ORIGINAL ARTICLE

Smile esthetics: Perception and comparison of

treated and untreated smiles
Erdal Işıksal,a Serpil Hazar,b and Sercan Akyalçınc
Izmir, Turkey

Introduction: Although orthodontic treatment is based primarily on occlusal relationships, greater attention
is now paid to enhancing dentofacial characteristics to produce optimal facial esthetics. The purposes of this
study were to compare smile esthetics among extraction and nonextraction patients and a control group,
assess certain dentofacial characteristics in those groups, and discuss how these features relate to smile
esthetics. Methods: Panels of orthodontists, plastic surgeons, artists, general dentists, dental professionals,
and parents used a 5-point scale to rate smiling photographs of 25 extraction, 25 nonextraction, and 25
untreated control subjects. Dentofacial characteristics of the 3 groups were obtained from lateral cephalo-
metric analyses, direct biometric measurements, and frontal photographs. Smile esthetics and differences
among the 3 groups were subjected to 1-way analysis of variance (ANOVA), and Pearson correlation
coefficients were calculated to determine the relationship of the variables to the esthetic score. Results: The
mean esthetic scores for the extraction, nonextraction, and control groups were 3.15, 3.12, and 3.26,
respectively. Visible dentition width relative to the smile width ratio and intercanine distance relative to smile
width ratio were significantly different among the groups, with extraction patients showing a slightly wider
dental arch relative to the soft tissue (P ⬍ .05). There was also a significant difference in the U1-SN angle
among the groups (P ⬍ .05), and this variable showed a strong correlation with the esthetic score as did
maxillary gingival display (P ⬍ .05). However, our study groups could not be differentiated in smile esthetics.
(Am J Orthod Dentofacial Orthop 2006;129:8-16)

a 3-dimensional canvas. Sarver and Ackerman10,11

C
osmetic dentistry has long been interested in
the esthetics of the smile. Recently, the topic
has become important for orthodontists be-
cause more orthodontic patients evaluate the outcome
of treatment by their smiles and the overall enhance-
ment in their facial appearance. Although orthodontic
treatment is based primarily on occlusal relationships,
greater attention is now paid to enhancing dentofacial
characteristics to produce optimal facial esthetics.
Current knowledge suggests that favorable treat-
ment changes are significant to patients, parents, and
friends, and are important aspects of orthodontic ther-
apy. Thus, it seems worthwhile to outline the common
denominators of an esthetically pleasing smile, which is
often a primary reason for seeking orthodontic care.
The literature contains noteworthy studies1-9 describing
the esthetic elements of the dentition and the surround-
ing soft tissues during smiling that can be evaluated on
From the Department of Orthodontics, Faculty of Dentistry, Ege University,
Izmir, Turkey.
a
Professor and chair.
b
Professor.
c
Postgraduate student.
Reprint requests to: Dr Serpil Hazar, Ege Universitesi Dishekimligi Fakultesi, Ortodonti
Anabilim Dali 35100 Bornova, Izmir, Turkey; e-mail, serpilhazar@yahoo.com.
Submitted, August 2004; revised and accepted, September 2004.
0889-5406/$32.00
Copyright © 2006 by the American Association of Orthodontists.
doi:10.1016/j.ajodo.2005.07.004
published instructive data about dynamic smile visual-
ization, quantification, and relevant treatment strate-
gies. They suggested that the orthodontist should add
another dimension—time—in evaluating smiles (how
smiles change over time because of aging).
In smile esthetics, 2 transverse characteristics—
arch form and buccal corridor— have gained greater
interest recently. It is said that extraction treatment
results in constriction of the dental arches and has
deleterious effects on the smile.12 he suspected arch-
width reduction by premolar extractions decreases the
buccal corridor ratio and leads to black triangles at the
corners of the mouth during smiling. Johnson and
Smith13 concluded that variables related to the buccal
corridor or other measures of the relationship between
the widths of the dentition and of the mouth during a
smile showed no relationship to extraction esthetics.
Furthermore, 2 other studies14,15 support these views by
comparing arch widths after extraction and nonextrac-
tion treatment in study casts. However, Spahl16 argued
that plaster models cannot tell us about human faces or
the lip support they do or do not provide. Similarly,
after evaluating the lip-teeth characteristics of the posed
smile in treated and untreated subjects, Ackerman et al9
concluded that not all orthodontically well-treated pa-

