ORIGINAL ARTICLE

Second molar intrusion: Continuous arch

or loop mechanics?
Renato Parsekian Martins,a Ricardo Lima Shintcovsk,b Luegya Knop Shintcovsk,c Rodrigo Viecilli,d
and Lidia Parsekian Martinsb
Araraquara, S~ao Paulo, and Salvador, Bahia, Brazil, and Loma Linda, Calif

Introduction: Our aim was to compare the load systems produced by rectangular loops and continuous arch
mechanics on intrusion of a maxillary second molar. Methods: An orthodontic appliance was passively bonded
from second molar to second molar on a patient's model fixed to an orthodontic force tester. The second molars
were extruded by 3 mm, and 2 mechanics systems were compared: beta-titanium rectangular loops and super-
elastic nickel-titanium continuous arch. Vertical (Fz) and mesial (Fy) forces, and tipping moments around the
buccal axis (Mx) at the tubes were collected at every 0.5 mm of deactivation. Repeated measures analysis of
variance was used to detect differences between mechanics. The net Mx was also estimated at the center of
resistance of each molar. Results: Significant differences were found between the mechanics at the tubes,
except for the Mx in the first molar, and more importantly, at the centers of resistance of the molars: Fz forces
were greater in the continuous arch than in the rectangular loops, which, on the other hand, produced lower
Fy forces than the continuous arch. In the second molar, net Mx was low initially and changed with deactivation.
In the continuous arch, net Mx decreased from –3.6 N.mm of mesial crown tipping to clinically insignificant levels
after 2 mm of deactivation. In the first molar, net Mx was lower on the rectangular loop than on the continuous
arch during the first 2 mm of deactivation and became similar in the last millimeter of deactivation. Additionally,
the net Mx decreased with deactivation with both mechanics. Conclusions: Both mechanics can intrude the
second molars. Although the rectangular loop produced less tipping in large deflections, the continuous arch
was a better approach when less intrusion was required. (Am J Orthod Dentofacial Orthop 2018;154:629-38)

M
axillary second molars may be malpositioned
for several reasons: dental loss, ectopic erup-
tion, agenesis, inadequate mechanics, and
incorrect positioning of tubes. All may lead to their
extrusion, which can complicate the orthodontic treat-
ment because intruding second molars is a difficult
problem. Because it is an unsupported cantilever, the
posterior terminal portion of an orthodontic archwire
has high flexibility, which complicates the correction of
the last tooth connected to the appliance—ie, an
a
Private practice, Araraquara, S~ao Paulo, Brazil; Faculdade de Odontologia de
Araraquara, Universidade Estadual Paulista, Araraquara, S~ao Paulo, Brazil.
b
Faculdade de Odontologia de Araraquara, Universidade Estadual Paulista, Ara-
raquara, S~ao Paulo, Brazil.
c
Private practice, Salvador, Bahia, Brazil.
d
Biomechanics and Microimaging Laboratory, Department of Orthodontics,
School of Dentistry, Loma Linda University, Loma Linda, Calif.
All authors have completed and submitted the ICMJE Form for Disclosure of Po-
tential Conflicts of Interest, and none were reported.
Address correspondence to: Renato Parsekian Martins, Rua Carlos Gomes, 2158,
14801320, Araraquara, SP, Brazil; e-mail, dr_renatopmartins@hotmail.com.
Submitted, November 2016; revised and accepted, January 2018.
0889-5406/$36.00
Ó 2018 by the American Association of Orthodontists. All rights reserved.
https://doi.org/10.1016/j.ajodo.2018.01.019
extruded second molar. Moreover, the absence of a distal
fulcrum makes it almost impossible for the archwire to
be used as a shape-driven approach to guide those
distal-end teeth to their correct positions.
In this situation, the step geometry between the tube
of an extruded second molar and the tube of a well-
positioned first molar produces a characteristic deflec-
tion of the archwire, which causes a crown-mesial
(forward) tipping tendency of both molars to equilibrate
the vertical forces acting on the tubes. This has already
been demonstrated in computer simulations of 2
brackets in that similar position, suggesting that contin-
uous arch mechanics could not produce intrusion of a
second molar without tipping the crown mesially along
with the first molar.1,2
One solution to this problem is to use a rectangular
loop, changing the insertion of the wire from going
through the mesial of the second molar tube to the
distal.3 As has been shown analytically and practically,
this loop can disassociate the tipping moments from
the vertical force, producing a different force system
from the one inevitably produced by continuous arch
mechanics.4-8 Thus, according to the orthodontist's
desire, a vertical force can be produced along a crown-
629
630
Fig 1. A patient's resin model with a passive orthodontic
appliance bonded. The model was secured to the ortho-
dontic force tester, and each molar was bonded to a
load cell. The molars were later separated from the
model, allowing the load system to be measured distinc-
tively. The orthodontic force tester software calibration
tool digitally rotates the load cells and translates the origin
of the measurement system to the center of the tubes
even when the load cells are not aligned with the tubes.
distal, crown-mesial, or even an absent tipping moment,
allowing intrusion of a second molar without tipping it.
There is little information available on how to use the
rectangular loop in a systematical way, even though it is
known that a pure force without moments can be pro-
duced in the center of its “box”8 with a high reactive
moment at its opposite end.6,9 Moreover, it has been
shown that its dimensions4,7 and activations5,7 may
change the load system, and that it may be used
with4,5,8 or without4,7 preactivation. However, the
literature fails to show any comparison of these
rectangular loops with continuous arch mechanics in
simulations with exact oral dimensions and brackets
bonded to all teeth.
The orthodontic force tester is an orthodontic mea-
surement system that allows nano-load cells to be con-
nected to specific teeth of a patient's model and to
acquire the tridimensional force system applied to
them.10 This system has been used in recent years to
November 2018  Vol 154  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Martins et al
Fig 2. The digital dial indicator and precision vertical
extension rod were attached to the orthodontic force
tester to allow precise vertical movement of the
second molar. a, Load cells; b, acrylic model; c, precision
rod; d, digital dial; e, steel table of the orthodontic force
tester; f, magnetic rod securing the digital dial; g, power
supply boxes of the load cells.
describe and compare several orthodontic mechanics in
a way that was not possible before, when the patients'
characteristics could not be replicated and simu-
lated.11-13 Therefore, the aim of this study was to
compare the load system produced by a preactivated
rectangular loop continuous vs a nickel-titanium wire
for the intrusion of a second molar.
MATERIAL AND METHODS
A maxillary arch acrylic resin model of an orthodontic
patient had the first molars banded with 0.018-in slot tri-
ple tubes (Roth prescription; Ormco, Glendora, Calif),
while the remaining teeth had 0.018-in brackets and
tubes (Opal Orthodontics, Sandy, Utah) of the same pre-
scription passively bonded with epoxy glue (J-B Weld,
Sulphur Springs, Tex) using a 0.017 3 0.025-in stainless
steel wire as a guide (Fig 1). The model was secured
into an orthodontic force tester, and the left first
and second molars were bonded to the 2 load cells
Martins et al 631

