ORIGINAL ARTICLE
Effects of cervical headgear and edgewise
appliances on growing patients
Marcia R. E. A. Schiavon Gandini, DDS, MS, PhD,a Luiz G. Gandini, Jr, DDS, MS, PhD,b Joel C. da Rosa
Martins, DDS, MS, PhD,† and Marinho Del Santo, Jr, DDS, MS, MSD, PhDc
Araraquara, São Paulo, Brazil
Maxillary basal bone, dentoalveolar, and dental changes in Class II Division 1 patients treated to normal occlusion
by using cervical headgear and edgewise appliances were retrospectively evaluated. A sample of 45 treated
patients was compared with a group of 30 untreated patients. Subjects were drawn from the Department of
Orthodontics, Araraquara School of Dentistry, Brazil, and ranged in age from 7.5 to 13.5 years. The groups were
matched based on age, gender, and malocclusion. Roughly 87% of the treated group had a mesocephalic or
brachicephalic pattern, and 13% had a dolicocephalic pattern. Cervical headgear was used until a Class I dental
relationship was achieved. Our results demonstrated that the malocclusions were probably corrected by
maintaining the maxillary first molars in position during maxillary growth. Maxillary basal bone changes (excluding
dentoalveolar changes) did not differ significantly between the treated and the untreated groups. Molar extrusion
after the use of cervical headgear was not supported by our data, and this must be considered in the treatment
plan of patients who present similar facial types. (Am J Orthod Dentofacial Orthop 2001;119:531-9)
S
keletal Class II malocclusion may be characterized by maxillary protrusion, mandibular retrusion, or a combination of both.1 Although there is
no rigorous operational rule for differentiating skeletal
and dental limits in Class II malocclusion,2 according
to Angle’s classification scheme, the Class II Division
1 malocclusion most likely includes some degree of
incisor overjet. Overjet of 3 mm or more is present in
55% of the US population,3 suggesting that the incidence of Class II Division 1 malocclusion in the US
population is significant. The incidence of skeletal
Class II malocclusion in growing Brazilian patients
who seek orthodontic treatment is also significant.4,5
When radiographic and clinical assessments show
maxillary skeletal protrusion, the main treatment goal
for growing patients is correction of the abnormal maxillofacial relationship. This is often done by applying
orthopedic forces to the maxilla.
Headgear appliances are frequently used to apply
orthopedic forces to the maxilla. It has been demon-
aPrivate
Practice, Araraquara, Brazil.
of Orthodontics, Araraquara School of Dentistry, UNESP,
Araraquara, Brazil.
cPrivate Practice, São Paulo, Brazil.
†Deceased.
Reprint requests to: Luiz G. Gandini, Jr, Faculdade de Odontologia de
Araraquara, Departamento de Clinica Infantil, Rua Humaita, 1680, 14801-903
Araraquara, Brazil.
Submitted, October 1999; revised and accepted, August 2000.
Copyright © 2001 by the American Association of Orthodontists.
0889-5406/2001/$35.00 + 0 8/1/113266
doi:10.1067/mod.2001.113266
bDepartment
strated that headgear appliances produce important
changes in growing and adult animals.6-12 Distal movement of the maxillary molars occurs, and the normal
downward and forward growth of the nasomaxillary
complex can be changed, with significant resorption
occurring at the maxillary sutures and the flexure of the
cranial base.6-12 However, human studies have not
clearly demonstrated that maxillary basal bone changes
(excluding dentoalveolar bone changes) are possible.