8
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 129, Number 1
Fig 1. Sample frontal and three-quarter facial photographs.
tients with exemplary plaster casts have desirable
anterior tooth displays while smiling.
Various studies13,17-19 have evaluated the treated
smiles by judgments of specialists and laypeople, but
none have documented the differences among extrac-
tion, nonextraction, and control subjects. A control
group can also provide data in addition to the compar-
ison of dentofacial characteristics between different
treatment protocols. Thus, the main purpose of this
study was to compare the esthetics of the smile in
extraction and nonextraction patients with a control
group with ideal occlusion. We also aimed to statisti-
cally determine certain dentofacial characteristics of the
3 groups and to discuss how these features relate to
smile esthetics.
MATERIAL AND METHODS
Frontal and three-quarter view smiling photo-
graphs, direct biometric measurements, and cephalo-
metric data were collected from 25 extraction patients,
25 nonextraction retention patients, and 25 untreated
participants with well-balanced faces and good occlu-
sions. The mean ages were 19.08 ⫾ 2.40 years in the
extraction, 19.04 ⫾ 1.97 years in the nonextraction, and
20.24 ⫾ 2.39 years in the control groups. The sex
distributions for the 3 groups were the same (13
women, 12 men).
The control subjects were dental students at the
dental school of Ege University, Izmir, Turkey, who
had not had orthodontic therapy. The treatment groups
Işıksal, Hazar, and Akyalçın 9
had Angle Class I malocclusions before treatment.
None of the treated subjects had a severe craniofacial
anomaly, and all were treated with standard 0.018 x
0.022-in edgewise appliances. The treatment objectives
for both the extraction and the nonextraction groups
were to ideally align the incisors, establish excellent
occlusions with teeth interdigitated, and resolve tooth
size-arch length discrepancies while maintaining the
original arch form. In the extraction group, 4 first
premolars were extracted. Mean values of crowding
were 4.15 ⫾ 1.76 mm in the maxillary arch and 3.16 ⫾
1.41 mm in the mandibular arch for the nonextraction
group. In the extraction group, crowding was 7.45 ⫾
2.12 in the maxillary arch and 5.02 ⫾ 2.14 mm in the
mandibular arch. The average treatment times were
27.10 ⫾ 14.23 months for the nonextraction group and
29.46 ⫾ 12.18 months for the extraction group.
Before the study, 198 retention patients and dental
students were carefully examined, and those with den-
tal features that would have deleterious effects on the
smile, such as diastema, deep bite, open bite, overjet,
and rotations, were excluded from the study. The final
sample of 75 subjects had excellent occlusions with
Angle Class I molar and canine relationships and
well-balanced faces.
Facial photographs were taken of each participant
during smiling, including frontal and three-quarter
views (Fig 1). All photographs were taken by the same
investigator (S.A.) at a constant object-to-lens distance
with a Coolpix digital camera (Nikon Photo Products,
10 Işıksal, Hazar, and Akyalçın
Fig 2. Images cropped so that only lower face is visible.
Tokyo, Japan). Several photos were made of each
subject so that we could choose unforced, natural
smiles. A frontal view and a three-quarter view of each
smile were selected for evaluation by the panel mem-
bers. These photographs were then transferred to Photo
Express 3.0 SE image processing software (Ulead
Systems, Torrance, Calif). The 2 facial photographs
were cropped so that only the lower face was shown
(Fig 2). The cropped images were converted to black
and white and copied to slides in PowerPoint (Mi-
crosoft, Redmond, Wash) for projection.
The panels consisted of 10 orthodontists, 10 plastic
surgeons, 10 dental specialists, 10 general dentists, 10
artists, and 10 parents. There were 5 men and 5 women
on each panel to eliminate sex bias. The ages of the
panel members varied from 36 to 56 years, with mean
ages of 44.3 years for the orthodontists, 41.8 for the
plastic surgeons, 46.2 for the dental specialists, 45.3 for
the general dentists, 44.4 for the artists, and 45.7 for the
parents. Three of the artists, 1 plastic surgeon, and 1
orthodontist had received orthodontic treatment.
The raters were told that they would see 75 slides,
each showing 2 views of the same person during
smiling. They were asked to rate the attractiveness of
the smiles on a 5-point scale, with 5 as “excellent” and
1 as “poor.” The raters were shown the 75 slides in
random order for 10 seconds. Each rater made his or
her evaluation privately with no information about the
subjects. The raters were allowed to view the slides
again and revise their scores, if they desired.
To further compare the dentofacial characteristics
of the 3 groups, additional data were obtained from
lateral cephalometrics (Fig 3), direct biometric mea-
surements, and frontal photographs (Table I). All ra-
diographs were taken on the same cephalostat with the
American Journal of Orthodontics and Dentofacial Orthopedics
January 2006
subjects standing, Frankfort plane parallel to the hori-
zontal, teeth in centric occlusion, and lips relaxed.
Direct biometric measurements were made with a
digital caliper with the subjects sitting and the Frankfort
plane parallel to the horizontal. Frontal photographs
were measured by computer with Image Tool for
Windows version 3.00 (UTHSCSA, San Antonio, Tex).
Spatial measurements were calibrated by drawing a line