Fig 3. Software developed at the biomechanics laboratory at the Department of Mechanical Engineer-
ing, Indiana University-Purdue University with the “calibration option” window open. Load cell number 1
rotation calibration around the z-axis was done by measuring the angular changes of the molar tube in
relation to the x-axis of the load cell with a protractor and adding it to the calibration window (blue arrow)
and fine tuning it until only Fx was registered by the software window when that force was applied to the
tube. Rotation calibrations were not necessary on the other axes. Displacement of the origin of mea-
surements was achieved by applying a force on the y-axis to the center of the tube and transferring
the values of Tz/Fy (the ratio between the moment in the z-axis to the force in the y-axis) to the x offset
(black arrows) and –Tx/Fy (the negative value of the ratio between the moment in the x-axis to the force
in the y-axis) to the z offset. The y offset was found and added by applying 1 force to the x-axis and
registering the –Tz/Fx ratio. Note: The values obtained for the ratios and offsets in the figure are illus-
trational and do not match.

(multi-axis force/torque Nano 17; ATI Industrial Automa-
tion, APEX, NC) of the orthodontic force tester, which has
amplitude of measurements of 0-20 N and 0-100 N-mm.
The molars were separated from the model, and the
second molar was moved 3 mm vertically using an adapted
precision rod and a digital dial (Mitutoyo, Aurora, Ill.) to the
orthodontic force tester (Fig 2). A custom modification of
the ATI transducer software (Department of Mechanical
Engineering, Indiana University-Purdue University, Indi-
anapolis, Ind) registered the moment/force ratios produced
when a force was applied to the tubes and away from the
load cells' origin (center of the load cell). This allowed the
insertion of those offset values into the software's calibra-
tion and the transference of the origin of the measurements
3 dimensionally to the center of the tube of each molar,
with the x-axis and z-axis perpendicular to the center of
the tubes (Fig 3). The software also allowed rotational
calibration of the origin of the measurements in all 3 di-
mensions.
Ten rectangular loops were made with a
0.017 3 0.025-in beta-titanium (TMA; Ormco) wire
on a custom-made template in the Loop software
(dHAL Software; Athens, Greece) and printed in a 1:1 ra-
tio, to standardize the loops. A photograph of the system
was inserted into the software in scale, and the loops
were drawn passively, with dimensions of 8 3 10 mm,
ensuring that the center of the second molar tube was
in the center of the box of the loop (Fig 4). The bent
loops were inserted in the model and adjusted until
they were passive, using the measurement indication
on the orthodontic force tester software.
Because the loops were built passively, they needed
to be preactivated. Preactivation is the shape modifica-
tion (eg, a group of bends) made into the loops to modify

