ORIGINAL ARTICLE

Profile changes in orthodontic patients treated

with mandibular advancement surgery
Susan T. Tsang,a Leland R. McFadden,b William A. Wiltshire,c Neeraj Pershad,d and Allan B. Bakere
Winnipeg, Manitoba, and St Catharines, Ontario, Canada

Introduction: The potential to improve facial esthetics is often the deciding factor in treatment planning of
borderline orthodontic patients who can be treated with either orthognathic surgery or dental camouflage.
The purpose of this study was to determine the degree of skeletal and soft-tissue Class II disharmony
necessary before a significant esthetic benefit is derived from mandibular advancement surgery. Methods:
Twenty laypeople, 20 orthodontists, and 20 oral surgeons rated the attractiveness of before and after
treatment profiles of 20 mandibular advancement patients using a 5-point Likert scale. The Spearman rank
correlation tested for relationships between amount of profile change and varying pretreatment ANB and
profile angles. Plots of the distribution of profile changes with varying ANB and profile angles were then
examined. Results: There was a tendency for inverse correlations between profile change and profile angle,
and for positive correlations between profile change and ANB angles, but only the relationship between
profile change and ANB angles judged by the orthodontists was statistically significant (P ⬍0.05).
Orthodontists, oral surgeons, and laypeople found that profiles consistently improved when profile angles
were ⱕ159°, ⱕ158°, and ⱕ157°, respectively. Orthodontists and oral surgeons found profiles consistently
improved when ANB angles were ⱖ5.5° and ⱖ6.5°, respectively, whereas laypeople showed no trend
between ANB angle and profile change. The incidence of having less desirable profiles after treatment was
2.6 to 5.0 times higher when the pretreatment profile angles were larger than the threshold profile angles, and
4.5 to 7.9 times higher when the pretreatment ANB angles were less than threshold ANB angles.
Conclusions: Pretreatment profile angles ⬍160° and ANB angles ⬎6° are necessary for profiles to be
consistently perceived as improved after surgery and to minimize the incidence of the profile worsening after
treatment. (Am J Orthod Dentofacial Orthop 2009;135:66-72)