of known length (width of the central incisors). Five
ratios (Table I), unaffected by magnification differ-
ences, were determined from these measurements. All
measurements were repeated 1 month later by the same
investigator (S.A.) to ensure accuracy with paired t
tests. No differences were found in the remeasurements
(P ⬎ .05).
Statistical analysis
Reliability of the study was evaluated with the
Cohen kappa. Kappa statistics are appropriate for test-
ing whether agreement exceeds chance levels for binary
and nominal ratings.20 One judge was randomly se-
lected from each panel to reevaluate the entire sample
2 months later. The second trials of these 6 judges were
determined to be in the range of good repeatability (P
⬍ .05). One-way analysis of variance (ANOVA) was
used to compare smile esthetics and differences among
the 3 groups. Additionally, multiple comparisons were
performed with the Bonferroni test. Repeated-measure
ANOVA was also applied to determine differences
among the panels. To further determine whether any
variables related to smile esthetics, Pearson correlation
coefficients were calculated. The level of significance
was established as P ⬍ .05 for all statistical tests.
RESULTS
According to the 1-way ANOVA, the 3 groups did
not differ statistically in mean esthetic score, as evalu-
ated by the 6 panels (P ⬎ .05) (Table II). The subjects
with the highest and the lowest scores were in the
nonextraction group. However, according to descrip-
tive statistics, there were no significant differences in
the distributions of the highest, lowest, and moderate
esthetic scores among the 3 groups (Table III). This
finding clearly shows that not only were the mean
scores similar, but also the distributions of subjects
with low and high scores were alike in the extraction,
nonextraction, and control groups (Fig 4).
Repeated-measure ANOVA showed significant dif-
ferences in the mean ratings for the whole sample
among the 6 panels (Table II). However, according to
pairwise comparisons, no differences were determined
between orthodontists and artists, and between plastics
surgeons and general dentists (P ⬎ .05). Other possible
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 129, Number 1
Fig 3. Cephalometric measurements included in study:
1, ANB angle; 2, SN-MP angle; 3, SN-PP angle; 4,
U1-SN angle; 5, U1-NA angle; 6, U1-NA distance; 7,
IMPA angle; 8, anterior maxillary height; 9, rima oris to
occlusal plane distance; 10, lower lip to H-line distance.
pairwise comparisons between panel groups were de-
termined significant at the P ⬍ .05 level. Parents, on
average, rated the smiling photographs significantly
more attractive than the other 5 panels. Repeated-
measure ANOVA also showed that there was no
interaction between panels and groups; this meant that
the panels did not rate any group better or worse
(P ⬎ .05). Additionally, Kendall coefficients of con-
cordance were calculated for the 6 panels. The Kendall
W can be interpreted as a coefficient of agreement
among raters.21 Interrater agreement existed in each
panel at significant levels (P ⬍.05). Although the mean
ratings for each panel differed, there was interrater
agreement among the 60 raters (P ⬍ .05).
Table IV lists descriptive statistics for comparison
of variables among the extraction, nonextraction, and
control groups, and Pearson correlation coefficients
between the variable and the esthetic score. Additional
multiple comparisons, determined by Bonferroni test,
are given in Table V. According to the statistical data,
cephalometric measurements including ANB, SN-MP,
SN-PP, and lower lip to H-line did not differ among the
3 groups (P ⬎ .05), indicating that both jaw orientation
and facial harmony were similar.
Visible dentition width relative to smile width ratio
Işıksal, Hazar, and Akyalçın 11
and intercanine distance relative to smile width ratio
were significantly different among the groups, with the
extraction patients having slightly wider dental arches
relative to the soft tissues (P ⬍ .05). There was also a
significant difference in U1-SN angle among the groups
(P ⬍ .05). Nonextraction patients had slightly increased
maxillary incisor inclinations in relation to anterior
cranial base compared with extraction patients; the
control subjects had ideal mean values. Although U1 to
NA distance, U1-NA angle, and IMPA showed similar
differences among the groups, these differences were
negligible according to the statistical tests. According
to Pearson correlation coefficients, no transverse
characteristic of the smile was related to the esthetic
score. However, the correlations between esthetic
scores and measurements including U1-SN angle and
maxillary gingival display were significant at P ⬍
.05 levels, indicating the importance of certain sag-
ittal and vertical characteristics in an esthetically
pleasing smile.
DISCUSSION
Harmony and balance are not fixed concepts. Stan-
dards of beauty vary tremendously among persons and
racial groups, and according to socioeconomic mores.22
However, the results showed that our study groups
were not differentiated in smile esthetics. The numbers
of subjects with high, moderate, and low scores had
similar distributions in each group. This is a logical
expectation because, in any group of subjects, there is
individual variability—shape of the teeth, curl of the
lips, and mouth expression—that would lead the smile
to be perceived as esthetically pleasing or not.
Hulsey17 stated that the mean rated smile scores of
orthodontically treated subjects were significantly
