ORIGINAL ARTICLE

Transverse skeletal and dentoalveolar changes

during growth
Richard M. Hesby,a Steven D. Marshall,b Deborah V. Dawson,c Karin A. Southard,d John S. Casko,e
Robert G. Franciscus,f and Thomas E. Southardg
Iowa City, Iowa

Introduction: We previously reported that, on average, maxillary molars erupt with buccal crown torque and
upright with age, mandibular molars erupt with lingual crown torque and upright with age, and these molar
crown torque changes are accompanied by concurrent increases in maxillary and mandibular intermolar
widths. Our purpose in this study was to determine the transverse skeletal changes that accompany these
molar movements during growth. Methods: Thirteen maxillary and mandibular transverse measurements
were made by using casts and posteroanterior radiographs from 36 Class I untreated subjects from the Iowa
Facial Growth Study at approximately ages 7.6, 10.3, 12.9, 16.5, and 26.4 years. Intermolar arch widths,
maxillary basal bone widths (bijugale), and mandibular basal bone widths (bigonion, biantegonion) were
recorded for each subject. Also recorded were the maxillary and mandibular cross-arch alveolar process
widths measured at the molar gingival crest and the midalveolar level. Results: The results indicate a pattern
of width changes in the maxilla, the maxillary alveolar process, the maxillary first molars, the mandibular first
molars, and the mandibular alveolar process that occurs as a gradient in the vertical dimension. The greatest
width change occurs superiorly (jugale point). The smallest width change occurs inferiorly (midalveolar point
of the mandible). On average, the basal bone of the maxilla increased in width by 5.37 mm; maxillary
cross-arch midalveolar process width increased by 3.20 mm (left buccal surface to right buccal surface) and
4.94 mm (left palatal surface to right palatal surface); maxillary intermolar width increased by 3.08 mm;
mandibular intermolar width increased by 2.05 mm; and mandibular cross-arch crest level alveolar width
increased by 1.60 mm (left buccal surface to right buccal surface) and 1.02 mm (left lingual surface to right
lingual surface). The basal bone of the mandible increased in width by 14.54 mm. Conclusions: We
concluded that transverse molar movements during growth mirror transverse maxillary basal bone width
increases, maxillary cross-arch alveolar process width increases, and mandibular cross-arch alveolar
process width increases. Furthermore, there is a pattern of width changes in the maxilla, the maxillary
alveolar process, the maxillary first molars, the mandibular first molars, and the mandibular alveolar process
that occurs as a gradient in the vertical dimension. On the other hand, transverse mandibular basal bone
changes, measured as bigonion and biantegonion, do not fit this pattern. (Am J Orthod Dentofacial Orthop
2006;130:721-31)

G
rowth-related changes in size, shape, and po-
sition of the facial skeleton and the associated
growth of the dentition have been well docu-
mented by longitudinal lateral cephalometry.1-4 This
From the University of Iowa, Iowa City.
a
Former resident, Department of Orthodontics; private practice, Bedford, Mass.
b
has been crucial to our understanding of normal and
abnormal facial growth in the vertical and sagittal
planes.5 However, facial growth changes in the coronal
plane (ie, changes in the transverse dimension) are less
well understood.
Transverse growth-related bony changes in maxil-

