ORIGINAL ARTICLE

Arch width after extraction and nonextraction

treatment
Anthony A. Gianelly, DMD, PhD, MD
Boston, Mass

An undocumented criticism of extraction treatment is that it results in narrower dental arches when
compared with nonextraction therapy. Anterior and posterior arch widths of the maxillary and mandibular
arches of 25 patients treated by 4 first-premolar extraction and 25 patients treated without extractions were
measured on posttreatment study models and compared statistically to determine whether the dental arches
were narrower after extraction treatment, to test the accuracy of this view. Measurements were made in the
canine and the molar regions from the most labial aspect of the buccal surfaces of the canines and the
molars. In both groups, anterior and posterior arch widths were the same except for the mandibular
intercanine dimension, which was 0.94 mm larger (P ⬍ .01) in the extraction group. This indicates that
extraction treatment does not result in narrower dental arches than nonextraction treatment. (Am J Orthod
Dentofacial Orthop 2003;123:25-8)

T
o paraphrase the Bard of Avon: to extract, or
not to extract: that is the question. This funda-
mental decision has spawned some intense de-
bates in orthodontics.1 One issue is the effect of
extraction on the buccal corridor, a recently introduced
landmark that represents the space between the buccal
surface of the dentition and the corresponding soft
tissues with particular emphasis on the corners of the
mouth.2-4 Presumably, the appearance of unesthetic
black triangles at the corners of the mouth during
smiling and dark shadows lateral to the buccal seg-
ments are expected sequellae of 4 first-premolar extrac-
tion treatment because this therapy narrows the width
of the dental arches and shrinks the arches, resulting in
a dentition that is too small to fill the oral slit during a
smile.5
Although the assumed relationship has been ac-
cepted as conventional wisdom by some,6 its validity is
questionable at best. Specifically, Johnson and Smith,2
in 1 of the few studies designed to evaluate this
relationship in the refereed literature, used a “buccal
corridor ratio,” defined as the maxillary intercanine
width divided by the width of the mouth during a smile,
to assess smile esthetics and found that the ratio was the
same after both extraction and nonextraction treatment.
Also, the width of the dental arches, at least in the
Professor and chairman, Department of Orthodontics, Boston University
School of Dental Medicine.
Reprint requests to: Anthony A. Gianelly, 100 E Newton St, Boston, MA
02118-2392.
Submitted, March 2002; revised and accepted, May 2002.
Copyright © 2003 by the American Association of Orthodontists.
0889-5406/2003/$30.00 ⫹ 0
doi:10.1067/mod.2003.57
canine region, is generally not smaller after extraction
treatment than after nonextraction therapy.7,8 For ex-
ample, the intercanine widths of both arches were
statistically the same after extraction and nonextraction
treatment of comparable groups of borderline subjects.7
And in patients whose long-term results were stable,
the mandibular intercanine dimension increased more
with extraction treatment9 than with nonextraction
therapy.10-13
Because a point of contention by those who espouse
and support the buccal-corridor relationship is a nar-
rowing of the dental arches, the purpose of this study
was to compare anterior and posterior widths of the
dental arches after both extraction and nonextraction
therapy to determine whether extraction treatment re-
sults in narrower dental arches and, by inference, larger
dark shadows in the buccal corridor.
MATERIAL AND METHODS
The posttreatment study models of 25 patients who
had 4 first-premolar extractions and 25 patients treated
without extractions were randomly selected and mea-
sured. The only inclusion criterion was that all second
molars were present and in alignment. Treatment was
performed at a university clinic and directed by various
instructors. All patients received fixed appliances; 49
were treated with the edgewise technique, and 1 with
the tip-edge technique. In the extraction group, there
were 12 Class I and 13 Class II malocclusions. In the
nonextraction sample, there were 9 Class I, 13 Class II,
and 3 Class III malocclusions.