poorer than the mean rated smile scores of the subjects
with normal occlusions. In contrast, according to Mack-
ley,18 to conclude that people with ideal occlusions and
accompanying ideal facial proportions who have had no
orthodontic treatment have more attractive smiles is an
unjust criticism of orthodontic treatment. In our study,
neither the treatment groups nor the control group was
seen as having superior smiles. These data corroborate
the findings of Johnston and Smith13 and Gianelly,15
who found no difference in smile esthetic scores
between extraction and nonextraction patients. Because
the art of esthetics lies in the clinicians’ hands and new
technologies have been developed to better visualize
and treat patients, it is not surprising to find no
difference in smile esthetics among untreated people
with ideal occlusions and patients treated either with or
without extractions.
Many studies3,6,23-25 have evaluated the perceptions
12 Işıksal, Hazar, and Akyalçın
Table I. Measurements used to evaluate smile esthetics
Measurements on frontal photographs
Smile width (mm)
Smile height (mm)
3-3 distance (mm)
4-4 distance (mm)
Visible dentition width (mm)
Maxillary gingival display (mm)
Smile index (ratio)
3-3 distance/smile width (ratio)
4-4 distance/smile width (ratio)
Visible dentition width/smile width (ratio)
3-3 distance/visible dentition width (ratio)
Direct biometric measurements
Width of central incisor (mm)
Height of central incisor (mm)
Height of central incisor during smiling (mm)
Upper lip length at rest (mm)
Upper lip length during smiling (mm)
Sn to incision distance (mm)
Maxillary incisor display (ratio %)
Upper lip length/Sn to incision (ratio)
Upper lip length during smiling/Sn to incision (ratio)
Upper lip length during smiling/upper lip length (ratio)
Measurements made on lateral cephalometrics
Linear measurements
Anterior maxillary height (mm)
Rima oris to cclusal plane (mm)
Lower lip to H-line (mm)
U1 to NA line (mm)
Angular measurements
SNA (°)
SNB (°)
ANB (°)
SN-MP (°)
SN-PP (°)
U1-SN (°)
U1-NA (°)
IMPA (°)
of different panels for dentofacial esthetic discrepan-
cies. Orthodontists on the average were found to be
more critical of dental esthetics than laypeople in
detecting minor discrepancies. In our study, even
though the mean rating scores were different among the
panels, according to the Kendall coefficients of concor-
dance, interrater agreement was determined for the 60
American Journal of Orthodontics and Dentofacial Orthopedics
January 2006
Intercommissure width as measured by distance between left cheilion to right cheilion
during smiling
Interlabial gap as measured by distance from upper stomion to lower stomion during
smiling
Distance measured between most distal points of maxillary canines
Distance measured between most distal points of maxillary first premolars (second
premolars in extraction group)
Distance between most lateral left and right points of visible maxillary dentition
during smiling
Amount of maxillary gingival exposure between inferior border of upper lip and
marginal gingiva of maxillary central incisors
Smile width divided by smile height
Intercanine distance divided by intercommissure width
Interpremolar distance divided by intercommisure width
Visible dentition width divided by smile width
Intercanine distance divided by visible dentition width
Distance measured between most distal and mesial points of maxillary central incisor
crowns
Distance measured between marginal gingiva and incisal edges of maxillary central
incisor crowns
Distance measured between most superior and inferior points on maxillary central
incisor crowns during smiling
With mandible and lips in rest position, distance from subnasale to inferior border of
upper lip
Distance from subnasale to inferior border of upper lip during smiling
Distance from subnasale to incisal edge of maxillary central incisor
Height of central incisor during smiling divided by actual height of central incisor
Lip curtain over incisors at rest: upper lip length distance divided by subnasale to
incision distance
Lip curtain over incisors during smiling: upper lip length during smiling divided by
subnasale to incision distance
Upper lip contraction during smiling: the ratio of upper lip smiling length to actual
upper lip length.
Perpendicular distance from ANS projected at right angle to occlusal plane
Relationship of occlusal plane to lip embrasure, measured as distance from rima oris
to occlusal plane
Distance of lower lip to harmony line, tangential to Pog’ and Ls
Linear distance between NA line and maxillary incisor.
Angle formed by intersection of SN and NA lines
Angle formed by intersection of SN and NB lines
Difference between SNA and SNB
Angle formed by intersection of SN line and mandibular plane (Go-Me)
Angle formed by intersection of SN line and palatal plane (ANS-PNS)
Angle formed by intersection of maxillary incisor to anterior cranial base
Angle formed by intersection of maxillary incisor to NA
Angle formed by intersection of mandibular incisor to mandibular plane
raters (P ⬍ .05). This finding enhanced the view that
our 3 groups were rated similarly in appearance. Boley
et al26 stated that orthodontic students and general
dentists could not identify the treatment modality when
assessing facial photographs.
Because some orthodontists believe that extraction
causes arch-width reduction that could lead to a de-
American Journal of Orthodontics and Dentofacial Orthopedics Işıksal, Hazar, and Akyalçın 13
Volume 129, Number 1