O
rthodontic treatment of up to 10% of patients
may be borderline between dental camouflage
and orthognathic surgery.1 The decision-mak-
ing process is influenced by variables such as esthetics;
patient desires, values, and cultural differences; orth-
odontist preferences; surgical expertise; and financial
considerations.2 When considering the effect of treat-
ment on facial esthetics, when should one treatment
method be preferred over the other in borderline cases?
Nongrowing skeletal Class II patients can be treated
with either dental camouflage or orthognathic surgery
to advance the mandible and improve the convexity of
a
Private practice, Winnipeg, Manitoba, Canada.
b
Assistant professor, Division of Orthodontics, University of Manitoba, Win-
nipeg, Manitoba, Canada.
c
Professor and chair, Division of Orthodontics, University of Manitoba,
Winnipeg, Manitoba, Canada.
d
Private practice, St. Catharines and Niagara Falls, Ontario, Canada.
e
Assistant professor, Division of Orthodontics, University of Manitoba, Win-
nipeg, Manitoba, Canada.
Reprint requests to: Susan Tsang, c/o University of Manitoba, Faculty of
Dentistry, 780 Bannatyne Ave, Winnipeg, Manitoba, R3E 0W2, Canada; email,
s_tsang2@hotmail.com.
Submitted, October 2006; revised and accepted, January 2007.
0889-5406/$36.00
Copyright © 2009 by the American Association of Orthodontists.
doi:10.1016/j.ajodo.2007.01.033
66
the profile. Although treatment can improve facial
esthetics, lack of improvement or even worsening of
esthetics can also occur with either treatment modality.
Class II patients with poorer pretreatment esthetics and
greater surgical advancements often benefit the most
esthetically from treatment.3-5 Shelley et al5 reported
that laypeople and orthodontic residents found a pre-
treatment ANB angle of 6° as the threshold for predict-
ing whether mandibular advancement surgery will pro-
duce significant esthetic changes. Profiles improved by
45% in patients with an initial ANB angle ⱖ6°, but
those with an initial ANB ⬍6° did not have statistically
significant overall posttreatment changes, and half of
them were perceived to have worse esthetics.
Because of the large variation in soft tissues,
measurements and guidelines based on hard-tissue
relationships might not necessarily correlate to the
actual soft tissues. Park and Burstone6 reported patients
with similar hard-tissue convexity angles but soft-tissue
profile angle (PA) values that differed by as much as
22°. The PA—the obtuse angle formed by lines con-
necting soft-tissue glabella, subnasale, and soft-tissue
pogonion— describes the general harmony of the up-
per, middle, and lower soft-tissue profile, and a Class I
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 135, Number 1
skeletal relationship is represented by a PA of 165° to
175°.7 Angles below or above this range represent
Class II or Class III skeletal relationships, respectively.
The PA has been suggested as the most critical deter-
minant of the need for anteroposterior surgical correc-
tion because variations in soft-tissue thickness are not
usually responsible for deviations beyond the normal
range, and departures from the norm indicate signifi-
cant underlying skeletal disharmony.7
When considering mandibular advancement to im-
prove facial appearance in borderline patients, it is
important to identify those who will most likely have a
clinically noticeable profile improvement. Therefore,
the purpose of this study was to establish guidelines to
recommend mandibular advancement surgery to pro-
duce noticeable profile improvement based on the
severity of the initial hard- and soft-tissue anteroposte-
rior disharmony.
MATERIAL AND METHODS
Charts of 20 patients treated with a combination of
orthodontics and mandibular advancement surgery at
the University of Manitoba Graduate Orthodontics
Clinic, Winnipeg, Manitoba, Canada, were selected.
Diagnostic records were taken before treatment (T1)
and at appliance removal (T2). No preference was
given to the amount of initial skeletal anteroposterior
discrepancy, patient sex or ethnicity, tooth extractions,
method of skeletal fixation, use of a surgical splint, or
type of occlusion at the end of treatment.
The sample was limited to patients with a mandib-
ular plane to sella-nasion angle (MPA) at T1 of 33° ⫾
6° to eliminate those with excessive or deficient vertical
face heights.8 Patients with craniofacial anomalies,
maxillary surgery, or genioplasty were excluded.
Charts with missing lateral cephalometric radiographs
and those with unclear soft-tissue contours of the
forehead and profile were also excluded.
T1 and T2 lateral cephalometric radiographs were
hand traced on acetate paper. The soft-tissue profile
was traced on acetate from a point superior to glabella
to a point past the neck-throat point. One operator
(S.T.) manually measured PA and ANB angles to the
nearest 0.5° using a protractor with 1° increments. To
orientate a subject’s T1 and T2 tracings to the same
head position, the 2 tracings were superimposed on the
anterior cranial base to confirm whether the Frankfort
horizontal (FH) planes were coincident.5 If the FH
planes at T1 and T2 were not coincident, the FH plane
at T1 was transferred to the T2 tracing. Profiles were
scanned at 200 dpi on a flatbed scanner (CanoScan Lide
30, Canon, Mississauga, Ontario, Canada) and im-
ported into Jasc Paint Shop Pro (Jasc Software, version
Tsang et al 67
8.1, Ottawa, Ontario, Canada). The images were ori-
ented with FH parallel to the top edge of the screen and
filled in with black to produce silhouettes (Fig 1).
The 20 pretreatment and 20 posttreatment profile
silhouettes were randomized and placed into a Power-
Point presentation (Microsoft, Mississauga, Ontario,
Canada). Three unrelated silhouettes were added at the
beginning of the slide show to familiarize participants
with the procedure but were not used for subsequent
data analysis. Seven of the 40 silhouettes were ran-
domly placed in the presentation a second time to test
for intraobserver reliability. The participants were un-
aware that there were duplicate images. Responses
from the first time the profiles were rated were used to
calculate changes in profile. The slide show was viewed
on a computer monitor, with the 3 introductory slides
shown for 20, 15, and 10 seconds each, and the
remaining 47 slides shown for 10 seconds each.
Twenty orthodontists, 20 oral and maxillofacial
surgeons who perform orthognathic surgery, and 20
laypeople without dental-related training comprised the