With an electronic caliper (Digit-Cal, Browne and
Sharpe, North Kingston, RI), the widths of the anterior
25
26 Gianelly American Journal of Orthodontics and Dentofacial Orthopedics
January 2003

and posterior parts of the maxillary and mandibular Table I. Pretreatment mandibular intercanine and
dental arches were measured in the canine and the intermolar arch widths: means and SD (mm)
molar regions from the most labial aspect of the buccal
Extraction Nonextraction
surfaces of those teeth. The caliper was placed at the (n ⫽ 25) (n ⫽ 25) Significance
best estimate of a right angle to the palatal suture in the
maxillary arch and to a line bisecting the incisor 3-3 31.47 ⫾ 2.10 31.24 ⫾ 1.32 NS
segment in the mandibular arch. The recorded widths 7-7 58.70 ⫾ 3.02 58.51 ⫾ 1.67 NS
between the molars were the widest distances between 3, Canine; 7, second molar; NS, not significant.
the first or second molars. The widest part of the
posterior part of the arch invariably was in the second Table II.Posttreatment maxillary and mandibular arch
molar region. Each distance was measured 3 times, and intercanine and intermolar widths: means and SD
the average of the 3 values was used as the final (mm)
measure.
Because arch width is ordinarily established by the Extraction Nonextraction
mandibular arch, pretreatment widths between the man- (n ⫽ 25) (n ⫽ 25) Difference Significance
dibular canines and molars were recorded, as described, 3-3(L) 32.87 ⫾ 1.74 31.93 ⫾ 1.20 0.94 .01
to establish that the arch widths of the mandibles of 3-3(U) 39.93 ⫾ 2.02 39.08 ⫾ 1.64 0.83 NS
both samples were similar at the start of treatment. The 7-7(L) 57.54 ⫾ 2.80 58.38 ⫾ 2.37 0.74 NS
treatment changes in the mandibular intercanine and 7-7(U) 60.92 ⫾ 2.00 60.88 ⫾ 2.91 0.04 NS
intermolar dimensions were also determined. The pre- 3, Canine; 7, second molar; NS, not significant; L, maxillary; U,
treatment maxillary intercanine and intermolar dimen- mandibular.
sions were not measured because they vary in different
malocclusions, and the distribution of malocclusions
was not the same in each group. width was increased a statistically significant 1.39 mm
The reproducibility of the measurements was eval- in the patients treated with extractions, but the mean
uated by analyzing the differences between 10 double 0.69-mm increase in the nonextraction group was not
measurements of intercanine and intermolar distances, statistically significant (Table III). The mandibular
randomly selected and taken at different times. The intermolar widths of both the extraction and the non-
error of measurement was assessed by Dahlberg’s extraction patients were essentially unchanged.
formula14:
DISCUSSION

Sx ⫽ 冑 兺D2
2N
where D is the difference between duplicate measure-
ments, and N is the number of double determinations.
The errors were 0.22 mm for mandibular intercanine
width, 0.50 mm for mandibular intermolar molar width,
0.26 mm for maxillary intercanine width, and 0.27 mm
for maxillary intermolar width. Means and standard
deviations were calculated, and a 2-tailed t test was
used to determine statistically significant differences
with P ⬍ .05.
RESULTS
At the start of treatment, the mandibular intercanine
and intermolar widths of both groups did not differ
statistically (Table I). At the end of treatment, the arch
widths of both groups were also statististically similar
with 1 exception (Table II). The average mandibular
intercanine dimension was 0.94 mm larger in the
extraction sample than in the nonextraction subjects.
During treatment, the mandibular intercanine mean
To some advocates of nonextraction therapy, 4
first-premolar extraction treatment is synonymous with
narrow dental arches with consequent unesthetic, large
black triangles in the buccal corridor.5 This is not an
accurate premise when describing treatment results
obtained by fixed appliances. A comparison of post-
treatment arch widths of patients treated by extraction
and nonextraction procedures failed to identify the
extraction patients because they had no characteristic
narrow dental arches. In fact, the arches were approx-
imately 1 mm wider in the mandibular canine area after
extraction therapy, and this region establishes the width
of the anterior part of the arches under normal condi-
tions because the maxillary arch is conventionally fit to
an idealized mandibular arch.
Although this finding might dissatisfy some nonex-
traction enthusiasts, particularly those who view extrac-
tion treatment negatively,5,6 it is not a surprising
observation if one examines the results of studies on
posttreatment long-term stability in which mandibular
incisor stability was acceptable. Specifically, the man-
dibular intercanine width increased 1.07 mm in an
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 123, Number 1
Table III. Mandibular intercanine and intermolar width changes: means and SD (mm)
Extraction (n ⫽ 25)
Pretreatment Posttreatment Difference Significance
3-3 31.48 ⫾ 2.10 32.87 ⫾ 1.74 1.39 .02
7-7 58.71 ⫾ 3.02 57.17 ⫾ 2.19 1.54 NS
3, Canine; 7, second molar; NS, not significant.