American Journal of Orthodontics and Dentofacial Orthopedics November 2018  Vol 154  Issue 5
632
Fig 4. A, Image of the simulation inserted and calibrated
to the Loop software. The software can be used to print in
a 1:1 ratio the loop constructed passively and with its box
centralized to the second molar tube. B, The virtual
bracket of the second molar tube was moved 3 mm verti-
cally to simulate a 3-mm vertical activation of the loop.
The deflection on a 3-mm loop produced a shape modifi-
cation that was frozen by the software and printed in a 1:1
ratio as the template for preactivation.
their shapes to make them active, before placing them
into the orthodontic appliance. The vertical preactiva-
tion of the loops was performed using a second tem-
plate, also made by the Loop software, which can
estimate wire deflections and freeze the acquired design
for a shape-type preactivation. The virtual tube of the
second molar was positioned 3 mm vertically without
angulation (Fig 4, B) into a level position, and the loop
(which was built passively to the extruded molar) was
activated. The deformation of the loop was frozen by
the software, and this preactivated loop shape was
printed in a 1:1 ratio.14 The real loops had their passive
shapes modified into this new preactivated shape and
subjected to stress relief through trial activations until
their new shapes were stabilized. Felt tip pen marks
were made on the loops to ensure centralization of the
second molar tube in the center of the loop box on
engagement in the maloclusion simulation model (Fig 5,
A and B).
November 2018  Vol 154  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Martins et al
Fig 5. Rectangular loop: A, preactivated and B, engaged
on a 3-mm vertical activation; C, 0.014-in super-elastic
nickel-titanium wire engaged on the 3-mm extruded
second molar for data collection.
The loops were inserted into the molar tubes with the
second molar tube centralized in the box of the loop. The
vertical (Fz) and horizontal (Fy) forces and the tipping
moments (Mx) were recorded by the orthodontic force
tester software at every 0.5 mm of deactivation until
the second molar leveled (from 3-0 mm extrusion).
The precise vertical movement of the second molar
was allowed by the rod coupled to the orthodontic force
tester, which moved only in the vertical direction by the
precision gear rotation and digital dial.
The second mechanics system for intrusion of the
second molar was the continuous arch. Ten 0.014-in
super-elastic nickel-titanium arches (Highland Metals,
Franklin, Ind) were inserted into the brackets and tubes,
from second molar to second molar, and secured in place
with elastic ligatures (Fig 5, C). Unlike the loops, the
wires did not require preactivation because they were
not made passive to the maloclusion. They were manu-
factured and already active once inserted into the unlev-
elled brackets. The second molar was also extruded by
3 mm. Force and moment (Fz, Fy, and Mx) values were
also recorded at every 0.5 mm of deactivation until full
alignment was reached (0 mm). Because nickel-
Martins et al
titanium wires were being evaluated, all mechanical tests
were performed at 37 C 6 1 C.
The data were compared with software (version 16.0;
SPSS, Chicago, Ill). Four repeated-measures analysis of
variance tests, with a significance level of 5%, were
used to detect differences in the moments and forces
produced between the loop and continuous arch me-
chanics in both first and second molars.
The center of resistance location was used to calculate
the net Mx, which is relevant to understanding the final
rotation direction of each molar and was calculated ac-
cording to the axes of resistance methodology proposed
by Viecilli et al.15 Therefore, to estimate the molar's center
of resistance, previously published finite element models
were used. As previously described, the center of resis-
tance for a movement was defined as the intersection
of the orthogonal projections of the 2 axes of resistance
perpendicular to the desired force direction.16 The num-
ber, displacement, and coordinates of the node that de-
fines 1 axis of rotation were recorded with an initial
couple load of 3 N-mm. The moment vector determined
the direction of each axis. Peak periodontal ligament
