils in simultaneous distal movement of
maxillary rst and second molars. Angle Orthod 1994;64:189-98.
16. Byloff FK, Darendeliler MA. Distal molar movement using the
pendulum appliance. Part 1: clinical and radiological evaluation.
Angle Orthod 1997;67:249-60.
17. Zentner A, Sergl HG, Filippidis G. A holographic study of variations
in bone deformations resulting from different headgear forces in a
macerated human skull. Angle Orthod 1996;66:463-72.
18. Andreasen G, Johnson P. Experimental ndings on tooth move-
ments under two conditions of applied force. Angle Orthod
1967;36:9-12.
19. Nolla C. The development of permanent teeth. J Dent Child 1960;
27:254-66.
20. Vardimon AD, Matsaev E, Lieberman M, Brosh T. Tightness of
dental contact points in spaced and non-spaced permanent den-
titions. Eur J Orthod 2001;23:305-14.
21. Brosh T, Machol IH, Vardimon AD. Deformation/recover cycle of
the periodontal ligament in human teeth with single or dual con-
tact points. Arch Oral Biol 2002;47:85-92.
22. Vardimon AD, Tryman Y, Brosh T. Changes over time in intra-arch
contact point tightness. Am J Dent 2003;16(Spec No):20A-4A.
23. Vardimon AD, Beckmann S, Shpack N, Sarne O, Brosh T. Posterior
and anterior components of force during bite loading. J Biomech
2007;40:820-7.
24. Houston WJB. The analysis of errors in orthodontic measurements.
Am J Orthod 1983;83:382-90.
25. Holtgrave EA, Kretschmer R, Moller R. Acceleration in dental devel-
opment: fact or ction. Eur J Orthod 1997;19:703-10.
26. Kurita LM, Menezes AV, Casanova MS, Haiter-Neto F. Dental
maturity as an indicator of chronological age: radiographic assess-
ment of dental age in a Brazilian population. J Appl Oral Sci 2007;
152:99-104.
27. Taner TU, Yukay F, Pehlivanoglu M, Cakirer B. A comparative
analysis of maxillary tooth movement produced by cervical head-
gear and pend-x appliance. Angle Orthod 2003;73:686-91.
October 2014  Vol 146  Issue 4 American Journal of Orthodontics and Dentofacial Orthopedics
Shpack et al
28. Hurng JM, Kurylo MP, Marshall GW, Webb SM, Ryder MI, Ho SP.
Discontinuities in the human bone-PDL-cementum complex.
Biomaterials 2011;32:7106-17.
29. Rygh P. Ultra structure changes in pressure zone of human perio-
dontium incident to orthodontic tooth movement. Acta Odontol
Scand 1973;31:109-22.
30. Iwasaki LR, Haack JE, Nickel JC, Morton J. Human tooth
movement in response to continuous stress of low magnitude.
Am J Orthod Dentofacial Orthop 2000;117:175-83.
31. Norton LA. Fundamental principles of the biology of tooth
movement. Semin Orthod 2000;6:139-44.
32. Slomka N, Vardimon AD, Gefen A, Pilo R, Bourauel C, Brosh T.
Time-related PDL: viscoelastic response during initial ortho-
dontic tooth movement of a tooth with functioning interprox-
imal contacta mathematical model. J Biomech 2008;41:
1871-7.
33. Keles A, Sayinsu K. A new approach in maxillary molar distaliza-
tion: intraoral bodily molar distalizer. Am J Orthod Dentofacial
Orthop 2000;117:39-48.
34. Schneevoigt R, Bourauel C, Harzer W, Eckardt L. Biomechanical
analysis of arch-guided molar distalization when employing
superelastic NiTi coil springs. J Orofac Orthop 1999;60:
124-35.
35. Abed Y, Brin I. Early headgear effect on the eruption pattern of
maxillary second molars. Angle Orthod 2010;80:454-60.
36. Thevissen PW, Alqerban A, Asaumi J, Kahveci F, Kaur J, Kim YK,
et al. Human dental age estimation using third molar develop-
mental stages: accuracy of age predictions not using country
specic information. Forensic Sci Int 2010;201:106-11.
37. Tanaka E, Ueki K, Kikuzaki M, Yamada E, Takeuchi M, Dalla-
Bona D, et al. Longitudinal measurements of tooth mobility during
orthodontic treatment using a periotest. Angle Orthod 2005;75:
101-5.
38. King GJ, Keeling SD, Wronski TJ. Histomorphometric study of
alveolar bone turnover in orthodontic tooth movement. Bone
1991;12:401-9.