3 panels. They evaluated the 50 profile silhouettes and
gave each an esthetic score on a 5-point Likert scale
having the terms “very unattractive,” “unattractive,”
“fair,” “attractive,” and “very attractive.”5,9
Statistical analyses were performed with statistical
software (version 9.1.3, SAS Institute, Cary, NC).
Treatment changes in MPA, SNA, SNB, ANB, and PA
from T1 to T2 were individually evaluated by using 5
paired Student t tests for repeated measurements (mul-
tiple univariate analysis). With 5 independent t tests,
statistical significance was taken at P ⬍0.05 divided by
5, or 0.01 level. Pearson correlation tests were con-
ducted between pretreatment ANB angle and PA using
a significance level of P ⬍0.05.
Changes in patient profiles were calculated as the
difference between T2 and T1 esthetic scores. Differ-
ences ⬎0 indicated profile improvement with treat-
ment, scores ⬍0 indicated worsening of the profile, and
differences equal to zero indicated no change with
treatment. With 20 persons in each panel evaluating 20
profiles, there were 400 judgments from each panel.
Each group of 400 judgments was plotted on 3-dimen-
sional scatterplots to graphically depict the frequency
of profile changes at varying ANB and PA at T1.
Similar to the study of Shelly et al,5 the plots were
examined for threshold ANB and PA values where the
esthetic scores from T1 to T2 consistently improved
(⬍1% of the 400 points are ⬍0). Based on the number
of evaluations that were either above or below the
threshold, the incidence of profile worsening at the
ANB and PA above or below the threshold was
calculated as the proportion in each circumstance ⬍0.
68 Tsang et al
Fig 1. Example of T1 and T2 profile silhouettes of a
subject treated with orthodontics and orthognathic sur-
gery.
Because of the nonparametric nature of this data,
the Spearman rank correlation test was used to deter-
mine the relationship between the changes in esthetic
scores and ANB and PA values at T1, and to assess
intraobserver reliability. A higher correlation between
an evaluator’s first and second ratings of a profile
implied better intraevaluator reliability.
RESULTS
Sixteen orthodontists (80%), 20 oral surgeons
(100%), and 10 laypeople (50%) were men. Four
orthodontists (20%) and 10 people (50%) were women,
but no female oral surgeons participated.
The surgical sample consisted of 3 (15%) male and
17 (85%) female subjects. The average ages of the
sample were 21 years 2 months at T1 and 24 years 3
months at T2. Nine patients in the surgical sample were
treated with mandibular extractions of either the first
premolars or a mandibular incisor. Five of the 9
extraction patients also had maxillary extractions. The
purpose of the extractions was not assessed.
Descriptive statistics for the cephalometric measure-
ments are given in Table I. The distributions of pretreat-
ment ANB and PA values are shown in Figures 2 and 3.
The Pearson correlation test found a nonstatistically sig-
nificant inverse correlation between pretreatment ANB
and PA (r ⫽ ⫺.40, P ⫽ 0.08).
Ninety percent of the lay evaluators and 85% of the
orthodontists and oral surgeons found negative correla-
tions between the patients’ changes in profile and initial
PA. A negative correlation implied that, as the initial PA
decreases (ie, profile becomes more convex), there is
increasing improvement in the profile after surgery. Cor-
relation values were ⫺0.64 to 0.08 among the lay evalu-
ators, ⫺0.39 to 0.23 among the orthodontists, and ⫺0.38
to 0.13 among the oral surgeons. Correlations between PA
American Journal of Orthodontics and Dentofacial Orthopedics
January 2009
and changes in the profile were statistically significant
(P ⬍0.05) for 2 evaluators in the lay group.
Negative profile changes accounted for 14 of the
400 orthodontists’ evaluations; 86% of these negative
changes occurred when pretreatment PA values were
ⱖ160° (Fig 4). When initial PA values were ⱕ159°,
profiles tended to consistently improve with treatment,
as shown by the large number of points in the shaded
area in Figure 4. Similarly, oral surgeons found 32 of
the 400 evaluations to be negative, but the most
negative changes (93.8%) occurred when the PA was
ⱖ159° (Fig 5). Laypeople found 55 of the 400 evalu-
ations to be negative, with 94.5% in those with pre-
treatment PA values ⱖ158° (Fig 6).
When initial PA values were above the threshold
PA, orthodontists found the incidence of negative
profile changes after treatment to be 2.6 times greater
than when the initial PA was less than or equal to the
thresholds (Table II). The oral surgeons found a 5-fold
increase in incidence of negative changes, whereas the
laypeople found a 4.3-fold increase when initial PA
values were above their threshold PA values.
Seventy-five percent of the lay evaluators and 90% of
the orthodontists and oral surgeons found positive corre-
lations between the patients’ changes in profile and their
initial ANB angles, implying that as initial ANB angle
increases (ie, increasing severity of the Class II skeletal
relationship), there is an improvement in profile after
surgery. Correlation values were ⫺0.21 to 0.50 among the
laypeople, ⫺0.26 to 0.6 among the orthodontists, and
⫺0.05 to 0.65 among the oral surgeons. These correla-
tions between changes in profile scores and initial ANB
were statistically significant (P ⬍0.05) for 1 oral surgeon,
3 laypeople, and 8 orthodontists.
Deterioration of esthetics after treatment occurred,
particularly at less severe ANB angles and milder Class II
skeletal relationships. When the orthodontists evaluated
the profiles, 14 of the 400 profile changes were negative,
but 11 (78.5%) occurred in patients with ANB angles of
ⱕ5° (Fig 7). When the oral surgeons evaluated profiles,
32 of the 400 evaluations had negative profile changes,
with 90.6% of these negative changes in those with an
initial ANB angle ⱕ6° (Fig 8). Orthodontists and oral
surgeons, therefore, found profiles consistently improved
when ANB angles were ⱖ5.5° and 6.5°, respectively. The
laypeople found 55 of the 400 evaluations to be negative,
but these scores occurred across a wide distribution of
ANB angles, and only at ANB angles of 9° or higher was
the occurrence of negative changes fewer than 3 (Fig 9).
There was no break in the distribution to indicate what
degree of ANB angle consistently improves or potentially
worsens with treatment.
When pretreatment ANB angles were less than the
American Journal of Orthodontics and Dentofacial Orthopedics Tsang et al 69
Volume 135, Number 1