principal strains were under about 7.5% in all directions
with this load, thus justifying the periodontal ligament's
linear properties at E 5 0.05 MPa. Cross-product matrix
calculations were performed by inputting nodal coordi-
nates and the final moment value, to output the neces-
sary nodal forces to obtain a moment vector parallel to
each coordinate system axis to the tooth. A moment of
a couple parallel to the x direction, for example, causes
a rotation axis that generates nodal displacement on
the y and z directions. This couple-derived axis of rotation
was then used as 1 of the 3 axes of resistance. The same
procedure was performed for the other axes. We further
approximated the coordinates of the node that defined
each axis by iterative refinement of the mesh around
the node with the lowest displacements in the plane of
movement, until the position varied within 60.05 mm.
Even though Fz and Fx measured by the load cells
were the same at the bracket and at the estimated cen-
ters of resistance of the molars, they will influence the
Mx measured by the load cell at the brackets, producing
at the center of resistance a net Mx that will be actually
the one felt by the molars. This (net) Mx produced at the
centers of resistance of the molars was calculated using
the following formulas.
For the second molar, net Mx (at the center of
resistance) 5 Fy*( 8.32) 1 Fz*( 1.34) 1 Mx (at the tube).
For the first molar, with the continuous arch me-
chanics, net Mx (at the center of resistance) 5
Fy*( 7.53) 1 Fz*( 0.6) 1 Mx (at the tube).
For the first molar, with the loop mechanics, a
different vertical distance was used because the auxiliary
American Journal of Orthodontics and Dentofacial Orthopedics
633
tube was located 1.9 mm more apically (where the
loop was engaged). Net Mx (at the center of
resistance) 5 Fy*( 5.63) 1 Fz*( 1.34) 1 Mx (at the
tube).
RESULTS
The 4 analysis of variance tests of repeated measure-
ments identified significant differences (P \0.001) be-
tween the Mx, Fz, and Fy values produced by the
rectangular loops and the continuous arch in both molar
tubes, except for the Mx values in the first molar tube
(P 5 0.232) that were similar between mechanics.
In the second molar tube, the rectangular loops
produced an insignificant Mx initially, whereas the
continuous arch produced 7.9 N-mm. Throughout
deactivation, Mx remained either very low or insignifi-
cant (0.5 and 0 mm) for the rectangular loop; for the
continuous arch, Mx dropped linearly (Table I). In the
first molar tube, the rectangular loop produced a Mx
of 9.5 N-mm at the 3-mm activation, whereas the
continuous arch produced 8.8 N-mm. In both me-
chanics, Mx dropped linearly with deactivation (Table I).
In the second molar tube, the rectangular loops pro-
duced an initial intrusive Fz of 0.94 N, and the contin-
uous wire produced 1.5 N (Fig 6). Forces dropped in
both mechanics, with deactivation linearly (Table II). In
the first molar tube, the rectangular loop produced an
extrusive Fz of 0.96 N initially, whereas the continuous
arch yielded a Fz of 1.5 N. Similarly to what happened
in the second molar, the forces dropped linearly with
deactivation in both mechanics (Table II).
In the second molar tube, the rectangular loops pro-
duced mesial forces from .12 to .08 N, without signifi-
cant variations with deactivation, whereas the
continuous arch produced distal forces from .28 N
decreasing to insignificant forces (Table III; Fig 7). In
the first molar tube, however, the loops produced distal
forces from .10 N to insignificant forces, and the
continuous arch produced mesial forces of .67 N
initially, gradually decreasing to insignificant forces
with deactivation.
The location of the center of resistance of the first
molar was estimated to be 7.53 mm apically to the center
of the main tube and 0.6 mm mesially. On the
second molar, it was estimated to be 8.32 mm apically
to the center of the tube and 1.34 mm mesially. When
the resulting net Mx was calculated in relation to the
center of resistance of the second molar, based on the
average forces and Mx values produced by the rectan-
gular loop (Table IV), the net Mx was low initially and
increased with deactivation (although its range was rela-
tively low). In the continuous mechanics, net Mx was
November 2018  Vol 154  Issue 5
634 Martins et al