Table II. Comparison of mean esthetic scores among extraction, nonextraction, and control groups as evaluated by 6
panels
Extraction Nonextraction Control Whole sample
Panel n ⫽ 25 n ⫽ 25 n ⫽ 25 P n ⫽ 75

Orthodontists 3.10 ⫾ 0.55 3.20 ⫾ 0.68 3.16 ⫾ 0.61 .85 3.160

Dental specialists 3.28 ⫾ 0.52 3.32 ⫾ 0.69 3.46 ⫾ 0.50 .53 3.356
General dentists 2.90 ⫾ 0.53 2.82 ⫾ 0.57 2.96 ⫾ 0.63 .68 2.899
Parents 3.65 ⫾ 0.63 3.64 ⫾ 0.68 3.90 ⫾ 0.42 .22 3.735
Plastic surgeons 2.71 ⫾ 0.45 2.69 ⫾ 0.68 2.87 ⫾ 0.54 .49 2.761
Artists 3.28 ⫾ 0.47 2.98 ⫾ 0.68 3.21 ⫾ 0.57 .17 3.160
Mean ratings of 6 panels 3.15 ⫾ 0.46 3.12 ⫾ 0.60 3.26 ⫾ 0.44 .55 3.178

Table III. Distribution of highest, lowest, and moderate scores among 3 groups
20% highest scores 60% moderate scores 20% lowest scores
Categories n ⫽ 15 n ⫽ 45 n ⫽ 15

Groups Extraction n 5 14 6
Within group 20% 56% 24%
Within category 33.3% 31.1% 40%
Nonextraction n 5 14 6
Within group 20% 56% 24%
Within category 33.3% 31.1% 40%
Control n 5 17 3
Within group 20% 68% 12%
Within category 33.3% 37.8% 20%

Based on mean ratings of 6 panels, all 75 subjects were categorized to be in 20% highest, 20% lowest, and 60% moderate scores. Distributions
of highest, lowest, and moderate esthetic scores were similar among 3 groups.

crease in the buccal corridor ratio, another criticism of
extraction treatment is that it results in poor smile
esthetics when compared with nonextraction therapy.
However, people with normal occlusions and balanced
faces could also have narrow arch forms related to their
wide lip extensions. In a maxilla of normal width,
lingual crown inclinations of posterior teeth might also
produce a narrow smile, whereas more upright crown
inclinations produce a broader smile. With regard to
jaw size, Zachrisson27-29 suggested that the smaller the
maxillary apical base, the more labial crown tip should
be given to the canines and the premolars for obtaining
optimal esthetics. We intentionally included untreated
subjects with ideal occlusions with the extraction and
nonextraction patients to document the relationships of
transverse variables with treatment modality and es-
thetic scores. Surprisingly, the extraction patients ex-
hibited slightly wider dental arches relative to the soft
tissue, but transverse characteristics of the smile ap-
peared to be of little significance to an attractive smile.
The conclusions of Hulsey,17 Rigsbee et al,30 and
Gianelly15 agreed with this finding.
According to Utley, cited by Mackley,18 facial
esthetics is not static, and its quality is not limited to the
measurements of a headfilm or to frontal and lateral
photographs. An orthodontist must have an awareness
of the proper vertical relationship of the denture and
soft tissues. With this in mind, we have included some
direct biometric measurements during a smile and at
rest in the vertical plane with other vertical measure-
ments made on lateral cephalometrics and frontal pho-
tographs. Biometric measurements, used in this study,
demonstrate the amount of lip contraction and the
amount of vertical lip drape over dentition at rest and
during smiling. However, no vertical measurement was
statistically significant, and only the maxillary gingival
display measurement appeared to be significant to an
attractive smile. A statistically significant correlation
(Pearson correlation coefficient, –.410) was determined
between the maxillary gingival display and the esthetic
score: the greater the maxillary gingival display, the
lower the esthetic score. Some authors17,18 concluded
that the upper lip should be at the height of the gingival
margin of the maxillary central incisors in an attractive
smile. Chiche and Pinault31 stated that the esthetically
ideal amount of visible gingiva was about 1 mm,
although 2 to 3 mm of gingiva might be esthetically
acceptable. However, this is very much of a function of
age, because children show more teeth at rest and have
more gingival display on smile than do adults.11
14 Işıksal, Hazar, and Akyalçın American Journal of Orthodontics and Dentofacial Orthopedics
January 2006