Table I. Descriptive statistics (°) of the surgical sample (n ⫽ 20)

Initial Final Treatment change

Angle Mean (SD) Min Max Mean (SD) Min Max Mean (SD) Min Max

MPA 32.7 (3.1) 28.0 38.0 36.3 (3.6) 29.0 43.0 3.6 (2.7)* ⫺1.0 9.0
SNA 80.3 (4.1) 74.0 92.0 80.3 (3.6) 75.0 91.0 ⫺0.1 (2.1) ⫺5.0 4.5
SNB 74.0 (3.6) 69.0 83.0 76.6 (3.0) 72.0 84.0 2.5 (1.8)* ⫺2.0 6.5
ANB 6.3 (1.7) 3.5 10.0 3.7 (1.6) 1.0 7.0 ⫺2.6 (1.2)* ⫺0.5 ⫺5.0
PA 161.0 (4.6) 152.0 170.0 165.3 (4.3) 156.0 172.0 4.3 (3.2)* ⫺2.0 12.0

Min, Minimum; Max, maximum.

*Significant at P ⬍ 0.01 level.

Fig 4. Orthodontists’ evaluations of profiles with vary-

ing initial PA values (n ⫽ 400). PA values ⱕ159° con-
sistently had a positive change in profile after treatment
(shaded area).

Fig 2. Distribution of pretreatment ANB in the surgical

sample (n ⫽ 20).

Fig 5. Oral surgeons’ evaluations of profiles with vary-

ing initial PA values (n ⫽ 400). PA values ⱕ158° con-
sistently had a positive change in profile after treatment
(shaded area).

orthodontist with no correlation was disregarded for this

Fig 3. Distribution of pretreatment PA in the surgical section of the analysis, the remaining 19 orthodontists had
sample (n ⫽ 20).
the narrowest range of correlations—from 0.69 to 1.0.

threshold ANB angle, the orthodontists and oral sur- DISCUSSION

geons found the incidence of negative profile changes
to be 4.5 and 7.9 times greater, respectively, than when
the pretreatment ANB angles were greater than or equal
to the threshold value (Table III).
For the laypeople, the Spearman correlation coeffi-
cients had the widest range, from 0.26 to 0.91. Correla-
tions among oral surgeons ranged from 0.44 to 1. Orth-
odontists had correlations from 0 to 1.0, but, if 1
The surgical sample was predominantly female
(85%), consistent with studies reporting that women are
more likely than men to seek orthodontic treatment and
accept orthognathic surgery treatment plans.10 With
particular facial characteristics distinctive to either sex,
profiles considered attractive in men might be less
desirable in women, and vice versa.11-13 Although our
evaluators were unaware of the sex of the patients,
70 Tsang et al American Journal of Orthodontics and Dentofacial Orthopedics
January 2009