Table I. Average moments (Mx) and standard deviations produced on the second and first molar tubes (Newtons/mm)
from 3 mm to zero of deactivation
Mechanics 3.0 mm 2.5 mm 2.0 mm 1.5 mm 1.0 mm 0.5 mm 0.0 mm
Second molars Rectangular loop \.01 (.06)* .14 (.21) .30 (.22) .34 (.27) .28 (.28) .13 (.33)* .18 (.33)*
Continuous arch 7.87 (.26) 6.51 (.22) 5.01 (.19) 2.93 (.14) .96 (.10) .11 (.18)* .14 (.20)
First molars Rectangular loop 9.55 (1.0) 7.17 (.99) 5.15 (.91) 3.12 (.85) 1.19 (.79) .30 (.60)* .36 (.48)
Continuous arch 8.85 (.76) 7.28 (.36) 5.13 (.29) 2.22 (.33) .36 (.28) .19 (.31) .33 (.37)
Negative values describe a moment in a crown-mesial direction. Multiply the values by 102 for an approximate conversion to gf-mm.
*Nonsignificant value.

Fig 6. Vertical forces (Fz) produced in the second and first molar tubes. Blue lines depict the rectan-
gular loop effects, and red lines indicate the continuous arch mechanics effects throughout the 3-mm
intrusion of a second molar.

Table II. Average forces (Fz) and standard deviations produced on the second and first molar tubes (Newtons) from
3 mm to zero of deactivation
Mechanics 3.0 mm 2.5 mm 2.0 mm 1.5 mm 1.0 mm 0.5 mm 0.0 mm
Second molars Rectangular loop .94 (.11) .72 (.10) .52 (.09) .33 (.08) .14 (.07) \.00 (.05)* .05 (.04)
Continuous arch 1.47 (.11) 1.20 (.09) .88 (.07) .53 (.03) .22 (.01) .02 (.01) .02 (.02)
First molars Rectangular loop .96 (.10) .74 (.09) .55 (.08) .36 (.07) .17 (.06) .02 (.04)* .02 (.02)
Continuous arch 1.55 (.09) 1.31 (.08) 1.03 (.07) .75 (.05) .39 (.04) .05 (.02) .02 (.03)*
Negative values describe intrusive forces. Multiply the values by 102 for an approximate conversion to gF.
*Nonsignificant value.

Table III. Average forces (Fy) and standard deviations produced on the second and first molar tubes (Newtons) from
3 mm to zero of deactivation
Mechanics 3.0 mm 2.5 mm 2.0 mm 1.5 mm 1.0 mm 0.5 mm 0.0 mm
Second molars Rectangular loop .12 (.09) .11 (.06) .09 (.06) .08 (.07) .09 (.07) .11 (.08) .11 (.08)
Continuous arch .28 (.17) .39 (.19) .55 (.19) .38 (.09) .14 (.03) .. 01 (\.01)* .. 01 (\.01)*
First molars Rectangular loop .07 (.09) .07 (.07) .05 (.07) .05 (.08)* .07 (.09) .09 (.09) .10 (.09)
Continuous arch .67 (.10) .51 (.09) .35 (.07) .13 (.03) .02 (.02) . .01 (\.01)* . .01 (\.01)*
Negative values describe a distal force. Multiply the values by 102 for an approximate conversion to gF.
*Nonsignificant value.

November 2018  Vol 154  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Martins et al
Fig 7. Horizontal forces (Fy) produced in the second and first molar tubes. Blue lines depict the rect-
angular loop effects, and red lines indicate the continuous arch mechanics effects throughout the 3-mm
intrusion of a second molar.
high initially ( 3.6 N.mm), decreased, changed direction
right before 1 mm of deactivation (Fig 8), and reached a
clinically insignificant level after 2 mm of deactivation.
The net Mx produced at the center of resistance of the
first molar was lower on the rectangular loop than on
the continuous arch during the first 2 mm of deactiva-
tion. The net Mx decreased with deactivation in both
mechanics, but after 2 mm of deactivation, it changed
direction on the rectangular loop. On the continuous
arch, net Mx of the first molar became insignificant after
2.5 mm of deactivation.
DISCUSSION
The rectangular loop did not produce significant
tipping moments on the second molar tube initially,
and produced low and clinically insignificant levels dur-
ing deactivation. This was different from the continuous
arch mechanics, where a large crown-mesial tipping
moment was produced, proportionally to the wire's ver-
tical activation. This means that the continuous arch
mechanics could not produce an intrusive force in a
second molar tube without a tendency to tip it. This dif-
ference in behavior is explained by the loop's design,
which allows a disassociation of the force and moment
produced,4,7 unlike what happens in continuous arch
mechanics, where the geometry between the brackets
dictates a specific force system.1,2 Even though the
force system of rectangular loops has been studied
analytically, its application was never tested in a real-
size model simulating intrusion of the second molars,
as we did, nor it has been compared with continuous
arch mechanics.3-7 Additionally, we measured the
horizontal forces produced by both mechanics at the
tubes, something unseen in the loop literature.
American Journal of Orthodontics and Dentofacial Orthopedics
635
Because the horizontal forces have their lines of action
below the centers of resistance of the molars, they can
add to or subtract from the tipping moments produced
at the tubes. On the other hand, it is thought that the
vertical forces produced will not alter significantly the
tipping of the molars, because tubes are generally
positioned in the middle of the teeth (and aligned with
the center of resistance). Based on our findings, the
tipping tendency will be influenced by both the
horizontal and vertical forces. Even though the center
of resistance will be close to the molar tubes, the
vertical forces are high enough to produce enough
tipping moments to add to or subtract from the ones
produced by the loop. Using the finite element
method, we could calculate the net tipping moments
produced at both molars' centers of resistance to
estimate the molar movements at each step of
deactivation. The vertical forces of both mechanics
were applied distally to the second molar's center of
resistance, which would tend to tip the second molar
crown distally, interfering with the small Mx at the
bracket of the loop but decreasing the mesial tipping
caused by the continuous arch. It appears that the
mesial force tendency created by the deactivation of
the loop (that would tend to tip the second molar
mesially), and the distal force tendency caused by the
continuous arch (that would tend to tip it distally)
(Fig 7), counteracted the tipping effects caused by verti-
cal forces.
For a large extrusion (3 mm), the rectangular loop ap-
pears to be a better approach than continuous arch me-
chanics because it will intrude second molars with
insignificant tipping rather than large tipping
(Table IV; Fig 6). Moreover, the rectangular loop would
cause fewer side effects to the first molar. On the other
November 2018  Vol 154  Issue 5
636 Martins et al