Fig 4. Distribution of esthetic scores of 75 subjects.

Table IV.Descriptive statistics for comparison of variables among extraction, nonextraction, and control groups and
Pearson correlation coefficients between variable and esthtetic score
Extraction Nonextraction Control
Variable Pearson correlation n ⫽ 25 n ⫽ 25 n ⫽ 25 P

Smile index (smile width/smile height) (ratio) .115 5.43 ⫾ 0.93 5.31 ⫾ 0.92 5.60 ⫾ 1.08 .602
3-3 distance/smile width (ratio) ⫺.157 0.72 ⫾ 0.030 0.70 ⫾ 0.047 0.68 ⫾ 0.041 .009†
4-4 distance/smile width (ratio) ⫺.097 0.83 ⫾ 0.034 0.81 ⫾ 0.050 0.80 ⫾ 0.040 .569
Visible dentition width/smile width (ratio) .034 0.9 ⫾ 0.040 0.88 ⫾ 0.054 0.90 ⫾ 0.034 .026†
3-3 distance/visible dentition width (ratio) ⫺.196 0.78 ⫾ 0.043 0.79 ⫾ 0.045 0.76 ⫾ 0.041 .058
Maxillary gingival display (mm) ⫺.410* 0.06 ⫾ 0.22 0.13 ⫾ 0.49 0.17 ⫾ 0.50 .645
Maxillary incisor display (ratio %) .100 97.58 ⫾ 3.69 96.40 ⫾ 7.20 95.07 ⫾ 6.62 .342
Upper lip length/Sn to incision (ratio) ⫺.061 0.82 ⫾ 0.034 0.81 ⫾ 0.053 0.82 ⫾ 0.054 .727
Upper lip length (smiling)/Sn to incision (ratio) .043 0.64 ⫾ 0.037 0.65 ⫾ 0.052 0.64 ⫾ 0.049 .363
Upper lip length (smiling)/upper lip length (ratio) .111 0.78 ⫾ 0.045 0.80 ⫾ 0.060 0.78 ⫾ 0.054 .199
ANB (°) ⫺.004 3.22 ⫾ 1.40 3.24 ⫾ 1.60 2.68 ⫾ 1.50 .333
SN-MP (°) ⫺.103 33.82 ⫾ 2.70 34.34 ⫾ 3.04 33.76 ⫾ 3.16 .322
SN-PP (°) .141 9.60 ⫾ 3.36 8.98 ⫾ 3.24 10.88 ⫾ 2.88 .103
U1-SN (°) ⫺.252* 100.2 ⫾ 5.25 104.7 ⫾ 6.12 103.3 ⫾ 4.83 .013†
U1-NA (°) ⫺.072 21.62 ⫾ 5.23 24.52 ⫾ 5.50 22.40 ⫾ 4.78 .130
U1-NA (mm) ⫺.113 3.88 ⫾ 2.10 5.26 ⫾ 2.95 4.45 ⫾ 1.77 .117
IMPA (°) .018 94.24 ⫾ 5.59 97.70 ⫾ 7.51 96.32 ⫾ 4.93 .139
Anterior maxillary height (mm) .002 31.32 ⫾ 2.90 30.72 ⫾ 2.94 30.10 ⫾ 2.77 .329
Rima oris to occlusal plane (mm) ⫺.198 2.91 ⫾ 1.61 2.19 ⫾ 1.62 2.13 ⫾ 1.22 .127
LL to H-line (mm) ⫺.023 0.49 ⫾ 1.64 1.18 ⫾ 1.39 0.73 ⫾ 0.97 .198

*P ⬍ .05.
†
P ⬍ .05.