c
Adjunct associate professor, Department of Orthodontics. lary basal width and maxillary alveolar process cross-
Professor, Department of Preventive and Community Dentistry, Director of
Biostatistics.
arch width are thought to result from growth at the
d
Professor, Department of Orthodontics. median palatal suture and appositional remodeling
e
Professor, B. F. and Helen E. Dewel Endowed Chair, Department of along the lateral aspects of the posterior region of the
Orthodontics.
f
Associate professor, Department of Anthropology.
maxilla and the maxillary tuberosity.6-9 Median palatal
g
Professor and head, Department of Orthodontics. sutural growth is greatest in the posterior portion of the
Supported by the Dr George Andreasen Memorial Fund. suture and occurs well into the second decade of
Reprint requests to: Thomas E. Southard, Department of Orthodontics, College of
Dentistry, University of Iowa, Iowa City, IA 52242; e-mail, tom-southard@ life,7-10 with sutural fusion occurring in the mid-
uiowa.edu. 20s.10,11 Increases in maxillary skeletal width have
Submitted, August 2004; revised and accepted, March 2005. been measured longitudinally into the mid and late teen
0889-5406/$32.00
Copyright © 2006 by the American Association of Orthodontists. years.8,9,12-17 In the mandible, growth of the corpus
doi:10.1016/j.ajodo.2005.03.026 width is accomplished by apposition and resorption of
721
722 Hesby et al
bone.5,18 Width increases of the corpus measured at
gonion (bigonial width) have been shown throughout
adolescence.15,16,19 Width increases measured along
the lateral surface of the mandibular body anterior to
attachments of the masseter muscle have also been
reported through adolescence9 and into adulthood.20
Increases in mandibular width at the level of the
alveolar process have not been reported.
With respect to the dentition, many longitudinal
studies of dental arch width showed that intermolar
width increases occur during the first 2 decades of
life.8,9,12,15,21,22-30 During the transitional dentition,
increases in width occur to a greater degree in the
maxilla than in the mandible.8,9,12,15,21,23-29 Fol-
lowed through adolescence into adulthood, intermo-
lar width changes vary depending on the longitudinal
sample, with some investigators reporting in-
creases,8,12,15,21-23,30 and others finding no significant
changes for males and females24,26,29,31,32 and de-
creases for females.28,33,34
In our recent longitudinal study of untreated sub-
jects, we identified transverse movements that first
permanent molars undergo from ages 7 to 26 years.21
As viewed in the coronal plane, maxillary molars were
found to erupt with buccal crown inclination and
upright lingually as maxillary intermolar width in-
creased. Concomitantly, mandibular molars were found
to erupt with lingual crown inclination and upright
buccally as mandibular intermolar width increased.
These findings raise a number of questions. Why
did the maxillary molars erupt buccally but then later
upright when the maxillary intermolar width was ex-
panding? In other words, as the maxillary molars were
uprighting, the intermolar width should have dimin-
ished, not increased. Furthermore, why were the max-
illary molars uprighting at all? Do the molar move-
ments we found reflect transverse skeletal changes
during growth? Previous studies were inadequate in
providing answers to these questions. The purpose of
this study was to determine the relationship between
transverse skeletal changes and transverse molar move-
ments during growth.
MATERIAL AND METHODS
The material for this research project was obtained
from the Iowa Facial Growth Study. Beginning in 1946,
this sample consisted of 89 boys and 86 girls of
primarily northern European descent and above-aver-
age socioeconomic status living near Iowa City, Iowa.
Orthodontic records were taken semiannually from
ages 3 to 12 years, annually during adolescence, and
once during early adulthood. All subjects were free of
facial or skeletal disharmony and had normal (Angle
American Journal of Orthodontics and Dentofacial Orthopedics
December 2006
Class I) molar and canine occlusions. Subjects were
chosen from this population for our study based on the
following criteria: (1) longitudinal study casts and
posteroanterior radiographs available from ages 7 to 30
years, (2) untreated Class I occlusion, (3) lack of
distortion or abrasion of the tooth portion of the dental
casts, (4) lack of excessive enamel wear or restorations
of the dental cusps of the measured teeth, (5) lack of
missing or extracted permanent teeth, and (6) lack of
gingival recession with time in the maxillary and
mandibular first molar regions. These criteria limited
our sample to 36 subjects (18 female, 18 male). The
most common reasons for exclusion of a subject from
the sample were cast distortion and restorations of teeth
at later ages.
Transverse measurements
The same subject population measured in this study
was also measured in our previous study on transverse
molar movements.21 Measurement time points for both
studies were identical. The subjects were grouped to
approximate 5 stages of dentition development: Early,
middle, and late mixed dentition; adolescent dentition;
and adult dentition. The time points of measurement
were the mean ages for the subject groups at these
stages: 7.6, 10.3, 12.9, 16.5, and 26.4 years. The age
ranges and standard deviations for the groups are
reported in our previous study.21 Transverse maxillary
and mandibular dentoalveolar and skeletal widths were
measured for each subject by using posteroanterior
radiographs and dental casts as shown in Figure 1.
1. Maxillary basilar width: width of the maxilla
measured radiographically from bilateral points on
the jugale processes at the intersection of the
outline of the tuberosity of the maxilla and the
zygomatic buttresses.
2. Maxillary cross-arch width midalveolar buccal:
width of the maxilla measured on casts from
bilateral points located 9.0 mm directly apical
from the most occlusal point of the buccal devel-
opmental groove viewed from the buccal aspect of
the maxillary permanent first molar.
3. Maxillary cross-arch width midalveolar palatal:
width of the palate measured on casts from bilat-
eral points located 9.0 mm directly apical from the
most occlusal point of the lingual developmental
groove viewed from the lingual aspect of the
maxillary permanent first molar.
4. Maxillary cross-arch width alveolar crest buccal:
width of the maxilla measured on casts from
bilateral points located at the gingival margin
directly apical from the most occlusal point of the
American Journal of Orthodontics and Dentofacial Orthopedics Hesby et al 723
Volume 130, Number 6

Fig 1. Transverse linear measurements recorded: 1, maxillary basilar width (bijugale); 2, maxillary
cross-arch width midalveolar buccal; 3, maxillary cross-arch width midalveolar palatal; 4, maxillary
cross-arch width alveolar crest buccal; 5, maxillary cross-arch width alveolar crest palatal; 6,
maxillary intermolar width; 7, mandibular intermolar width; 8, mandibular cross-arch width alveolar
crest buccal; 9, mandibular cross-arch width alveolar crest lingual; 10, mandibular cross-arch
width midalveolar buccal; 11, mandibular cross-arch width midalveolar lingual; 12, mandibular
basilar width (bigonion); 13, mandibular basilar width (biantegonion). Measurements 1, 12, and 13
were measured radiographically; measurements 2-11 were measured on study casts.