Table IV. Average moments (Mx) at the center of resistance produced on the second and first molars (Newtons/mm)
from 3 mm to zero of deactivation
Mechanics 3.0 mm 2.5mm 2.0 mm 1.5 mm 1.0 mm 0.5 mm 0.0 mm
Second molars Rectangular loop 0.2 (.7) 0.1 (.6) 0.3 (.5) 0.6 (.6) 0.8 (.5) 1.1 (.5) 0.8 (.55)
Continuous arch 3.6 (1.3) 1.7 (1.6) 0.7 (1.6) 0.9 (.8) 0.5 (.3) 0.2 (.2) 0.2 (.3)
First molars Rectangular loop 9.7 (1.1) 7.2 (1.2) 5.2 (1.1) 3.0 (1.2) 0.9 (1.2) 0.8 (1.0) 0.9 (1.0)
Continuous arch 14.9 (1.2) 11.9 (.7) 8.4 (.7) 3.6 (.5) .8 (.4) 0.2 (.4) 0.4 (.4)
Negative values describe a mesial-crown tipping tendency. Multiply the values by 102 for an approximate conversion to gf-mm.

Fig 8. Net tipping moments (Mx) produced in the second and first molars calculated at the centers of
resistance of the molars. Blue lines depict the rectangular loop effects, and red lines indicate the contin-
uous arch mechanics effects throughout the 3-mm intrusion of a second molar.

hand, with a small extrusion, such as .75 to 0 mm, the
simpler mechanics of a continuous arch would not
only intrude a second molar without tipping it, but
also cause fewer side effects than a rectangular loop.
Therefore, when trying to intrude a second molar in
the conditions tested by this experiment, the orthodon-
tist could use the rectangular loop for the first 2.0 mm of
intrusion and reactivate it properly to be used for the re-
maining amount of intrusion, or the mechanics could be
changed at this point to a continuous arch. However,
this information must be used carefully because the
threshold limit of force to achieve molar intrusion is
not clearly known. The optimal force clinically suggested
in the literature for intruding 1 molar is about 50 gf and,
even though the continuous arch could produce a small
Mx at the center of resistance, the vertical force pro-
duced by a .014-in nickel-titanium wire at that deflec-
tion might be subthreshold.17
The rectangular loop that we tested produced intru-
sive forces without a significant tipping tendency. This
occurred because the net Mx measured at the center of
resistance of the second molar was low, especially in
the beginning of the loop's deactivation. A similar effect
could even be achieved with a rectangular loop without
preactivation, but only in specific situations that could
not be replicated in every patient: eg, if a 6 3 10-mm
loop is used, on a 2- to 3-mm extrusion, in a specific
interbracket distance of 7 mm.7 When a customized
loop for each distinct clinical situation is needed, the
answer is to simulate the desired movement in a passive
loop and replicate its deformation as an individualized
preactivation, as we did in this study by simulating dig-
ital intrusion without tipping. Clinically, this can be done
with pliers, and the loop's shape can be captured with a
camera. With proper preactivation in the loop, it will not
produce tipping in the second molar tube because its
angle of entry is parallel to the molar tube slot orienta-
tion and should remain relatively parallel during deacti-
vation (Fig 9). In contrast, this parallel orientation
between the wire and tube does not occur when a
continuous arch is engaged in the same situation, since
the step geometry between the molar tubes causes
mesial-crown tipping of both molars because of the
wire deflection (Fig 5, C).1,2
The centralization of the loop box with the
second molar tube had great importance for this study.
An incorrect position of the loop can produce unwanted
moments in both directions, preventing it from produc-
ing vertical forces and insignificant moments at the
same time.8 In this simulation, .018-in tubes were used
to decrease second-order clearance to increase the accu-
racy of the measurements, enabling even small moments
to be measured. Because positioning errors can cause sig-
nificant negative effects, the clinical use of a .022-in slot