Of all the factors related to a balanced smile, 1
can easily be controlled—maxillary incisor position.
These teeth should be angulated and also positioned
most favorably in their anteroposterior and vertical
relationships to all facial structures to ensure maxi-
mum facial harmony.32 In this study, the inclinations
of the maxillary central incisors as measured by
U1-SN angle were statistically different among the 3
groups. It seems that the maxillary incisors needed
more labial crown torque after retraction in the
extraction group. However, the difference did not
affect smile esthetics in the 3 groups. Furthermore,
our results showed that increasing the U1-SN angle
would cause smile esthetics to deteriorate (Pearson
correlation coefficient, –252). For instance, the use
of a high-torque bracket system particularly in non-
extraction treatment with anterior crowding and ini-
tial tooth torques that are close to the desired finished
American Journal of Orthodontics and Dentofacial Orthopedics Işıksal, Hazar, and Akyalçın 15
Volume 129, Number 1

Table V. Multiple comparisons

Variable Extraction-nonextraction Extraction-control Nonextraction-control

Smile index (smile width/smile height) (ratio) NS NS NS

3-3 distance/smile width (ratio) NS * NS
4-4 distance/smile width (ratio) * NS NS
Visible dentition width/smile width (ratio) NS NS NS
3-3 distance/visible dentition width (ratio) NS NS NS
Maxillary gingival display (mm) NS NS NS
Maxillary incisor display (ratio %) NS NS NS
Upper lip length/Sn to incision (ratio) NS NS NS
Upper lip length (smiling)/Sn to incision (ratio) NS NS NS
Upper lip length (smiling)/upper lip length (ratio) NS NS NS
ANB (°) NS NS NS
SN-MP (°) NS NS NS
SN-PP (°) NS NS NS
U1-SN (°) * NS NS
U1-NA (°) NS NS NS
U1-NA (mm) NS NS NS
IMPA (°) NS NS NS
Anterior maxillary height (mm) NS NS NS
Incision to stomion (mm) NS NS NS
LL to H-line (mm) NS NS NS

NS, Not significant.

*P ⬍ .05.

angles would be inappropriate.33 Overexpansion of (orthodontists, plastic surgeons, artists, general

the maxillary dental arch with increased maxillary
incisor torque might flatten the smile arc, which is
best described by the relationship of the curvature of
the incisal edges of the maxillary incisors and the
canines to the curvature of the lower lip,5,9 and lead
to negative esthetic consequences. Proclined maxil-
lary incisors might also reduce incisor display. If a
patient shows less than 75% of the central incisor
crowns at smile, tooth display is considered inade-
quate.7 However, inadequate torque of both the
maxillary and mandibular incisors can also result in
undesirable effects regardless of the treatment mo-
dality.
As a result of our study, no statistically significant
differences were determined in smile esthetics among
subjects with ideal occlusions and Class I patients
treated with or without extractions. However, this does
not mean that extraction of premolars is justified. Jaw
size, labial crown angulations, initial incisor torques,
initial crowding, and soft tissue extensions should be
carefully evaluated before treatment planning. Each
subject has different problems, and, to create natural
dental esthetics, the entire dentition must be carefully
considered.
CONCLUSIONS
1. Subjects with ideal occlusions and Class I patients
treated with or without extractions were not differ-
entiated in smile esthetics by 6 panels of judges
dentists, dental professionals, and parents).
2. Transverse characteristics of the smile appeared to
be of little significance to an attractive smile.
3. Maxillary gingival display and the ultimate posi-
tions of the anterior teeth have definite effects on
smile esthetics.
4. Treatment modality alone has no predictable effect
on the overall esthetic assessment of a smile.
REFERENCES
1. Zachrisson BU. Esthetic factors involved in anterior tooth
display and the smile: vertical dimension. J Clin Orthod 1998;
32:432-5.
2. Phillips E. The classification of smile patterns. J Can Dent Assoc
1999;65:252-4.
3. Johnston CD, Burden DJ, Stevenson MR. The influence of dental
to facial midline discrepancies on dental attractiveness ratings.
Eur J Orthod 1999;21:517-22.
4. Benson KJ, Laskin DM. Upper lip asymmetry in adults during
smiling. J Oral Maxillofac Surg 2001;59:396-8.
5. Sarver DM. The importance of incisor positioning in the esthetic
smile: the smile arc. Am J Orthod Dentofacial Orthop 2001;120:
98-111.
6. Thomas JL, Hayes C, Zawaideh S. The effect of axial midline
angulation on dental esthetics. Angle Orthod 2003;73:359-64.
7. Morley J, Eubank J. Macroesthetic elements of smile design.
J Am Dent Assoc 2001;132:39-45.
8. Ward DH. Proportional smile design using the recurring esthetic
dental (RED) proportion. Dent Clin N Am 2001;45:143-54.
9. Ackerman JL, Ackerman MB, Bresinger CM, Landis JR. A
morphometric analysis of the posed smile. Clin Orthod Res
1998;1:2-11.
16 Işıksal, Hazar, and Akyalçın American Journal of Orthodontics and Dentofacial Orthopedics
January 2006