buccal developmental groove viewed from the
buccal aspect of the maxillary permanent first
molar.
5. Maxillary cross-arch width alveolar crest palatal:
width of the palate measured on casts from bilat-
eral points located at the gingival margin directly
apical from the most occlusal point of the lingual
developmental groove viewed from the lingual
aspect of the maxillary permanent first molar.
6. Maxillary intermolar width: width of the maxillary
dental arch measured on casts from bilateral points
located at the most gingival point on the contour
of the distal marginal ridge of the maxillary
permanent first molar.
7. Mandibular intermolar width: width of the man-
dibular dental arch measured on casts from bilat-
eral points located at the most gingival point on
the contour of the distal marginal ridge of the
mandibular permanent first molar.
8. Mandibular cross-arch width alveolar crest buccal:
width of the mandible measured on casts from
bilateral points located at the gingival margin
directly apical from the most occlusal point of the
buccal developmental groove viewed from the
buccal aspect of the mandibular permanent first
molar.
9. Mandibular cross-arch width alveolar crest lingual:
width of the mandible measured on casts from
bilateral points located at the gingival margin di-
rectly apical from the most occlusal point of the
lingual developmental groove viewed from the lin-
gual aspect of the mandibular permanent first molar.
10. Mandibular cross-arch width midalveolar buccal:
width of the mandible measured on casts from
bilateral points located 9.0 mm directly apical
from the most occlusal point of the buccal devel-
opmental groove viewed from the buccal aspect of
the mandibular permanent first molar.
11. Mandibular cross-arch width midalveolar lingual:
width of the mandible measured on casts from
bilateral points located 9.0 mm directly apical
from the most occlusal point of the lingual devel-
opmental groove viewed from the lingual aspect
of the mandibular permanent first molar.
12. Mandibular basilar width bigonial: width of the
mandible measured radiographically from bilat-
724 Hesby et al
eral points at the most lateral margin of the angle
of the mandible.
13. Mandibular basilar width biantegonial: width of
the mandible measured radiographically from bi-
lateral points at the most superior point of the
antegonial notch.
For measurements on casts, midalveolar and gingi-
val points were located and marked with a 0.5-mm lead
pencil. A digital caliper was used to measure the linear
distance between bilateral points. The rationale for
choosing the midalveolar point 9.0 mm apical to the
most occlusal point of the buccal and lingual grooves
was to ensure that the measurement point would be well
within the alveolar process portion of the cast recorded
by the impression and in the approximate root-trunk
area of the molar roots.
We assumed that interarch cast measurements made
from the buccal, palatal, and lingual gingivae repre-
sented the underlying alveolar process bone measure-
ments. That is, we assumed that the gingival soft tissue
covering this bone maintained a relatively constant
thickness throughout the age range studied. This was a
reasonable assumption. Measurements of palatal gingi-
val thickness near the first molar were reported to
change by only 0.1 mm over the ages of 14 to 59
years.35 Furthermore, no significant differences in buc-
cal gingival thickness of the first molar region were
noted by ultrasonic measurement of the maxilla and the
mandible in subjects of widely differing ages.36 We
therefore assumed that our choice of measurement
points on casts represented the underlying alveolar
process bone.
Passive eruption of the permanent molars with age
was noted for these subjects, and the gingival margin
measurement point changed compared with the tooth
crown at the later ages of measurement. It is not
possible to have a fixed reference point on a study cast
that is transferable to serial casts for the same subject.
Thus, our gingival measurement point changed with
time compared with the tooth crown in an indetermin-
able way. We think, however, that the gingival margin
point is a reference to the alveolar crest that is constant
with age, and we assumed that the width changes we
recorded were also accompanied by changes in other
dimensions that we could not record.
Corrections for radiographic enlargement of bilat-
eral linear measurements were made for each subject at
each time point as previously reported for the Iowa
Growth Study.19 The method was well documented for
the Iowa Growth Study. Lateral and posteroanterior
cephalograms were traced to include the landmarks
(jugale, gonion, and antegonion) with cephalometric
American Journal of Orthodontics and Dentofacial Orthopedics
December 2006
tracing acetate and a 0.5-mm lead pencil. Distances
from the posteroanterior cephalogram landmarks to the
midcoronal plane were measured on the lateral cepha-
logram to correct for enlargement of measurements on
the posteroanterior cephalogram.
Reliability of cast and radiographic measurements
To assess the reliability of cast and radiographic
measurements, 30 units (a unit is the cast or radiograph
from 1 visit for 1 subject) were randomly chosen, and
all cast or radiographic measures were duplicated by
the investigator (R.M.H.) who made all the original
measurements. Selection of these units to assess reli-
ability was carried out separately for cast and radio-
graphic measures; this allowed for a greater distribution
over the sample of subjects and the entire period of
observation. Intraclass correlation coefficients were
obtained for each width measure as measures of reli-
ability. The intraclass correlation is typically used in
situations such as this, when it is of interest to obtain a
measure of intrarater agreement for quantitative out-
comes.37,38 Perfect agreement corresponds to an intra-
class correlation coefficient of 1; intraclass correlations
of zero indicate lack of agreement between duplicate
measures. All intraclass correlation coefficients were
greater than 0.99 and differed significantly from zero
(with significance of P ⬍.00001 in each instance). This
indicates almost perfect agreement between the dupli-
cate measurements with 1 rater.
Statistical methods
To assess age trends, we used nonparametric
methods to analyze longitudinal data, using the
approach of Ghosh et al39 and Dawson and Siegler.40
These statistical methods are useful when applied to
modest subject numbers and when data are missing;
they also apply in instances such as this one, when at
least some measurements of interest do not meet the
stringent distributional requirements of traditional
parametric methods. Few distributional assumptions
are required because they emphasize nonparametric
approaches based on rank procedures. To assess age
trends, the Spearman rank correlation of a given
radiographic or cast measurement and time of mea-
surement was computed as a summary measure of
each subject’s change in that measurement over time.
These independent summary reexpressions of the
data were used to address the null hypothesis of zero
correlation (no change over time). The Wilcoxon
signed rank test was used to test the null hypothesis
that the median rank correlation is zero—ie, that
there was no increasing or decreasing trend over
time.41 This same approach was used to assess
American Journal of Orthodontics and Dentofacial Orthopedics Hesby et al 725
Volume 130, Number 6

Table I. Transverse skeletal and dentoalveolar width measurements (in mm) during growth
7.6 years 10.3 years 12.9 years 16.5 years 26.4 years