extraction sample.9 In contrast, in the nonextraction
subjects, the increase in mandibular intercanine dimen-
sion was less than 1 mm in both Class I13 and Class II
patients.10-12
In borderline patients, the long-term increase of the
mandibular intercanine width in those treated by ex-
traction therapy was 1.0 mm, whereas the increase in
the nonextraction sample was only 0.5 mm.7 At the end
of treatment, the intercanine widths of both groups
were the same. Luppanapornlarp and Johnston8 evalu-
ated the posttreatment and long-term results of treat-
ment in clear-cut extraction and nonextraction patients
and noted that the mandibular intercanine dimension of
the extraction subjects was greater at all stages exam-
ined than the same parameter in the nonextraction
patients. These data indicate that there is no systematic
narrowing of the dental arches as a result of 4 first-
premolar extraction treatment.
The increase in the mandibular intercanine width in
this study in both extraction (1.4 mm) and nonextrac-
tion (0.69 mm) subjects compares with the previously
cited increases recorded by others who evaluated pa-
tients treated by both strategies.7,8,15,16 In these reports,
a slightly larger increase occurred in those treated with
extractions, possibly reflecting minor lateral movement
as the canines are moved distally into the premolar
sites. As an example, BeGole et al15 observed that the
mandibular intercanine widths increased 1.58 mm in an
extraction sample and 0.95 mm in nonextraction sub-
jects. Uhde et al16 illustrated treatment changes in
graph form in extraction and nonextraction subjects and
demonstrated that the pretreatment mandibular interca-
nine widths were comparable in both groups, and,
during treatment, a larger increase occurred in the
extraction sample.
An unanticipated and surprising observation was
that the intermolar widths of both groups of subjects
were the same after treatment, because the usual ex-
pectation is that the posterior part of the arch becomes
narrower as the molars move mesially into a narrower
part of the arch during space closure. When including
second molars in the assessment, this narrowing appar-
ently does not occur. This observation also supports the
view of Johnson and Smith2 that “typically, transverse
Gianelly 27
Nonextraction (n ⫽ 25)
Pretreatment Posttreatment Difference Significance
31.24 ⫾ 1.74 31.93 ⫾ 1.20 0.69 NS (.06)
58.51 ⫾ 1.72 58.38 ⫾ 2.37 ⫺0.13 NS
arch width at any particular location in the buccal
segments is maintained or slightly enlarged after ex-
traction.” They also opined that the logic of those who
believe that premolar extraction results in a reduction of
the radius of the curve of the dental arch is faulty
because the dental arch is not a circle and does not
behave as a circle. Therefore, a reduction of the radius
is not a necessary outcome of extracting a premolar on
each side of the arch.
To some, maxillary arch width is a determinant of
smile esthetics.4-6 Because maxillary arch width in the
extraction and nonextraction subjects was the same, the
effect of the 2 treatment strategies on smile esthetics
would be similar. This conclusion is consistent with the
results of a study that evaluated smile esthetics after
both extraction and nonextraction treatment.2 There
were no differences in esthetic scores between the
groups. Interestingly, the investigators found that inter-
canine width, relative to the visible dentition, was wider
in the extraction patients.
The labial surfaces of the canines and the molars
were chosen as the measuring landmarks instead of the
more customary cusp tips for several reasons: (1) to
determine the widest possible widths of the arches, (2)
to prevent confusion when selected cusps tips were not
distinct, and (3) to avoid identifying a specific molar for
measurement because in extraction treatment, as noted
by Johnson and Smith,2 the first molars might have
moved forward into a narrower part of the arch. For this
reason, arch width measured between first molars
would not adequately represent the width of the poste-
rior aspect of the dental arches.
Changes in maxillary canine and molar widths were
not recorded because the maxillary dentition does not
customarily establish the widths of the dental arches.
Also, these measurements can produce confusing infor-
mation unless the groups compared are essentially
identical at the outset. For instance, a pretreatment
comparison of Class II Division 1 and Division 2
malocclusions with well-aligned mandibular arches and
identical mandibular intercanine widths might show
that the maxillary intercanine dimension is narrower in
the Class II Division 1 malocclusion, particularly if the
arch form in the overjet region is tapered. Under these
28 Gianelly
conditions, the dimensional change resulting from
treatment would be greater for the Class II Division 1
malocclusion even though the posttreatment interca-
nine distance in both examples would be the same
because the mandibular intercanine distance is the
same.