Experimental models provide important biological
information, but they cannot be directly compared with
human studies. First, the force applied in patients is
lighter than the force used in animal protocols.13 Second, treated patients have malocclusions; experimental
animals do not. Third, the therapeutic limits are different because in human studies the application of force is
deemphasized when a dental Class I relationship is
achieved.13 Some authors claim that headgear therapy
causes skeletal maxillary changes in humans, but others deny its effects.2,13-17 In fact, dentoalveolar changes
have been clearly demonstrated concomitantly with the
downward relocation of the palatal plane in the anterior
region and remodeling of A point.2,14,15,17-21 In growing patients who have skeletal Class II malocclusions,
dentoalveolar headgear effects alone appear to be less
than the ideal goal; however, changes in the eruption
pattern may be a suitable mechanism to compensate for
skeletal maxillomandibular discrepancies.22,23
Normal dental eruption contributes significantly to
individual facial features.23 Posterior alveolar height is
directly related to mandibular rotation, which influences
531
532 Gandini et al
Table I. Age
American Journal of Orthodontics and Dentofacial Orthopedics
May 2001
and length of observation of treated patients and control subjects (in years)
Age T1
Group
Gender
Treated
Female
Male
Combined
Female
Male
Combined
Untreated
Age T2
Period observed
Sample
(N)
Mean
SD
Mean
SD
Mean
SD
26
19
45
18
12
30
10.9
11.2
11.0
10.2
10.3
10.2
1.1
1.5
1.3
1.8
1.3
1.6
14.4
15.0
14.6
11.5
11.7
11.6
1.1
1.6
1.4
1.5
1.2
1.4
3.5
3.8
3.6
1.3
1.4
1.3
1.3
1.3
1.3
0.8
0.9
0.8
the anteroposterior chin projection and, consequently,
the overall facial profile.24-26 Because dental eruption
does not strictly follow genetic patterns but is strongly
influenced by forces governing occlusal development,23
headgear appliances may affect the path and the degree
of dental eruption and thus the facial growth pattern.
Headgear dental effects can be academically divided
into horizontal and vertical components. Horizontal
effects, maintaining or moving maxillary first molars
distally, have been extensively described.2,13-21,27-33
However, vertical effects, especially when cervical
headgear appliances are used, are not well understood.
Extrusion of the maxillary molars has been described as
an important side effect of cervical headgear.2,21 In
other studies, extrusion of the maxillary molars has
been suspected or negated.16,19,27 Clinically, understanding the effects of cervical headgear is vital for its
correct application.
The purpose of this study was to evaluate skeletal and
dental changes in the maxillary and mandibular first molar
regions of growing Class II Division 1 patients treated
with cervical headgear and edgewise appliances.27
MATERIAL AND METHODS
A sample of 75 Brazilians of European descent
with ages ranging from 7.5 to 13.5 years was divided
into 2 groups: 45 subjects who had been treated with
cervical headgear and fixed edgewise appliances and
30 subjects who had not been treated (Table I).27
Records were collected retrospectively from the database of the Department of Orthodontics, Araraquara
Dental School (UNESP), São Paulo, Brazil. The
patients were treated by graduate students enrolled in
the orthodontic program.
The criteria for selection in the treated group were
that the patient have a Class II Division 1 (molar and
canine) malocclusion, with an overjet of 3 mm or
greater, that had been treated to a Class I relationship
with nonextraction therapy. The untreated group was
matched to the treated group by age, gender, and mal-
occlusion. Rejection criteria included poor-quality
records, craniofacial disorders, or previous orthodontic or orthopedic treatment. The cephalometric landmarks and measurements used for this study are
described in Table II.34
After the selection, the sample was classified
according to the Siriwat and Jarabak index.35 This
index is defined as the proportion between posterior
facial height (sella to gonion) and anterior facial height
(nasion to menton) × 100. Dolicocephalic subjects tend
to have a smaller proportion of posterior facial height
to anterior facial height than do mesocephalic subjects,
and brachicephalic subjects tend to have a larger proportion of posterior facial height to anterior facial
height than do mesocephalic individuals. The treated
group included 26 mesocephalic (58%), 6 dolicocephalic (13%), and 13 brachicephalic (29%) patients;
the untreated group included 20 mesocephalic (67%), 5
dolicocephalic (17%), and 5 brachicephalic (17%) subjects. The cephalometric indices measured in the
treated and untreated groups did not present a statistically significant difference between the groups. All the
cephalograms were traced by the same operator (M. S.
G.). Ten cephalograms were randomly chosen and
measured twice, 1 week apart, to calculate the systematic and random errors. Systematic error was not significant for any of the variables. Random error was calculated by the Dahlberg36 method
(ME = ∑
(x
1–2
x/2
)2n)
and ranged from 0.16 to 0.38. The changes were annualized, that is, alterations in millimeters were divided
by the observation period, which allowed for comparison of the changes observed in different intervals.