Mean SD Mean SD Mean SD Mean SD Mean SD P*

Transverse measurement
1. Maxillary basilar width bijugale 56.24 2.76 59.26 2.85 60.59 3.10 62.32 3.62 61.57 3.92 ⬍.0001
2. Maxillary cross-arch width midalveolar buccal 56.60 2.96 58.37 3.10 59.29 3.10 59.84 3.66 59.80 3.71 ⬍.0001
3. Maxillary cross-arch width midalveolar palatal 26.42 2.39 28.75 2.35 30.05 2.47 31.22 2.91 31.36 3.15 ⬍.0001
4. Maxillary cross-arch width alveolar crest buccal 54.22 2.82 55.70 2.77 56.59 2.86 57.31 3.67 57.60 3.59 ⬍.0001
5. Maxillary cross-arch width alveolar crest palatal 33.11 2.62 34.07 2.37 34.69 2.59 35.36 3.09 35.62 3.18 ⬍.0001
6. Maxillary intermolar width 46.50 2.18 48.24 2.48 48.78 2.62 49.25 3.04 49.58 2.98 ⬍.0001
7. Mandibular intermolar width 41.54 2.16 42.43 2.35 42.93 2.57 43.08 3.05 43.59 3.24 ⬍.0001
8. Mandibular cross-arch width alveolar crest buccal 51.95 2.47 52.78 2.59 53.27 2.70 53.46 3.01 53.55 2.85 ⬍.0001
9. Mandibular cross-arch width alveolar crest lingual 32.00 2.14 32.40 2.28 32.50 2.41 32.59 2.64 33.02 2.41 .0046
10. Mandibular cross-arch width midalveolar buccal 56.13 2.20 56.07 2.66 55.79 2.55 55.68 2.92 55.63 3.02 .0278
11. Mandibular cross-arch width midalveolar lingual 31.66 2.16 32.08 2.33 31.86 2.23 31.75 2.40 31.69 2.44 .9200
12. Mandibular basilar width bigonial 81.05 3.52 85.38 36.3 89.09 4.10 93.70 5.58 95.59 6.71 ⬍.0001
13. Mandibular basilar width biantegonial 72.03 3.26 75.95 3.15 78.80 3.35 81.90 3.79 83.04 4.42 ⬍.0001
Maxillary first molar torque‡ 10.30 3.64 7.62 4.49 7.10 4.18 5.94 3.90 4.83 4.35 ⬍.0001
Mandibular first molar torque‡ –10.18 5.20 –6.64 5.72 –3.30 5.17 –3.23 4.50 –2.27 5.25 ⬍.0001

*Significance probability associated with test for correlation between width change and time, using nonparametric approach of Ghosh et al39 for
longitudinal data.
‡
Mean torque of paired right and left first molars in degrees from previous study.21

Fig 2. Mean transverse maxillary basilar, cross-arch alveolar, and intermolar changes with time.
Significance probability associated with Wilcoxon signed rank test of null hypothesis that median
correlation among subjects studied is equal to zero (ie, no correlation between width change and
time) is shown for significant changes in width with time as **P ⬍.0001. These results are significant
after adjustment for multiple comparisons. Measurement numbers are as described in Figure 1.

correlations between pairs of measurements within
subjects over time. The Wilcoxon signed rank pro-
cedure for matched pairs was also used to evaluate
whether median net change over the time period
differed from zero for each of the 13 measures and to
compare net change between a given pair of mea-
surements. The latter set of analyses included 53
comparisons, including all pairwise comparisons
among the 6 maxillary measures, all pairwise com-
parisons among the 7 mandibular measures, and
additional comparisons contrasting the jugale-jugale
width, the bigonial, and the biantegonial widths with
the other measures. All available pairwise data were
used for these analyses. The modified Bonferroni
method of Holm42 was used to adjust for multiple
comparisons with an overall type I error level of .05.
726 Hesby et al American Journal of Orthodontics and Dentofacial Orthopedics
December 2006

Fig 3. Mean transverse mandibular basilar, cross-arch alveolar, and intermolar changes with time.
Significance probability associated with Wilcoxon signed rank test of null hypothesis that median
correlation among subjects studied is equal to zero (ie, no correlation between width change and
time) is shown for significant changes in width with time as **P ⬍.0001 and *P ⬍.005. Results are
significant after adjustment for multiple comparisons. Measurement numbers are as described in
Figure 1.

RESULTS comparisons), which appeared to closely follow the

Mean transverse skeletal and dentoalveolar changes
during growth are given in Table I. Figure 2 illustrates
change with time for each transverse maxillary mea-
surement referenced to the first time point (mean age,
7.6 years). Figure 3 illustrates change with time for
each transverse mandibular measurement referenced to
the first time point (mean age, 7.6 years). Table II
illustrates the relationship between width changes at the
13 measurement points. The results in Table II include
adjustments made with the modified Bonferroni
method for multiple comparisons,42 with an overall .05
level of statistical significance.
Strong evidence for age-related increases in
width was identified at all 6 maxillary measurement
points with significant median Spearman rank corre-
lations (P ⫽ .8 to .9, P ⬍.0001). All 6 measurements
had similar patterns of change during growth, appear-
ing to reach adult dimensions by age 16.5 years. As
shown in Figure 2, changes in maxillary basilar width
(bijugale) and maxillary cross-arch width midalveolar
palatal followed a similar pattern. On average, maxil-
lary basilar width appeared to decrease slightly after
age 16.5 years. Change in palatal cross-arch width at
the midalveolar point appeared to be slightly less than
maxillary basilar change but significantly greater than
the other 3 maxillary measurements (P ⬍.0001 by
Wilcoxon signed rank test for the 3 comparisons,
significant at the .05 level after adjustment for multiple
change in intermolar width. Maxillary basilar width
change is similar to palatal width change, and these
changes are significantly different from all other
maxillary and mandibular measures, after adjustment
for all pairwise comparisons of measures performed
(Table II).
As illustrated in Figure 3 and Table I, increases
with age in mandibular intermolar width and cross-
arch alveolar width at the buccal and lingual crests
were highly significant (median P ⫽ .7, P ⬍.005).
However, the cross-arch alveolar measurements
taken more apically (midalveolar buccal and lingual
measurements, 10 and 11) showed equivocal or no
significant change in width with time. Mandibular
basilar widths measured as bigonion and biantego-
nion increased dramatically with age (P ⬍.0001).
Mandibular basilar width change, measured as bigo-
nion and biantegonion, was 5 to 6 times greater than
width changes in the cross-arch alveolar buccal crest
and lingual crest regions. Mandibular intermolar
width change and mandibular cross-arch alveolar
crest width change were similar, but both were
significantly different from width change at antegonion
and gonion (P ⬍.0001, remaining significant at the .05
level after multiple comparisons adjustment). Bigonial
and biantegonial width changes differed significantly from
those of all other measurements (Table II).
American Journal of Orthodontics and Dentofacial Orthopedics Hesby et al 727
Volume 130, Number 6