No patient in this study was treated with appliances
specifically designed to increase mandibular arch
width. The instability of excessive intercanine expan-
sion (⬎ 1-1.5 mm) in the mandibular arch17 is 1 reason
that our customary treatment does not include these
appliances.
In a more global sense, the contention that extrac-
tion treatment results in less esthetic smiles because of
narrowing of the dental arches is, to an extent, similar
to a once-presumed damaging effect of extraction
therapy on the temporomandibular joint.18 In both
instances, the consequences of extraction treatment are
considered to be decidedly negative. The imputed
injurious relationship between extraction treatment and
temporomandibular disease has largely been discredit-
ed.19 The available data indicate that narrow dental
arches are not the expected consequence of extraction
treatment. Therefore, the esthetically compromising
effect of narrow dental arches on smiles is not a
systematic outcome of extraction treatment.
REFERENCES
1. Case CS. The question of extraction in orthodontia (reprint).
Am J Orthod 1964;50:660-91.
2. 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.
3. Dougherty HL. Clubs, quips, phrases, and hype: musings for the
new millennium. Am J Orthod Dentofacial Orthop 2000;117:
586-8.
4. McNamara JA Jr. Maxillary transverse deficiency. Am J Orthod
Dentofacial Orthop 2000;117:567-70.
5. Spahl TJ, Witzig JW. The clinical management of basic maxil-
lofacial orthopedic appliances. Littleton (Mass): PSG Publishing
Co; 1987.
American Journal of Orthodontics and Dentofacial Orthopedics
January 2003
6. Dierkes JM. The beauty of the face: an orthodontic perspective.
J Am Dent Assoc 1987;(special number):89-95E.
7. Paquette DE, Beattie JR, Johnston LE. A long-term comparison
of nonextraction and premolar extraction edgewise therapy in
“borderline” Class II patients. Am J Orthod Dentofacial Orthop
1992;102:1-14.
8. Luppanapornlarp S, Johnston LE. The effects of premolar ex-
traction: a long-term comparison of outcomes in “clear-cut”
extraction and nonextraction Class II patients. Angle Orthod
1993;63:257-72.
9. Vaden JL, Harris EF, Ziegler Garner R. Relapse revisited. Am J
Orthod Dentofacial Orthop 1997;111:543-53.
10. Glenn G, Sinclair P, Alexander R. Nonextraction orthodontic
therapy: posttreatment dental and skeletal stability. Am J Orthod
Dentofacial Orthop 1987;92:321-8.
11. Moussa R, O’Reilly MT, Close JM. Long-term stability of rapid
palatal expander and edgewise mechanotherapy. Am J Orthod
Dentofacial Orthop 1995;108:478-88.
12. Yavari J, Shrout MK, Russell CM, Haas AJ, Hamilton EH.
Relapse in Angle Class II Division 1 malocclusion treated by
tandem mechanics without extraction of permanent teeth: a
retrospective analysis. Am J Orthod Dentofacial Orthop 2000;
118:34-42.
13. Azizi M, Shrout MK, Haas AJ, Russell CM, Hamilton EH. A
retrospective study of Angle Class I malocclusions treated
orthodontically without extractions using two palatal expansion
methods. Am J Orthod Dentofacial Orthop 1999;116:101-7.
14. Dahlberg G. Statistical methods for medical and biological
students. London: George Allen and Unwin Ltd; 1940. p. 122-32.
15. BeGole EA, Fox DL, Sadowsky C. Analysis of change in arch
form with premolar expansion. Am J Orthod Dentofacial Orthop
1999;113:307-15.
16. Uhde MD, Sadowsky C, BeGole EA. Long-term stability of
dental relationships after orthodontic treatment. Angle Orthod
1983;53:240-52.
17. Burke SP, Silveira AM, Goldsmith LJ, Yancey JM, Stewart A,
Scarfe WC. A meta-analysis of mandibular intercanine width in
treatment and post retention. Angle Orthod 1998;68:53-60.
18. Farrar WB, McCarty WL. A clinical outline of temporomandib-
ular joint diagnosis and treatment. Montgomery (Ala): Walker
Printing Co; 1983. p. 84-5.
19. McNamara JA Jr, Seligman DA, Okeson JP. Orthodontics,
occlusion and temporomandibular disorders. In: McNamara JA
Jr, Bruden WL. Orthodontics and dentofacial orthopedics. Ann
Arbor (Mich): Needham Press; 2001. p. 519-44.