Cervical headgear appliances were adjusted to have
a 20° upward angulation of the headgear external bow,
to apply 400 g of force per side on the maxillary first
molars for 14 to 18 hours per day until a Class I relationship was achieved, and to apply the same force for
Gandini et al 533
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 119, Number 5
Table II. Definitions
of landmarks used
Landmark
Abbreviation
Definition
Sella turcica34
Nasion34
S
N
Sella–nasion –7°
Sella–nasion –7°
perpendicular
Upper first molar apex34
Upper first molar center of
resistance
Upper first molor cusp34
Lower first molar apex34
Lower first molar center of
resistance
Lower first molar cusp34
A point34
SN–7°
SN–7° Perp
Point in center of pituitary fossa of sphenoid bone
Junction of frontonasal suture at most posterior point on curve
at bridge of nose
Constructed line from SN line minus 7°
Constructed line 90° to SN minus 7°
UA
UCR
Mesiobuccal root apex of maxillary first molar
Furcation; union of 3 roots of maxillary first molar
UC
LA
LCR
Mesial cusp tip of maxillary first molar
Apex of the mesial root of mandibular first molar
Furcation; union of 2 roots of mandibular first molar
LC
A
B point34
B
SNA
SNB
ANB
SN.palatal plane
SN.occlusal plane
SN. mandibular plane
AO point
BO point
AO-BO distance
SNA
SNB
ANB
SN.PP
SN.OP
SN.GoMe
AO
BO
Wits
Mesial cusp tip of mandibular first molar
Most anterior point on curve of maxilla between anterior nasal
spine and supradentale
Most posterior point to line from infradentale to pogonion on
anterior surface of symphyseal outline of mandible
Angle between SN line and A point
Angle between SN line and B point
Difference between SNA and SNB angles
Angle between SN line and palatal plane
Angle between SN line and occlusal plane
Angle between SN line and mandibular plane
Projection of A point onto occlusal plane
Projection of B point onto occlusal plane
Distance from AO point to BO point
8 to 10 hours per day thereafter.27 The edgewise appliance prescription presented a –6° angulation (distal tip)
on the mandibular first molars.
Lateral cephalograms were taken at the beginning
and at the end of treatment. Cranial base (Fig 1), maxillary (Fig 2), and mandibular superimpositions were
performed to evaluate maxillary basal bone and dental
changes.37,38 Total superimposition was based on the
cranial base37 and included dental, dentoalveolar, and
maxillary basal bone changes. Partial superimpositions were used to evaluate dentoalveolar and dental
changes in the maxilla and the mandible because differentiation between the two was not possible. Maxillary basal bone changes were calculated as the difference between the total and the partial superimpositions
and represent the skeletal maxillary alteration or, in
other words, the repositioning of the upper jaw. The
results were represented by pitchfork diagrams.39 DFP
Plus software (Dentofacial Software, Toronto, Ontario,
Canada) was used to digitize the landmarks, and all
data were computed with the use of SPSS 9.0 software
(SPSS, Chicago, Ill).
Mean and SDs were used to describe central tendencies and dispersion of the variables. Variables that presented positive skewness or kurtosis were compared by
the Mann-Whitney U test. A Student t test was used for
variables that presented normal distribution (P < .05).
RESULTS
SNA and ANB angles and Wits distances changed
significantly (Table III). Moreover, slight counterclockwise rotation of the occlusal plane and clockwise
rotation of the palatal plane were observed (Table III).
Partial superimposition showed significant distal
relocation (negative sign) in the apex, center of resistance (furcation region), and cusp of the maxillary first
molars, including dentoalveolar and dental changes.
However, maxillary basal bone changes were not significantly different between the treated and untreated
groups (Table IV). Distal dental relocation was more
significant in the apex of the maxillary molars and
gradually decreased in the center of resistance and in
the molar cusp (Table IV). On the other hand, potential
skeletal changes were more pronounced in the region
of the molar cusp and gradually decreased in the region
of the center of resistance and the apex. Vertically,
none of the skeletal or dental changes in the apex, center of resistance, or molar cusp were significant (Table
V). Combined horizontal and vertical effects on the
maxillary molars are illustrated in Figure 3.