Table II.Pairwise comparison of 13 maxillary and crest level (means, 4.94 and 2.51 mm, respectively),
mandibular transverse measurements with respect to suggesting that the lateral drift of the palatal surface
change over time away from the midsagittal plane is greater at the level
of the first molar roots than at the level of the first molar
Maxillary measurements Mandibular measurements
gingival crest. Along the buccal alveolar surface of the
Comparison P value* Comparison P value* maxilla, width increase in the region of the first molar
1 vs 2 .0002 7 vs 8 .3420
is similar at the midalveolar and alveolar crest (means,
1 vs 3 .4957 7 vs 9 .0026 3.38 and 3.20 mm, respectively), and both points
1 vs 4 .0004 7 vs 10 ⬍.0001 showed lateral drift greater than that of the maxillary
1 vs 5 ⬍.0001 7 vs 11 ⬍.0001 molar crowns (3.04 mm).
1 vs 6 ⬍.0001 7 vs 12 ⬍.0001 We also reported that mandibular molars upright
2 vs 3 ⬍.0001 7 vs 13 ⬍.0001
2 vs 4 .5116 8 vs 9 .0249
transversely with age by tipping buccally, but without
2 vs 5 .0346 8 vs 10 ⬍.0001 the lateral translation of the roots seen in the maxillary
2 vs 6 .5197 8 vs 11 ⬍.0001 molars.21 In this study, the alveolar crest measurement
3 vs 4 ⬍.0001 8 vs 12 ⬍.0001 points had mean increases in width of 1.60 mm (right
3 vs 5 ⬍.0001 8 vs 13 ⬍.0001 buccal surface to left buccal surface) and 1.02 mm
3 vs 6 ⬍.0001 9 vs 10 ⬍.0001
4 vs 5 .0111 9 vs 11 .0031
(right lingual surface to left lingual surface), carrying
4 vs 6 .3708 9 vs 12 ⬍.0001 the alveolar surfaces away from the midsagittal plane.
5 vs 6 .0727 9 vs 13 ⬍.0001 In contrast, the more apical measurement points at the
1 vs 7-13 ⬍.0001 10 vs 11 .0106 mandibular midalveolar process had no significant
10 vs 12 ⬍.0001 change in width. This pattern is consistent with the
10 vs 13 ⬍.0001
11 vs 12 ⬍.0001
mean change in mandibular first molar intermolar width
11 vs 13 ⬍.0001 (2.02 mm) and suggests transverse uprighting that
12 vs 13 ⬍.0001 occurs by rotation in the coronal plane around a point in
*P value ⬍.01163 indicates significant difference between measure-
the body of the tooth root. However, these mandibular
ments after adjustment for multiple comparisons, using overall .05 width increases in the dentoalveolar region, proximal to
level of significance. the first molars, were far less than the mean changes in
the basilar width measured at the angle of the mandible,
representing less than 15% of mean basilar width
DISCUSSION increase measured at antegonion (10.81 mm).
A composite illustration of the mean transverse If we reflect on the transverse changes in our study
basal skeletal, dentoalveolar, and molar changes from with respect to growth changes in bone that is proximal
ages 7.6 to 26.4 years is provided in Figure 4, which to the permanent first molars (ie, excluding changes at
summarizes our principal findings: Transverse maxil- gonion and antegonion), the pattern of width change
lary basal bone width increases, maxillary alveolar reflects the transverse molar movements occurring with
process width increases, and mandibular alveolar pro- growth for this sample. Thus, width changes in the
cess bone width increases are consistent with transverse basal maxilla (jugale), the maxillary alveolar process,
molar movements during growth. In contrast, trans- the maxillary first molars, the mandibular first molars,
verse mandibular basal bone width increases, measured and the mandibular alveolar process occur as a gradient
as either gonion-gonion or as antegonion-antegonion in the vertical dimension (jugale-mandibular alveolar
are considerably greater, and transverse mandibular process). The greatest width change occurred superi-
molar movements do not reflect this magnitude of orly (jugale point). The smallest change occurred infe-
basilar mandibular change during growth. riorly (midalveolar point of the mandible). In the
We previously reported that maxillary molars un- maxilla, transverse basilar drift and displacement ap-
dergo uprighting in the transverse dimension as their peared to highly influence transverse molar move-
crowns expand laterally with age and rotate in the ments, whereas the mandibular dentoalveolar complex
coronal plane to increase molar torque.21 The age- in the region of the first molars appeared to be restricted
related changes seen in maxillary basilar and alveolar in transverse growth compared with the mandibular
arch widths in this study are congruent with this basal bone in the region of gonion and antegonion,
maxillary transverse molar uprighting. Basilar maxil- which appeared to grow independently.
lary width underwent the greatest increase (mean, 5.37 How might this be explained? According to the
mm), followed by palatal width increase, which is twice equilibrium theory, the tongue, the muscles and the soft
as great at the midalveolar level than at the alveolar tissues of the cheeks, and occlusal function play roles in
728 Hesby et al American Journal of Orthodontics and Dentofacial Orthopedics
December 2006

Fig 4. Composite illustration of mean transverse basilar, cross-arch alveolar, and intermolar
changes from ages 7.5 to 26.4 years. 1, maxillary basilar width; 2, maxillary cross-arch width
midalveolar buccal; 3, maxillary cross-arch width midalveolar palatal; 4, maxillary cross-arch width
alveolar crest buccal; 5, maxillary cross-arch width alveolar crest palatal; 6, maxillary intermolar
width; 7, mandibular intermolar width; 8, mandibular cross-arch width alveolar crest buccal;
9, mandibular cross-arch width alveolar crest lingual; 10, mandibular cross-arch width midalveolar
buccal; 11, mandibular cross-arch width midalveolar lingual; 12, mandibular basilar width bigonion;
13, mandibular basilar width biantegonial. Measurements 1, 12, and 13 were measured radiograph-
ically; measurements 2-11 were measured on study casts. Overall pattern of changes includes
changes in molar torque reported previously.21 Horizontal bars indicate average change in width
measured in millimeters from 7.5 to 26.4 years of age.