Fig 6. Laypeople’s evaluations of profiles with varying Fig 8. Oral surgeons’ evaluations of profiles with vary-
initial PA values (n ⫽ 400). PA values ⱕ157° consistently ing initial ANB values (n ⫽ 400). ANB values ⱖ6.5°
had a positive change in profile after treatment (shaded consistently had a positive change in profile after treat-
area). ment (shaded area).

Table II. Incidence of negative profile changes when

pretreatment PA values are above and below threshold PA
Incidence of negative profile
changes

Threshold Less than or equal Greater than

PA to the threshold the threshold

Orthodontists 159° 1.67% 4.29%

Oral surgeons 158° 2.00% 10.00%
Laypeople 157° 3.75% 16.25% Fig 9. Laypeople’s evaluations of profiles with varying
initial ANB values (n ⫽ 400). Negative profile changes
occurred over a wide rage of ANB values.

Table III. Incidence of negative profile change when

pretreatment ANB values are above and below thresh-
old ANB
Incidence of negative profile
changes

Threshold Less than Greater than or equal

ANB threshold to threshold

Fig 7. Orthodontists’ evaluations of profiles with vary-
ing initial ANB values (n ⫽ 400). ANB values ⱖ5.5°
consistently had a positive change in profile after treat-
ment (shaded area).
DeSmit and Dermaut14 reported no significant differ-
ences in the esthetic evaluation of profiles regardless of
whether they are evaluated as male or female.
The nonstatistically significant relationship be-
tween the ANB and PA reinforces the importance of
expanding treatment planning and cephalometrics be-
yond just the hard-tissue skeleton, particularly when
treatment objectives are aimed both to correct a skeletal
malocclusion and change the soft-tissue profile. The
definition of PA by Arnett and Bergman7 would clas-
sify the 161° mean PA at T1 of the sample as indicating
underlying Class II skeletal patterns that are most
appropriately treated surgically. As expected after man-
dibular advancement, treatment had no effect on max-
Orthodontists 5.5° 6.11% 1.36%
Oral surgeons 6.5° 13.2% 1.67%
illa position, but it increased mandibular prominence
and decreased facial convexity.
Limiting the study sample to those with an average
MPA was done to exclude extreme forward or back-
ward rotators and to minimize the confounding effects
of surgically induced vertical changes. Increases in
facial height with mandibular advancement are favor-
able with a short face height but detrimental if the
preexisting excessive face height is further increased.
Vertical profile changes have a greater effect on the
perception of a profile than anteroposterior changes,
with negative effects particularly pronounced when
face height is excessively increased.14
Figure 10 shows a patient from the sample who had
minimal or negative profile changes with treatment,
American Journal of Orthodontics and Dentofacial Orthopedics Tsang et al 71
Volume 135, Number 1