534 Gandini et al
Fig 1. Superimposition on cranial base (total superimposition). Superimposition made on bold areas. SN-7°
was determined to be the horizontal plane, with vertical
plane perpendicular to horizontal. ∆X is anteroposterior
difference and ∆Y is vertical difference between initial
and final cephalograms.
American Journal of Orthodontics and Dentofacial Orthopedics
May 2001
A
Mandibular skeletal or dental horizontal changes
were not significant when the treated and untreated
groups were compared (Table VI). Furthermore, the
mandibular molars did not present any significant vertical change (Table VII). The mandibular molar cusp
showed a tendency to move back, but the center of
resistance and especially the apex displayed a tendency
to move forward. Combined horizontal and vertical
effects on the mandibular molars are shown in Figure 4.
Horizontal maxillomandibular changes (skeletal and
dental) are shown in Figures 5 and 6, and vertical changes
(skeletal and dental) are shown in Figures 5 and 7.
DISCUSSION
Angular and linear changes support the notion that
cervical headgear appliances may have corrected the
Class II Division 1 malocclusion to Class I. Such a suggestion must be differentiated from a clear conclusion.
The limitation is based on a lack of independence
between our selection criteria and our results. Because
only Class II Division I patients who had been successfully treated by cervical headgear were included in our
sample, the possible efficiency of the cervical headgear
was known before the material was analyzed, and such
bias does not allow certainty.
The changes observed in A point do not necessarily confirm alterations of the maxillary basal bone
B
Fig 2. A and B, Superimposition on maxillary and
mandibular basal bone (partial superimposition). ∆X
refers to horizontal differences, and ∆Y refers to vertical
differences. Differences calculated for 3 areas: dental
apex, dental center of resistance (CR), and dental cusp.
because they may be due to distal relocation of the
dentoalveolar bone.40
The differences between experimental and clinical
studies are explained by the lack of matched protocols,
and results cannot be directly compared.13 Skeletal
Gandini et al 535
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 119, Number 5
Table III. Jaw
relationship (degrees per year and Wits in
millimeters per year)
Treated
Untreated
Variables
Mean
SD
SNA
SNB
ANB
SN.PP
SN.OP
SN.GoMe
Wits
–0.58
0.23
–0.81
0.25
–0.81
–0.07
–0.44
0.53
0.38
0.53
0.60
0.94
0.74
0.63
Mean
SD
Difference
0.28
0.35
–0.07
0.09
–0.46
0.03
0.12
0.66
0.69
0.60
0.76
0.78
1.2
0.71
0.86**
0.12
0.88**
0.16*
0.35*
0.1
0.56**
*P < .05; **P < .01.
Horizontal changes of maxillary molars (millimeters per year)
Table IV.
Treated
Fig 3. Each small diagram represents spatial change for
dental apex, dental center of resistance, and dental
cusp (in millimeters) for maxillary molar.
changes have been described in human studies; however,
the term skeletal may include alterations on the maxillary basal bone or in the dentoalveolar region.2,14-16,21
There is no clear description of the nature of the effect.
Our results did not show significant maxillary basal bone
differences (again, excluding dentoalveolar bone
changes) between the treated and the untreated groups,
supporting the results of Bernstein et al13 and Sandusky,17 who evaluated the repositioning of the maxillary basal bone, but not dentoalveolar changes.
Conversely, maxillary dental changes (including
dentoalveolar changes) were confirmed, agreeing with
results extensively demonstrated in other studies.13-20
Maxillary superimposition showed significant distal
relocation in all 3 dental areas (ie, the apex, the center
of resistance, and the molar cusp). The most significant
change occurred in the apex area, which suggests distal
apex tipping. Changes observed in the apex are due to
the 20° upward angulation of the headgear external
bow combined with the backward and downward pull
of the cervical headgear. The overall results show that
the headgear appliances maintained the position of the
maxillary molars but did not significantly influence the
Untreated
Variables
Mean
SD
Mean
SD
Difference
A, skeletal
A, dental
A, total
CR, skeletal
CR, dental
CR, total
C, skeletal
C, dental
C, total
0.46
–0.47
0.03
0.48
–0.34
0.17
0.45
–0.25
0.31
0.78
0.75
0.79
0.77
0.71
0.79
0.82
0.75
0.85
0.85
0.33
0.97
0.88
0.29
1.01
1.14
0.29
1.27
1.81
1.33
1.19
1.46
1.20
1.14
1.80
1.22
1.21
–0.39
–0.80*
–0.94*
–0.40
–0.63*
–0.93*
–0.69
–0.54*
–0.96*
*P < .01.