determining the buccolingual position of teeth.43 It has
been demonstrated that, when cheek pressure and
posterior occlusal function are reduced by a Fränkel
appliance, both the molar crowns and their buccal
alveolar bones drift farther laterally during growth than
if the appliance was not used.44 However, with a
Fränkel, the lateral tipping of the molar crowns is
greater than the lateral drift of the associated buccal
alveolar bone. In other words, transverse maxillary
molar uprighting (increased lingual crown torque),
identified in our sample of untreated subjects, does not
occur in patients whose cheek and occlusal forces are
reduced. This suggests that these forces are important
influences in transverse molar uprighting. In addition,
application of a Fränkel appliance produces a change in
the transverse relationship of the mandibular molars
that mirrors the findings of our study. In subjects
treated with a Fränkel, the mandibular molars show
lateral drift of the molar crowns greater than the lateral
drift of the alveolus.44 Thus, under conditions of
reduced forces from occlusal function and cheek-
muscle function, the area of least lateral drift is the
mandibular buccal midalveolar process. This suggests
some limitation to lateral alveolar drift and lateral drift
of the dentition in the mandible.
If we assume that the mandibular dentoalveolar
complex is limited in growth-driven lateral drift, then
the transverse molar movements in our sample can be
explained by occlusal function and cheek pressure
during development. Erupting maxillary molars move
laterally as the suture widens and encounter cheek
muscles that force the crowns lingually. In addition,
occlusal function between the maxillary and mandibu-
lar molars could also act to upright the maxillary molars
in the lingual direction while uprighting the mandibular
molars in the buccal direction. Because of the con-
straints suggested above, the mandibular molars are
limited in lateral drift, and the maxillary molars re-
American Journal of Orthodontics and Dentofacial Orthopedics Hesby et al 729
Volume 130, Number 6

Table III. Comparison of maxillary and mandibular longitudinal change in transverse width
†
Authors Landmark for bilateral measurement Mean annual change*

Woods12 Most lateral aspect of zygomatic arch 1.91

Singh and Savara13 Zygomaticomaxillary suture 1.33
Savara and Singh14
Singh and Savara13 Pterygomaxillary fissure 0.88
Savara and Singh14
Present study Jugale 0.76
Cortella et al16 Jugale 0.76
Snodell et al15 Jugale 0.67
Korn and Baumrind9 Implant in zygomatic process of maxilla 0.43
Björk and Skieller8 Implant in zygomatic process of maxilla 0.42
Gandini and Buschang17 Implant in zygomatic process of maxilla 0.27
Present study Alveolar bone proximal to maxillary first molars 0.38
Various authors8,15,21,23-29 Maxillary first molars 0.30-0.48
Various authors15,21,23-29 Mandibular first molars 0.18-0.32
Present study Alveolar bone proximal to mandibular first molars 0.18
Gandini and Buschang17 Implant placed inferior to mandibular first molar 0.19
Cortella et al16 Antegonion 1.23
Present study Antegonion 1.24
Snodell et al15 Antegonion 1.41
Newman and Meredith19 Gonion 1.47
Woods12 Gonion 1.71

*Change in mm/year based on calculation directly from data provided for total mean change in bilateral width between right and left implants or
between right and left radiographic or study cast landmarks over age range of study.
†
Bilateral widths measured between right and left landmarks.