compared with a patient whose profile significantly

improved with treatment. Negative correlations be-
tween profile change with PA and positive correlations
between profile change with ANB angle reflect trends
for patients with more acute pretreatment PA values
and severe Class II skeletal relationships to have greater
esthetic improvement after surgery. Correlations, how-
ever, were statistically significant for only the orth-
odontist group and ANB angle, with 40% of the
correlations reaching statistical significance. The lack
of clinical significance in the correlations might be due
to the ordinal scale requiring nonparametric statistics.
Studies with visual analog scales15-18 can use more
powerful parametric statistics, but it cannot be assumed
that anchor terms such as “very unattractive” and “very
attractive” are interpreted similarly by different people,
that identical positions on the scale express comparable
intensities of feelings of different people, or that a multiple
of a particular rating represents a multiple of the intensity
of the feeling.15,17-19 A visual analog scale also does not
identify how many millimeters of difference are required
for a significant clinical difference.18,19
Arnett and Bergman7 recommended a PA of 165°
as the threshold for requiring surgical treatment, but we Fig 10. A, A patient with minimal or negative facial
changes with treatment; B, a patient with a positive
found that a more acute PA threshold of 160° might be
change in profile after treatment.
necessary for profiles to be consistently perceived as
improved with surgery. PA values ⱖ160° were associ-
ated with a 2.6- to 5-fold increase in the incidence of skeletal relationships, whereas the laypeople were less
profile deterioration. Profile worsening occurred 4.5 to aware of changes and rated some patients with severe
7.9 times more frequently in patients with initial ANB initial Class II skeletal patterns as still having poor
angles less than 5.5° and 6.5° when evaluated by esthetics at T2. For both PA and ANB, orthodontists had
orthodontists and oral surgeons, respectively. This is the smallest proportion of subjects with negative profile
similar to the findings of Shelly et al,5 thus supporting changes (3.5%), followed by oral surgeons (8%) and
the threshold ANB angle of 6° as a guideline in esthetic laypeople (13%-14%), suggesting that orthodontists most
treatment planning of Class II patients. critically evaluate profiles and notice subtle changes,
In spite of the increased risk, however, profile whereas laypeople are less likely to see changes. Detailed
deterioration is a relatively uncommon occurrence, and information on the characteristics of our evaluators (age,
83% to 95% of profiles more orthognathic than the PA education, cultural background, length of time since grad-
thresholds and 86% to 93% of patients with less severe uation) was not collected in this study, but these variables
initial ANB values than the threshold still esthetically can influence treatment planning decisions and should be
benefited from treatment. examined in future studies.
Nonetheless, as profiles become more orthognathic, Other studies have reported that laypersons rank 25%
greater consideration should be directed toward non- of patients as “unimproved” even when there are great
surgical options, if feasible. Cassidy et al20 showed that surgical changes, whereas dental professionals rank only
Class II Division 1 adults who were borderline between 25% of patients with the least amount of surgical change
camouflage and surgery had no significant differences as “unimproved.”3 Other found that dental professionals
in satisfaction with their profile improvement, temporo- prefer Class I profiles more than the general public,
mandibular joint function, or incisor stability with possibly because the professionals are conditioned to
either treatment choice; they suggest that orthodontic focus more on the features in the lower third of the face,
treatment alone is more appropriate and conservative whereas laypersons rarely focus on profiles and are more
for many borderline Class II adults. influenced by other general facial features.18
Dental specialists were more critical and noted Orthodontists tended to be the most reliable in their
changes in the T2 profiles of patients with milder Class II ratings, with higher intraobserver correlations and the
72 Tsang et al
narrowest range. The laypeople showed a tendency to
be less consistent than the other 2 groups in how they
rated the profiles, with lower correlations and the
widest range. Correlation coefficients from this study
are similar to findings of other profile evaluation
studies, whose average correlations ranged from 0.46 to
0.78.18,21 Maple et al,18 however, reported that layper-
sons had the highest correlation coefficients because
professionals “overevaluate” profiles rather than giving
an initial reaction to the overall profile.
Genioplasty procedures are often done with mandib-
ular advancement if further chin projection is needed and