10. Sarver DM, Ackerman MB. Dynamic smile visualization and
quantification: part 1. Evolution of the concept and dynamic records
for smile capture. Am J Orthod Dentofacial Orthop 2003;124:4-12.
11. Sarver DM, Ackerman MB. Dynamic smile visualization and
quantification: part 2. Smile analysis and treatment strategies.
Am J Orthod Dentofacial Orthop 2003;124:116-27.
12. Spahl TJ, Witzig JW. The clinical management of basic maxillofacial
orthopedic appliances. Littleton (Mass): PSG Publishing Co; 1987.
13. Johnson DK, Smith RJ. Smile esthetics after orthodontic treat-
ment with and without extraction of four first premolars. Am J
Orthod Dentofacial Orthop 1995;108:162-7.
14. Gianelly AA. Arch width after extraction and nonextraction
treatment. Am J Orthod Dentofacial Orthop 2003;123:25-8.
15. Kim E, Gianelly AA. Extraction vs nonextraction: arch widths
and smile esthetics. Angle Orthod 2003;73:354-8.
16. Spahl TJ. Premolar extractions and smile esthetics. Am J Orthod
Dentofacial Orthop 2003;124(1):16A-7A.
17. Hulsey CM. An esthetic evaluation of lip-teeth relationships
present in the smile. Am J Orthod 1970;57:132-44.
18. Mackley RJ. An evaluation of smiles before and after orthodontic
treatment. Angle Orthod 1993;63:183-90.
19. Kerns LL, Silveira AM, Kerns DG, Regennitter FJ. Esthetic
preference of the frontal and profile views of the same smile. J
Esthet Dent 1997;9:76-85.
20. Cook RJ. Kappa. In: Armitage TP, Colton T, editors. The
encyclopedia of biostatistics. New York: Wiley; 1998. p. 2160-6.
21. Kendall MG, Smith BB. The problems of m ranksings. Annals of
mathematical statistics 1939;10:275-87.
22. Czarnecki ST, Nanda RS, Currier GF. Perceptions of a balanced
facial profile. Am J Orthod Dentofacial Orthop 1993;104:180-7.
23. Kerr WJS, O’Donnell JM. Panel perception of facial attractive-
ness. Br J Orthod 1990;17:299-304.
24. Phillips C, Tulloch C, Dann C. Rating of facial attractiveness.
Community Dent Oral Epidemiol 1992;20:214-20.
25. Kokich VO, Kiyak HA, Shapiro PA. Comparing the perception
of dentists and lay people to altered dental esthetics. J Esthet
Dent 1999;11:311-24.
26. Boley JC, Pontier JP, Smith S, Fulbright M. Facial changes in
extraction and nonextraction patients. Angle Orthod 1998;68:
539-46.
27. Zachrisson BU. Making the premolar extraction smile full and
radiant. World J Orthod 2002;3:260-5.
28. Zachrisson BU. Maxillary expansion: long-term stability and
smile esthetics. World J Orthod 2001;2:266-72.
29. Zachrisson BU. Premolar extraction and smile esthetics. Am J
Orthod Dentofacial Orthop 2003;124(6):11A-2A.
30. Rigsbee OH III, Sperry TP, BeGole EA. The influence of facial
animation on smile characteristics. Int J Adult Orthod Orthog
Surg 1988;3:233-9.
31. Chiche GJ, Pinault A. Esthetics of anterior fixed prosthodontics.
Chicago: Quintessence; 1994.
32. Janzen EK. A balanced smile: a most important treatment
objective. Am J Orthod 1977;72:359-72.
33. Cordato M. Variation in torque expression. Am J Orthod Dento-
facial Orthop 2004;126(1):18A.

BOUND VOLUMES AVAILABLE TO SUBSCRIBERS

Bound volumes of the American Journal of Orthodontics and Dentofacial Orthopedics are
available to subscribers (only) for the 2005 issues from the Publisher, at a cost of $96.00 ($115.56
Canada and $108.00 international) for Vol. 127 (January-June) and Vol. 128 (July-December).
Shipping charges are included. Each bound volume contains subject and author indexes. The
binding is durable buckram, with the journal name, volume number, and year stamped in gold on
the spine. Payment must accompany all orders. Contact Elsevier Inc., Subscription Customer
Service, 6277 Sea Harbor Dr, Orlando, FL 32887; phone 800-654-2452 or 407-345-4000.
Subscriptions must be in force to qualify. Bound volumes are not available in place of a
regular Journal subscription.