spond by uprighting as they increase their transverse
width due to continued maxillary basilar transverse
expansion. In other words, the transverse molar move-
ments compensate for the differential transverse growth
of the maxillary and mandibular basilar structures near
the first permanent molars (not antegonion or gonion) that
directly influence alveolar process cross-arch width. Sim-
ilar compensation in the anteroposterior dimension was
shown during growth of untreated Class II subjects.45,46
Our transverse skeletal data agree with previous lon-
gitudinal studies of skeletal width changes.8,9,12-17,19 Ta-
ble III compares these studies with our data by calcu-
lation of annual rates of width changes for various
landmarks, including the permanent first molars. It is
interesting to note how the growth change in these
structures compares with growth change in intermolar
width. Relative to maxillary intermolar width change,
width change from right to left jugale point is approx-
imately 2-fold greater; width change from right to left
pterygomaxillary fissure point is more than 2-fold
greater; width change from right to left zygomatico-
maxillary suture point is more that 3-fold greater; and
width change from the most lateral aspect of the right
and left zygomatic process is approximately 5-fold
greater (Table III).
Width change of the basilar portion of the jaws
measured by implant studies8,9,17 shows a smaller
annualized rate of change compared with our data and
those of other cephalometric studies.15,16,19 Implants
measure bone displacement longitudinally, whereas
cephalometric landmarks measure drift (remodeling)
and displacement longitudinally.
Our maxillary buccal midalveolar points grew in
width at the rate of 0.40 mm per year over the first 10
years of measurement (ages 7.6 to 16.5 years); this is
similar to the rate of maxillary basilar transverse
displacement seen in implant studies when implants
were placed in the zygomatic process of the maxilla
proximate to the first molars.8,9
Our basal skeletal data agree with those of others
who reported that mandibular basal bone, measured at
gonion or antegonion, increases in width approximately
twice as much as maxillary basal bone, measured at
jugale, over the first 2 decades of life, growing approx-
imately 20 mm wider in the transverse dimen-
sion.9,15,16,47 In some studies, with implants in the
buccal cortical plate of the mandibular body in the
region apical to the first molar, the authors reported
lateral displacement of the right and left corpora of the
mandible during growth.9,17 Our measurements, made
at a more occlusal level in the alveolar bone, do not
support these findings.
We are unaware of any other longitudinal growth
studies reporting alveolar process transverse changes
measured from either the lateral aspect of the alveolar
process (buccal surface to buccal surface) or the medial
730 Hesby et al
aspect of the alveolar process (palatal surface to palatal
surface or lingual surface to lingual surface) for both
the maxilla and the mandible. Knott48 reported an
increase in palatal width from ages 5 to 17 years in
untreated females. Our finding of 4.94 mm is similar,
considering that our study did not extend to an earlier
age when rapid growth in width occurs.
Authors have previously proposed comparing ef-
fective maxillary width (right to left jugale width) with
effective mandibular width (right to left antegonial
width) to indicate the presence or absence of an
underlying transverse skeletal discrepancy.47,49 How-
ever, Newman and Meredith19 reported only a weak,
clinically insignificant correlation between these
mandibular and maxillary basal bone measures. Our
findings—that transverse molar movement with
growth generally reflects the corresponding alveolar
process and maxillary basal but not mandibular basal
skeletal change—support the findings of Newman and
Meredith.19 Simply put, gonion and antegonion appear
to be too far removed from the maxillary or mandibular
alveolar process bone to be of much value in defining
transverse skeletal discrepancies. Gonion and antego-
nion positions might be more reflective of medial
pterygoid and masseter muscle dynamics than the other
landmarks studied.
How do our findings translate into clinical practice?
These results support the concept that the limiting
factor for either orthopedic or surgical expansion of the
maxilla is the mandibular posterior dentition. Although
the mandibular basilar bone widens at the angle of the
mandible during growth, this region is essentially
divorced from the maxillary basal bone, the maxillary
alveolar process bone, and the mandibular alveolar
process bone. Mandibular basal bone at the first molars
does not appear to widen substantially. An increase in
mandibular intermolar width with age results from
slight widening of the alveolar process bone and tipping
of the molars buccally. In our opinion, the greatest
amount of expansion an orthodontist should attempt to
achieve in the mandibular molar region is the point at
which the mandibular molars are upright. Short of a
mandibular midline surgical procedure to widen the
mandible, this mandibular limit establishes a maxillary
expansion limit.
The American Board of Orthodontics requires can-
didates for board certification to assess molar bucco-
lingual inclinations for finished cases. The board states
that, “In order to establish proper occlusion in maxi-
mum intercuspation and avoid balancing interferences,
there should not be a significant difference between
heights of the buccal and lingual cusps of the maxillary
and mandibular molars and premolars.”50 The molar
American Journal of Orthodontics and Dentofacial Orthopedics
December 2006
uprighting that we found naturally occurring during
normal growth supports the board’s criteria. As we
noted in our earlier study,21 our findings justify the goal
of achieving upright posterior teeth during orthodontic
treatment.
Another implication of our investigation concerns
studies of spontaneous mandibular arch expansion and
transverse expansion stability. Because the posterior
teeth are both uprighting and expanding (on average, 3
mm in the maxilla and 2 mm in the mandible) during
normal growth, it should not be surprising to find
reports of spontaneous mandibular molar uprighting
after rapid maxillary expansion or reports of relative
long-term stability of small amounts of arch expansion
in childhood or adolescence.51
CONCLUSIONS
● Transverse molar movements during growth mirror
transverse maxillary basal bone width increases,
maxillary cross-arch alveolar process width in-
creases, and mandibular cross-arch alveolar process
width increases.
● There is a pattern of width changes in the maxilla, the
maxillary alveolar process, the maxillary first molars,
the mandibular first molars, and the mandibular
alveolar process that occurs as a gradient in the
vertical dimension (jugale-mandibular alveolar pro-
cess). The greatest width change occurs more supe-
rior (jugale point). The smallest width change occurs
inferiorly (midalveolar point of the mandible).