patients have a flat or insufficient mentolabial contour.
Since our sample included patients treated without genio-
plasty, the addition of advancement genioplasty in some
patients might have resulted in more positive changes.
Treatment, however, can still be considered successful in
spite of a lack of significant anteroposterior change if there
was functional or vertical improvement.5
Although the use of profile silhouettes minimizes
confounding factors, such as sex, complexion, hair color
and style, and facial expression when evaluating treatment
effects on facial profiles, it does not prevent variables such
as lip protrusion, lip competence, and nasolabial angle
from affecting the evaluation. If orthodontic treatment
altered incisor position and affected overlying soft tissues,
esthetic scores might reflect positive or negative changes
in these soft tissues, rather than reflecting the changes (or
lack of changes) after mandibular surgery.
CONCLUSIONS
1. There is a trend for more acute PA values and severe
Class II skeletal relationships to have increasing es-
thetic profile improvements after mandibular ad-
vancement, but this was statistically significant only
between 40% of the orthodontists and ANB angles.
2. Patients with a PA ⬍160° or an ANB of ⬎6° are
more likely to have profile improvements after
treatment with orthodontics and mandibular ad-
vancement surgery.
3. Although most patients have profile improvements
after mandibular advancement surgery, there is a
2.6 to 5 times increase in the incidence of profile
worsening when the initial PA values are above the
threshold and a 4.5 to 7.9 times increase when the
initial ANB is below the threshold.
REFERENCES
1. Weaver NE, Major PW, Glover KE, Varnhaen CK, Grace M.
Orthodontists’ perception of need for jaw surgery. Int J Adult
Orthod Orthognath Surg 1996;11:49-56.
American Journal of Orthodontics and Dentofacial Orthopedics
January 2009
2. Arnett GW, McLaughlin RP. Overview—treatment goals restated.
In: Arnett GW, McLaughlin RP, editors. Facial and dental planning
for orthodontists and oral surgeons. St Louis: Mosby; 2004. p 1-11.
3. Dunlevy HA, White RP, Turvey TA. Professional and lay
judgment of facial esthetic changes following orthognathic
surgery. Int J Adult Orthod Orthognath Surg 1987;2:151-8.
4. Proffit WR, Phillips C, Douvartzidis N. A comparison of
outcomes of orthodontic and surgical-orthodontic treatment of
Class II malocclusion in adults. Am J Orthod Dentofacial Orthop
1992;101:556-65.
5. Shelly AD, Southard TE, Southard KA, Casko JS, Jakobson JR,
Fridrich KL, et al. Evaluation of profile esthetic change with
mandibular advancement surgery. Am J Orthod Dentofacial
Orthop 2000;117:630-7.
6. Park YC, Burstone CJ. Soft-tissue profile—fallacies of hard-
tissue standards in treatment planning. Am J Orthod Dentofacial
Orthop 1986;90:52-62.
7. Arnett GW, Bergman RT. Facial keys to orthodontic diagnosis
and treatment planning. Part II. Am J Orthod Dentofacial Orthop
1993;103:395-411.
8. Björk A. The relationship of the jaws to the cranium. In:
Lundstrom A, editor. Introduction to orthodontics. New York:
McGraw Hill; 1960. p. 104-40.
9. Dongieux J, Sassouni V. The contribution of mandibular positioned
variation to facial esthetics. Angle Orthod 1980;50:334-9.
10. Bailey LJ, Haltiwanger LH, Blakey GH, Proffit WR. Who seeks
surgical-orthodontic treatment: a current review. Int J Adult
Orthod Orthognath Surg 2001;16:280-92.
11. Sergl HG, Zentner A, Krause G. An experimental study of the
esthetic effect of facial profiles. J Orofac Orthop 1998;59:116-26.
12. McCollum T. TOMAC: an orthognathic treatment planning
system. Part 1 soft tissue analysis. J Clin Orthod 2001;35:356-64.
13. Sarver DM, Proffit WR, Ackerman JL. Evaluation of facial soft
tissues. In: Proffit WR, White RP, Sarver DM, editors. Contem-
porary treatment of dentofacial deformity. 1st ed. St Louis:
Mosby; 2003. p. 92-126.
14. De Smit A, Dermaut L. Soft-tissue profile preferences. Am J
Orthod 1984;86:67-73.
15. Phillips C, Tulloch C, Dann C. Rating facial attractiveness.
Community Dent Oral Epidemiol 1992;20:214-20.
16. Michiels G, Sather AH. Determinants of facial attractiveness in a
sample of white women. Int J Orthod Orthognath Surg 1994;9:95-
103.
17. O’Neil KO, Harkness M, Knight R. Ratings of profile attractive-
ness after functional appliance treatment. Am J Orthod Dento-
facial Orthop 2000;118:371-6.
18. Maple JR, Vig KWL, Beck M, Larsen PE, Shanker S. A
comparison of providers’ and consumers’ perceptions of facial-
profile attractiveness. Am J Orthod Dentofacial Orthop 2005;
128:690-6.
19. Aitkin RCB. Measurement of feelings using visual analogue
scales. Proc R Soc Med 1969;62:989-93.
20. Cassidy DW, Herbosa EG, Rotskoff KS, Johnston LE. A compar-
ison of surgery and orthodontics in “borderline” adults with Class II,
Division 1 malocclusions. Am J Orthod Dentofacial Orthop 1993;
104:455-70.
21. Johnston C, Hunt O, Burden D, Stevenson M, Hepper P. The
influence of mandibular prominence on facial attractiveness. Eur
J Orthod 2005;27:129-33.