A, Apex; CR, center of resistance; C, cusp.
Vertical changes of maxillary molars (millimeters per year)
Table V.
Treated
Variables
A, skeletal
A, dental
A, total
CR, skeletal
CR, dental
CR, total
C, skeletal
C, dental
C, total
Untreated
Mean
SD
Mean
SD
Difference
0.66
1.16
1.82
0.67
1.19
1.87
0.73
1.16
1.86
1.10
0.75
0.73
1.09
0.76
0.71
0.94
0.78
0.71
0.49
1.14
1.81
0.44
1.07
1.78
0.45
1.12
1.77
1.09
0.95
1.07
0.93
0.94
1.03
1.12
1.02
1.08
0.17*
0.02*
–0.01*
0.23*
0.12*
0.09*
0.28*
0.04*
0.09*
*NS.
A, Apex; CR, center of resistance; C, cusp.
growth of the maxillary jaw. Our results could not
negate the null hypothesis that the headgear appliance
does not modify the growth pattern of the skeleton;
however, lack of rejection of the null hypothesis is not
proof that the opposite is true.
536 Gandini et al
American Journal of Orthodontics and Dentofacial Orthopedics
May 2001
Table VI. Horizontal
changes of mandibular molars (mil-
limeters per year)
Treated
Variables
A, skeletal
A, dental
A, total
CR, skeletal
CR, dental
CR, total
C, skeletal
C, dental
C, total
Untreated
Mean
SD
Mean
0.94
0.93
1.77
0.85
0.74
1.59
0.81
0.58
1.39
1.33
0.79
1.20
1.09
0.64
1.01
1.05
0.64
0.93
1.12
0.53
1.48
1.01
0.60
1.44
0.83
0.63
1.30
SD
1.30
0.99
1.25
1.33
0.86
1.21
1.18
0.88
1.03
Difference
–0.18*
0.40*
0.29*
–0.16*
0.14*
0.15*
–0.02*
–0.05*
0.09*
*Not significant.
A, Apex; CR, center of resistance; C, cusp.
Vertical changes of mandibular molars (millimeters per year)
Table VII.
Treated
Fig 4. Each small diagram represents spatial changes
for dental apex, dental center of resistance, and dental
cusp (in millimeters) for mandibular molar.
Untreated
Variables
Mean
SD
Mean
SD
Difference
A, skeletal
A, dental
A, total
CR, skeletal
CR, dental
CR, total
C, skeletal
C, dental
C, total
3.12
–0.83
2.29
3.09
–0.89
2.20
3.03
–0.99
2.03
1.25
0.80
0.86
1.25
0.84
0.85
1.25
0.89
0.80
3.06
–0.97
2.16
2.97
–0.92
2.05
2.87
–0.85
2.02
1.45
0.94
1.13
1.42
0.99
1.13
1.48
0.86
1.22
0.06*
0.14*
0.13*
0.12*
0.03*
0.15*
0.16*
–0.14*
0.01*
*Not significant.
A, Apex; CR, center of resistance; C, cusp.
Fig 5. Template describes numbers found in Figs 6 and 7.
Vertical extrusion of the maxillary molars has been
reported as a detrimental effect of cervical headgear
appliances.2,15,19,32 Such an effect was not confirmed by
our study, which agrees with the results of Boecler,16
Weinberger,20 and Ringenberg and Butts.28 Extrusion of
the maxillary molars is not necessarily unfavorable.
Brachicephalic patients have low anterior–posterior
height proportions, and extrusion of the maxillary molars
would compensate for excessive vertical mandibular
ramus growth, increasing the anterior facial height and
consequently improving the facial profile.25,26
Because the maxillary molars did not significantly
extrude, mandibular molars are not expected to change
vertically, and the mandibular rotation pattern is not
expected to change. Mandibular molars did not present
horizontal or vertical changes. The tendency for forward movement of the mandibular molar apex is possibly due to the 6° angulation in the edgewise prescription and may not be related to headgear mechanics.