November 2018  Vol 154  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Martins et al
Fig 9. Superimposition of a preactivated loop disen-
gaged and engaged. There is a difference in length of
the posterior horizontal extremity of the loop with activa-
tion. Cinching or tying the loop back will keep the loop
from functioning properly.
on the second molar instead of an .018-in slot would in-
crease second-order clearance and could, therefore, mask
small positional errors of the loop with the second molar
tube. Again, this information should not be evaluated by
itself without considering the load system at the center of
resistance of the molar.
Mesial-crown tipping moments were present at the
first molar tube in both mechanics, due to equilibrium.
Moments were similar in both molar tubes with the
continuous arch mechanics, but when the loop was
used, they were only detected in the first molar tube. It
has been already shown that rectangular loops can pro-
duce insignificant tipping moments within their “box” at
the expense of a high tipping moment at their other
end,6,9 suggesting the need for reinforced anchorage
in that reactive unit: eg, a rigid transpalatal bar.6
When evaluating the net effect of the rectangular loop
at the first molar's center of resistance with the rectan-
gular loop (Fig 8), the tipping was in a mesial-crown di-
rection during the first 2 mm of deactivation, and then it
changed to a crown-distal direction. Clinically, it would
be wise to use a transpalatal bar on the first molars
because that tipping would modify the geometry be-
tween the tubes and change the force system. In the
continuous arch mechanics, the side effect on the first
molar was even greater initially, but it decreased to ne-
glegable values in a small 0.5-mm wire deflection, where
it would not require a transpalatal bar. Thus, using a
continuous arch to intrude by 3 mm the extruded
second molar, it would cause not only side effects on
the second molar by tipping it, but also a large tipping
moment on the first molar. That would bring the distal
end of the first molar tube occlusally and closer to the
mesial end of the second molar tube, decreasing the
American Journal of Orthodontics and Dentofacial Orthopedics
637
vertical activation of the wire along with the undesirable
tipping side effects.
Vertical forces were smaller when the rectangular
loop was used than in the continuous arch group. This
was due to the higher load-deflection ratio of the
.014-in nickel-titanium wire (.73 N/mm vs .31 N/mm
[75 gf/mm vs 32 gf/mm] compared with the rectangular
loop); since both mechanics were activated at 3 mm, the
initial force was higher in the continuous wire. However,
if a different force is needed clinically on the loop, its
vertical preactivation can be modified regardless of the
moments produced, due to its capacity of disassociating
force from moments produced. This is different from
what happens with the continuous arch, where the force
cannot be altered in the same wire because it depends on
the distance between the tubes and their geometry.
However, in both situations, the vertical forces produced
were proportional to their activations.
An unusual result was found in relation to the vertical
forces produced. In the rectangular loops, the vertical
forces in both molars were similar, as expected, but in
the continuous arch, they were not—something not ex-
pected due to the step geometry.1,2 Initially, it could
be assumed that the intrusive force in the
second molar should equal the sum of the extrusive
forces distributed throughout the arch, decreasing
gradually as with the first molar (Fig 10), but instead,
a larger extrusive force was found in the first molar
than the intrusive one in the second molar (Table II).
The different vertical forces detected in both first and
second molars occurred because there were more
brackets involved in the mechanics and not only 2. The
step geometry and the deflection of the wire can explain
this finding. The deflected wire produced a large mesial-
crown moment in the first molar and had an exit angle
from the first molar tube in the apical direction, produc-
ing an intrusive force (not an extrusive force, as ex-
pected) in the second premolar, adding to the extrusive
force in the first molar itself (Fig 11). This deflection
was unnoticed because in a straight-wire system, the di-
mensions of the wires, brackets, tubes, and interbracket
distances are too small for the deflection to be seen by