● Transverse mandibular basal bone changes, mea-
sured as bigonion and biantegonion, do not fit this
pattern.
REFERENCES
1. Björk A. Facial growth in man studied with the aid of metallic
implants. Acta Odontol Scand 1955;13:9-34.
2. Björk A. Variations in the growth pattern of the human mandible:
longitudinal radiographic study by the implant method. J Dent
Res 1963;42:400-11.
3. Björk A. The use of metallic implants in the study of facial
growth in children: method and application. Am J Phys An-
thropol 1968;29:243-54.
4. Björk 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;5:16-46.
5. Enlow DH. Handbook of facial growth. Philadelphia: Saunders:
1975.
6. Enlow DH, Bang S. Growth and remodeling of the human
maxilla. Am J Orthod 1965;51:446-56.
7. Krebs A. Midpalatal suture expansion studied by the implant
method over a 7-year period. Trans Eur Orthod Soc 1964;40:
131-42
8. Björk A, Skieller V. Growth of the maxilla in three dimensions
as revealed radiographically by the implant method. Br J Orthod
1977;4:53-64.
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 130, Number 6
9. Korn EL, Baumrind S. Transverse development of the human
jaws between the ages of 8.5 and 15.5 years, studied longitudi-
nally with use of implants. J Dent Res 1990;69:1298-306.
10. Melsen B. Palatal growth studied on human autopsy material: a
histologic microradiographic study. Am J Orthod 1975;68:42-54.
11. Persson M, Thilander B. Palatal suture closing in man from
15-35 years of age. Am J Orthod 1977;72:42-52.
12. Woods GA Jr. Changes in width dimensions between certain
teeth and facial points during human growth. Am J Orthod
1950;36:676-700.
13. Singh IJ, Savara BS. Norms of size and annual increments of
seven anatomical measures of maxillae in girls from 3 to 16 years
of age. Angle Orthod 1966;36:312-24.
14. Savara BS, Singh IJ. Norms of size and annual increments of
seven anatomical measures of maxillae in boys from 3 to 16
years of age. Angle Orthod 1968;38:104-20.
15. Snodell SF, Nanda FS, Currier GF. A longitudinal cephalometric
study of transverse and vertical craniofacial growth. Am J Orthod
Dentofacial Orthop 1993;104:471-83.
16. Cortella S, Shofer F, Ghafari J. Transverse development of the
jaws: norms for the posteroanterior cephalometric analysis. Am J
Orthod Dentofacial Orthop 1997;112:519-22.
17. Gandini LG, Buschang PH. Maxillary and mandibular width
changes studied using metallic implants. Am J Orthod Dentofa-
cial Orthop 2000;117:75-80.
18. Enlow DH, Harris DB. A study of the postnatal growth of the
human mandible. Am J Orthod 1965;50:25-50.
19. Newman K, Meredith H. Individual growth in skeletal bigonial
diameter during the childhood period from 5 to 11 years of age.
Am J Anat 1956;99:157-88.
20. Israel H. Evidence for continued apposition of adult mandibular
bone from skeletalized materials. J Prosthet Dent 1979;41:101-4.
21. Marshall S, Dawson D, Southard KA, Lee AN, Casko JS,
Southard TE. Transverse molar movements during growth. Am J
Orthod Dentofacial Orthop 2003;124:615-24.
22. Harris EF. A longitudinal study of arch size and form in untreated
adults. Am J Orthod Dentofacial Orthop 1997;111:419-27.
23. Moorrees CFA. The dentition of the growing child: a longitudi-
nal study of dental development between 3 and 18 years of age.
Cambridge: Harvard University Press; 1959.
24. Sillman JH. Dimensional changes of the dental arches: a
longitudinal study from birth to 25 years. Am J Orthod
1964;50:824-42.
25. Knott VB. Size and form of the dental arches in children with
good occlusion: studied longitudinally from 9 years to late
adolescence. Am J Phys Anthropol 1961;19:263-84.
26. DeKock WH. Dental arch depth and width studied longitudinally
from 12 years of age to adulthood. Am J Orthod 1972;62:56-66.
27. Bishara SE, Treder JE, Damon P, Olsen M. Changes in the dental
arches and dentition between 25 and 45 years of age. Angle
Orthod 1996;66:417-22.
28. Bishara SE, Jakobsen JR, Treder J, Nowak A. Arch width
changes from 6 weeks to 45 years of age. Am J Orthod
Dentofacial Orthop 1997;111:401-9.
29. Tsujino K, Machida Y. A longitudinal study of the growth and
development of the dental arch width from childhood to adoles-
cence in Japanese. Bull Tokyo Dent Coll 1998;39:75-89.
Hesby et al 731
30. Bondevik O. Changes in occlusion between 23 and 34 years.
Angle Orthod 1998;68:75-80.
31. Brown VP, Daugaard-Jensen I. Changes in the dentition from the
early teens to the early twenties. Acta Odontol Scand 1951;9:
177-92.
32. Lundstrom A. Changes in crowding and spacing of the teeth with
age. Dent Practit 1969;19:218-24.
33. Carter GA, McNamara JA Jr. Longitudinal dental arch changes
in adults. Am J Orthod Dentofacial Orthop 1998;114:88-99.
34. Sinclair PM, Little RM. Maturation of untreated normal occlu-
sions. Am J Orthod 1983;83:114-23.
35. Waraaswapati N, Pitiphat W, Chandrapho N, Rattanayatikul C,
Karimbux N. Thickness of palatal masticatory mucosa associated
with age. J Periodontol 2001;72:1407-12.
36. Eger T, Muller HP, Heinecke A. Ultrasonic determination of
gingival thickness. J Clin Periodontol 1996;23:839-45.
37. Zar JH. 1984. Biostatistical analysis. 2nd ed. Englewood Cliffs,
NJ: Prentice-Hall; 1984. p. 323-5.
38. Hunt RJ. Percent agreement, Pearson’s correlation, and kappa
as measures of inter-examiner reliability. J Dent Res 1986;
65:128-30.
39. Ghosh M, Grizzle JE, Sen PK. Nonparametric methods in
longitudinal studies. J Am Stat Assoc 1973;68:29-36.
40. Dawson DV, Siegler IC. 1996. Approaches to the nonparametric
analysis of limited longitudinal data sets. Exp Aging Res 1996;
22:33-57.
41. Conover WJ. Practical nonparametric statistics. 2nd ed. New
York: Wiley; 1980.
42. Holm S. A simple sequentially rejective multiple test procedure.
Scand J Stat 1979;6:65-70.
43. Proffit WR. Equilibrium theory revisited, factors influencing
position of the teeth. Angle Orthod 1978;48:175-86.
44. McWade RA, Mamandras AH, Hunter WS. The effects of
Fränkel II treatment on arch width and arch perimeter. Am J
Orthod Dentofacial Orthop 1987;92:313-20.
45. You ZH, Fishman LS, Rosenblum RE, Subtelny JD. Dentoalve-
olar changes related to mandibular forward growth in untreated
Class II persons. Am J Orthod Dentofacial Orthop 2001;120:598-
607.
46. Lager H. The individual growth pattern and stage of maturation
as a basis for treatment of distal occlusion with overjet. Trans Eur
Orthod Soc 1967;135-45.
47. Ricketts RM. Perspectives in the clinical application of cepha-
lometrics, the first fifty years. Angle Orthod 1981;51:115-50.
48. Knott VB. Height and shape of the palate in girls; a longitudinal
study. Arch Oral Biol 1970;15:849-60.
49. Vanarsdall R. Transverse dimension and long-term stability.
Semin Orthod 1999;5:171-80.
50. The American Board of Orthodontics grading system for dental
casts and panoramic radiographs. St Louis: American Board of
Orthodontics; 1999.
51. Lima A, Lima A, Filho R, Oyen O. Spontaneous mandibular arch
response after rapid palatal expansion: a long-term study on
Class I malocclusion. Am J Orthod Dentofacial Orthop 2004;
126:576-82.