Clinically, our results demonstrated that the cervical
headgear appliances corrected the Class II malocclusion
to a Class I relationship (molars and canines) in the successful cases. The lack of statistically significant maxillary basal bone changes does not necessarily mean that
the treatment was unsuccessful because dentoalveolar
corrections may camouflage skeletal discrepancies.1
A limitation of this study was that the changes were
annualized, a choice of method based on the assumption
Gandini et al 537
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 119, Number 5
Fig 6. Horizontal changes for the maxillary basal bone, mandibular basal bone, and maxillary and
mandibular first molars. Arrows represent repositioning in space (forward or backward). Circles represent differences between maxillary and mandibular arches. Difference diagram shows differences
between untreated and treated groups.
Fig 7. Vertical changes for maxillary basal bone, mandibular basal bone, and maxillary and madibular first molars. Arrows represent the repositioning in space (upward or downward). Circles represent
differences between the maxillary and mandibular arches. Difference diagram shows differences
between untreated and treated groups.
that growth rate is linear, which may not be true. However, this was the best way to compare the experimental
and the control groups that had had significantly different
observation periods. Furthermore, our data do not evaluate pure headgear effects because lateral cephalograms
taken immediately after the Class I relationship had been
achieved were not available. The overall change for the
entire period of treatment may hide the headgear effects
because patients may have resumed normal growth patterns during the period in which headgear was used only
to maintain the achieved Class I relationship.
Another limitation is the potential treatment bias;
mesocephalic or brachicephalic patients, for whom the
cervical headgear is indicated, may have less of a tendency for extrusion of the maxillary molars because
they have different masticatory forces than do dolicocephalic patients. Future studies evaluating a larger
number of all facial types are suggested.
CONCLUSION
In this study, statistically significant maxillary basal
bone changes did not occur. Cervical headgear appliances corrected the Class II Division 1 malocclusion to a
Class I relationship by maintaining the maxillary first
molars and redirecting dentoalveolar growth in the maxilla, rather than by significantly changing the growth of
the maxillary jaw base. Vertical changes were not supported by our data. Absence of statistically significant
extrusion of the maxillary molars after the use of cervical
headgear in patients with Class II Division 1 malocclusion is a valuable piece of information for the clinician.
REFERENCES
1. Proffit WR, Fields HW, Ackerman JL, Sinclair PM, Thomas PM,
Tulloch JFC. Contemporary orthodontics. St. Louis: Mosby–Year
Book; 1993.
2. Baumrind S, Korn EL, Isaacson RJ, West EE, Molthen R. Quantitive analysis of the orthodontic and orthopedic effects of maxillary traction. Am J Orthod 1983;84:384-98.
3. Brunelle JA, Bhat M, Lipton JA. Prevalence and distribution of
selected occlusal characteristics in the US population, 19881991. J Dent Res 1996;75 Spec No:706-13.
4. Gandini MS. Estudo da oclusão dentária de escolares na fase da
dentadura mista da cidade de Araraquara, São Paulo [thesis]. 1993.
5. Silva Filho OG, Freitas SF, Cavassan AO. Prevalência de oclusão
normal e maloclusão na dentadura mista em escolares da cidade de
Bauru (São Paulo). Rev Odontol Unive São Paulo 1989;43:287-90.
6. Droschl H. The effect of heavy orthopedic forces on the maxilla
538 Gandini et al
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
in the growing Saimiri sciureus (squirrel monkey). Am J Orthod
1973;63:449-61.
Droschl H. The effect of heavy orthopedic forces on the sutures
of facial bones. Angle Orthod 1975;45:26-33.
Elder JR, Tuenge RH. Cepahlometric and histologic changes
produced by extraoral high-pull traction to the maxilla in
Macaca mulatta. Am J Orthod 1974;66:599-617.
Thompson RW. Extraoral high-pull forces with rapid palatal
expansion in the Macaca mulatta. Am J Orthod 1974;66:302-17.