the naked eye, but it could be successfully measured
by the load cells of the orthodontic force tester.
CONCLUSIONS
When we intruded a 3-mm extruded second molar,
the rectangular loop did not initially produce tipping,
but as it was deactivated, a small tipping tendency
was produced. In contrast, the continuous arch pro-
duced a large tipping moment, but in small extrusions
(.75-.5 mm), it may not tip the second molar.
November 2018  Vol 154  Issue 5
638
Fig 10. Normally estimated vertical forces that would act
on the simulation tested, which are different from the
measured forces. F1 5 – (F2 1 F3 1 F4). Red circles
are rough estimates of the location of the centers of resis-
tance. (Other forces and moments are purposely not
drawn.)
Fig 11. A, Force system of 2-bracket step geometry; yel-
low arrows indicate forces, and blue arrows indicate mo-
ments that should be of the same magnitude to be in
equilibrium. B, Black arrows show the couple on the molar
tube to produce the tipping moment. The dashed white
line shows the estimated deflection of the wire when dis-
engaged from the premolar brackets, which is respon-
sible for the added extrusive force on the first molar.
When using the loop for a 3-mm intrusion, it should
be reactivated halfway to keep the second molar from
tipping significantly.
Both types of mechanics will tip the first molar,
except for the continuous arch in smaller activations.
In the continuous arch, the wire deflection caused
an unexpected difference between the extrusive force
November 2018  Vol 154  Issue 5 American Journal of Orthodontics and Dentofacial Orthopedics
Martins et al
in the first molar and the intrusive force in the
second molar.
The rectangular loop produced lower forces than did
the 0.014-in nickel-titanium super-elastic wire.
SUPPLEMENTARY DATA
Supplementary data related to this article can be
found online at https://doi.org/10.1016/j.ajodo.2018.
01.019.
REFERENCES
1. Burstone CJ, Koenig HA. Force systems from an ideal arch. Am J
Orthod 1974;65:270-89.
2. Koenig HA, Burstone CJ. Force systems from an ideal arch—large
deflection considerations. Angle Orthod 1989;59:11-6.
3. DeFranco JC, Koenig HA, Burstone CJ. Three-dimensional large
displacement analysis of orthodontic appliances. J Biomech
1976;9:793-801.
4. Raboud D, Faulkner G, Lipsett B, Haberstock D. Three-dimensional
force systems from vertically activated orthodontic loops. Am J Or-
thod Dentofacial Orthop 2001;119:21-9.
5. Menghi C, Planert J, Melsen B. 3-D experimental identification of
force systems from orthodontic loops activated for first order cor-
rections. Angle Orthod 1999;69:49-57.
6. Crismani AG, Freudenthaler JW, Bernhart T, Hofshneider UB,
Kole EG. Rectangular loops in inconsistent force systems. J Orofac
Orthop 2002;63:105-12.
7. Vanderby R Jr, Burstone CJ, Solonche DJ, Ratches JA. Experimen-
tally determined force systems from vertically activated orthodon-
tic loops. Angle Orthod 1977;47:272-9.
8. Cacciafesta V, Melsen B. The rectangular loop: biomechanical
principles and clinical applications in three dimensional control
of single-tooth discrepancies. Prog Orthod 1999;1:23-36.
9. Koenig HA, Burstone CJ. Analysis of generalized curved beams for
orthodontic applications. J Biomech 1974;7:429-35.
10. Chen J. Apparatus and method for measuring orthodontic force
applied by an orthodontic appliance. United States Patent Number
6.120.287.
11. Xia Z, Chen J, Jiangc F, Li S, Viecilli RF, Liu SY. Load system of
segmental T-loops for canine retraction. Am J Orthod Dentofacial
Orthop 2013;144:548-56.
12. Almeida L, Ribeiro A, Parsekian Martins R, Viecilli R, Parsekian
Martins L. Nickel titanium T-loop wire dimensions for en masse
retraction. Angle Orthod 2016;86:810-7.
13. Katona TR, Isikbay SC, Chen J. Effects of first- and second-order
gable bends on the orthodontic load systems produced by T-
loop archwires. Angle Orthod 2014;84:350-7.
14. Halazonetis DJ. Understanding orthodontic loop preactivation.
Am J Orthod Dentofacial Orthop 1998;113:237-41.
15. Viecilli RF, Budiman A, Burstone CJ. Axes of resistance for
tooth movement: does the center of resistance exist in 3-
dimensional space? Am J Orthod Dentofacial Orthop 2013;
143:163-72.
16. Viecilli RF, Burstone CJ. Ideal orthodontic alignment load relation-
ships based on periodontal ligament stress. Orthod Craniofac Res
2015;18(Suppl 1):180-6.
17. Melsen B, Fiorelli G. Upper molar intrusion. J Clin Orthod 1996;30:
91-6.