Meldrum RJ. Alterations in the upper facial growth of Macaca
mulatta resulting from high-pull headgear. Am J Orthod
1975;67:393-411.
Triftshauser R, Walters RD. Cervical retraction of the maxillae
in the Macaca mulatta monkey using heavy orthopedic force.
Angle Orthod 1976;46:37-46.
Brandt HC, Shapiro PA, Kokich VG. Experimental and postexperimental effects of posteriorly directed extraoral traction in the
adult Macaca fascicularis. Am J Orthod 1979;75:301-17.
Bernstein L, Ulbrich WR, Gianelly A. Orthopedics versus orthodontics in Class II treatment: an implant study. Am J Orthod
1977;72:549-59.
Klein PL. An evaluation of cervical traction on maxilla and the
upper first permanent molar. Angle Orthod 1957;27:61-8.
Jakobsson SO. Cephalometric evaluation of treatment effect on
Class II, division 1 malocclusions. Am J Orthod 1967:53:446-57.
Boecler PR. Skeletal changes associated with extraoral appliance therapy: an evaluation of 200 consecutively treated cases.
Angle Orthod 1989;59:263-70.
Sandusky WC. Cephalometric evaluation of the effects of the
Kloehn type of cervical traction used as an auxiliary with the
edgewise mechanism following Tweed’s principles for correction of Class II, division 1 malocclusion. Am J Orthod
1965;51:262-87.
King EW. Cervical anchorage in Class II, division 1 treatment, a
cephalometric appraisal. Angle Orthod 1957;27:98-104.
Poulton DR. The influence of extraoral traction. Am J Orthod
1967;53:8-18.
Weinberger TW. Extra-oral traction and functional appliances—
a cephalometric comparison. Br J Orthod 1974;1:35-9.
Brown PA. A cephalometric evaluation of high-pull molar headgear
and face bow neck strap therapy. Am J Orthod 1978:74;621-32.
Solow B. The dentoalveolar compensatory mechanism: background and clinical implications. Br J Orthod 1980;7:145-61.
Solow B, Iseri H. Maxillary growth revisited: an update based on
recent implant studies. In: Davidovitch Z, Norton LA, editors.
Biological mechanisms of tooth movement and craniofacial
adaptation. Boston: Harvard Society for the Advancement of
Orthodontics; 1996.
Schudy F. Cant of the occlusal plane and axial inclinations of
teeth. Angle Orthod 1963;33:69-82.
Schudy F. The rotation of the mandible resulting from growth: its
implications in orthodontic treatment. Angle Orthod 1965;35:36-50.
Bjork A, Skieller V. Facial development and tooth eruption. An
implant study at the age of puberty. Am J Orthod 1972;62:339-83.
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.
Ringenberg QM, Butts WC. A controlled cephalometric evaluation of single arch cervical traction therapy. Am J Orthod
1970;57:179-85.
Barton JJ. High-pull headgear versus cervical traction: a
cephalometric comparison. Am J Orthod 1972;62:517-29.
American Journal of Orthodontics and Dentofacial Orthopedics
May 2001
30. Wieslander L. Early or late cervical traction therapy of Class II
malocclusion in the mixed dentition. Am J Orthod 1975;67:432-9.
31. Chaconas SJ, Caputo AA, Davis JC. The effects of orthopedic
forces on the craniofacial complex utilizing cervical and headgear appliances. Am J Orthod 1976;69:527-39.
32. Cangialosi TJ, Meistrell ME Jr, Leung MA, Ko JY. A cephalometric appraisal of edgewise Class II nonextraction treatment with
extraoral force. Am J Orthod Dentofacial Orthop 1988;93:315-24.
33. Aelbers CMF. Orthopedics in orthodontics: Part I, fiction or
reality—a review of the literature. Am J Orthod Dentofacial
Orthop 1996;110:513-9.
34. Behrents RG. An atlas of growth in the aging craniofacial skeleton. Monograph 18. Ann Arbor: Center for Human Growth and
Development, University of Michigan; 1985. p. 1-13.
35. Siriwat P, Jarabak J. Malocclusion and facial morphology. Is
there a relationship? An epidemiologic study. Angle Orthod
1985;55:127-38.
36. Dahlberg G. Statistical methods for medical and biological students. London: George Allen and Unwin; 1940.
37. Bjork A, Skieller V. Normal and abnormal growth of the
mandible. A synthesis of longitudinal cephalometric implant
studies over a period of 25 years. Eur J Orthod 1983;5:1-46.
38. Bjork A, Skieller V. Roentgen cephalometric growth analysis of
the maxilla. Trans Eur Orthod Soc 1977;53:51-5.
39. Goldrich HN. The effects of a modified maxillary splint combined with a high pull headgear [thesis]. Dallas: Baylor College
of Dentistry; 1994.
40. Taylor CM. Changes in the relationship of nasion, point A, and
point B and the effect upon ANB. Am J Orthod 1969;56:143-63.
COMMENTARY: RETROSPECTIVE STUDY
REQUIRES CAREFUL INTERPRETATION
The authors of this article have collected data from
two types of Class II Division 1 subjects: those who
wore cervical headgear and those who remained
untreated. Although the groups were matched “by age,
gender, and malocclusion,” we are not told specifically
what that means. At the end of the observation periods,
certain differences between the groups were observed,
and the authors suggest that these differences can be
attributed to the headgear. However, the use of cervical
headgear was not the only difference between the
groups. The time interval for the observation group was
longer than that for the control subjects, and only those
subjects who successfully achieved a Class I relationship were allowed to remain in the treated group. This
limits the conclusions as to a “cause” that can appropriately be drawn from the study.
The error the authors have made is somewhat subtle
and is extremely common—my impression is that the
same mistake has been made in more than half the clinical studies published during the past 30 years. It is
good that the sophistication of our readers is such that
we now detect errors of this sort. On the other hand, I
believe it is dangerous to dismiss a report out of hand
because it is not perfect. It is all very well to say that, on
theoretical grounds, randomized clinical trials are the
Gandini et al 539
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 119, Number 5
gold standard in human research and, by implication,
are the best or even the only way to go. Twelve years
ago, I believed that the best answers for orthodontics
could be achieved only by textbook-perfect prospective
studies, and I strongly disagreed with those who
claimed such studies were not feasible in orthodontics.
But experience has convinced me that I was wrong.
To be sure, prospective clinical trials play a valuable role, but they are no more likely to produce perfect
answers than are well-conducted retrospective studies.
Indeed, there are two practical reasons why orthodontics is unlikely to see large prospective trials in the near
future. The first is that, to be meaningful, prospective
trials would have to span at least 15 to 20 years,
whereas experience shows us that treatment strategies
change sufficiently over such a time interval to preclude or seriously limit the generalization of results
from one period to another. The second reason is that
funding for well-controlled long-term prospective trials
in orthodontics is simply not available—and is not
likely to become available during the next decade.
This leaves us, as a specialty, with the task of learning how to draw limited conclusions from limited studies, albeit with greater sophistication than was employed
earlier. Future studies need to address questions of selection bias much more carefully than the present authors
have done. There are strategies for retrospectively controlling for or minimizing the effects of such bias, both
at the beginning of studies and, as in the present case, at
the end—by limiting our post hoc generalizations.
It seems evident to this reviewer that much careful
work has gone into this study, and some meaningful findings regarding patterns of change within the treated group
can be gleaned from it. However, I believe there is not sufficient evidence to allow the observed between-group differences to be ascribed to the use of cervical headgear.
Shelly Baumrind, DDS, MS
University of the Pacific, San Francisco, Calif
0889-5406/2001/$35.00 + 0 8/1/114283
doi:10.1067/mod.2001.114283
Am J Orthod Dentofacial Orthop 2001;119:538-9
Copyright © 2001 by the American Association of Orthodontists.
RECEIVE THE JOURNAL’S TABLE OF CONTENTS EACH MONTH BY E-MAIL
To receive the tables of contents by e-mail, send an e-mail message to
[email protected]
Leave the subject line blank, and type the following as the body of your message:
Subscribe ajodo_toc
You can also sign up through our website at http://www.mosby.com/ajodo.
You will receive an e-mail message confirming that you have been added to the mailing
list. Note that TOC e-mails will be sent when a new issue is posted to the website.