ORIGINAL ARTICLE

Midpalatal suture density ratio: A novel

predictor of skeletal response to rapid
maxillary expansion
Thorsten Gru € nheid,a Chad E. Larson,b and Brent E. Larsona
Minneapolis and Rochester, Minn

Introduction: During adolescence, increasing interdigitation of the midpalatal suture increases resistance to
rapid maxillary expansion (RME); this decreases its skeletal effect. In this study, we aimed at determining
whether a novel measure of midpalatal suture maturity, the midpalatal suture density ratio, can be used as a valid
predictor of the skeletal response to RME. Methods: The midpalatal suture density ratio, chronologic age, cer-
vical vertebral maturation, and the stage of midpalatal suture maturation were assessed before treatment for 30
patients (ages, 12.9 6 2.1 years) who underwent RME as part of comprehensive orthodontic treatment. Mea-
surements on cone-beam computed tomography scans were used to determine the proportions of prescribed
expansion achieved at the greater palatine foramina, the nasal cavity, and the infraorbital foramina. Results:
There was a statistically significant negative correlation between the midpalatal suture density ratio and both
the greater palatine foramina and the infraorbital foramina (r 5 0.7877 and 0.3647, respectively; P \0.05).
In contrast, chronologic age, cervical vertebral maturation, and stage of midpalatal suture maturation were
not significantly correlated to any of the assessed measures of skeletal expansion (r range, 0.2209 to
0.0831; P .0.05). Conclusions: The midpalatal suture density ratio has the potential to become a useful clinical
predictor of the skeletal response to RME. Conversely, chronologic age, cervical vertebral maturation, and stage
of midpalatal suture maturation cannot be considered useful parameters to predict the skeletal effects of RME.
(Am J Orthod Dentofacial Orthop 2017;151:267-76)

R
apid maxillary expansion (RME) is a commonly
used procedure to correct a transverse constric-
tion of the maxillary arch.1,2 The heavy forces
generated by the expander transmit through the teeth
into the maxillary bones and are intended to open the
midpalatal suture by separating the hemimaxillae; this
leads to subsequent bone deposition. These effects are
considered skeletal expansion and, in most cases, are
desired to be the sole effects of the force application.
However, dental tipping and bending of the alveolar
process, which are considered dentoalveolar expansion,
occur as well2-4 and have been reported to account for
a
Division of Orthodontics, School of Dentistry, University of Minnesota, Minne-
apolis, Minn.
b
Private practice, Rochester, Minn.
All authors have completed and submitted the ICMJE Form for Disclosure of Po-
tential Conflicts of Interest, and none were reported.
Address correspondence to: Thorsten Gr€ unheid, Division of Orthodontics, School
of Dentistry, University of Minnesota, 6-320 Moos Health Science Tower, 515
Delaware St SE, Minneapolis, MN 55455; e-mail, tgruenhe@umn.edu.
Submitted, December 2015; revised and accepted, June 2016.
0889-5406/$36.00
Ó 2017 by the American Association of Orthodontists. All rights reserved.
http://dx.doi.org/10.1016/j.ajodo.2016.06.043
39% to 49% and 6% to 13% of the total expansion,
respectively.5-7 Dentoalveolar expansion is often
unwanted, since dental tipping may lead to loss of
periodontal attachment level,8,9 fenestrations of the
buccal cortical bone,10 and dental root resorption.11
For these reasons, the goals of RME are typically to
maximize skeletal expansion and to minimize dentoal-
veolar expansion.
Closure of the craniofacial sutures, especially the
midpalatal suture, eventually makes skeletal expansion
by conventional RME impossible,4 necessitating surgi-
cally assisted maxillary expansion.12 The surgical
approach can involve either intraoperative widening of
the maxilla through a multisegment LeFort osteotomy
or surgically assisted RME.10,13 Both procedures are
invasive, costly, and associated with surgical risks.10,14
To help with the clinical decision whether correction of
a transverse discrepancy should be attempted with
conventional RME or whether surgically assisted
expansion is necessary, indicators of midpalatal suture
maturation have been proposed. These indicators
include sutural morphology as assessed on occlusal
radiographs,15 skeletal maturity indicators on
267
268
hand-wrist radiographs,16,17 cervical vertebral maturation
(CVM) indicators on lateral cephalograms,18 and a
recently suggested 5-stage classification of midpalatal
suture maturation.19 Up to now, however, none of these
indicators has been validated to predict the amount of skel-
etal expansion achieved with RME. A reliable way to
predict a patient's skeletal and dentoalveolar response to
RME before treatment would add a new parameter for
orthodontic treatment success. For instance, predicting
the response would allow an adolescent with early
closure of the suture to avoid the potential negative side
effects of conventional RME. In contrast, in a young
adult with late closure of the suture, surgically assisted
RME may be avoided in lieu of conventional RME, which
would prevent the added cost and risks of the surgical
procedure.
In the past decade, the use of cone-beam computed
tomography (CBCT) in clinical practice has increased as a
consequence of its diagnostic advantages over tradi-
tional 2-dimensional imaging for orthodontic treatment
planning.20,21 With CBCT, it is possible to visualize the
midpalatal suture in vivo without any overlapping
anatomic structures,4 which may allow the development
of a qualitative or quantitative assessment of midpalatal
suture maturation to assist with the decision about
whether conventional or surgically assisted maxillary
expansion is more appropriate.
The primary aim of this study was to test whether a
novel midpalatal suture maturation measure, the midpa-
latal suture density ratio, obtained from a pretreatment
CBCT image, can be used as a valid predictor of the
amount of skeletal response to RME treatment. A sec-
ondary and subsidiary aim was to assess the correlations
of the amount of skeletal response to RME treatment
with the midpalatal suture density ratio, chronologic
age, CVM, and stage of midpalatal suture maturation
to determine the relative clinical usefulness of each po-
tential predictor.
MATERIAL AND METHODS
The research protocol including the use of existing
CBCT scans was approved by the Institutional Review
Board at the University of Minnesota (study number
1402M48043). A total of 30 patients (13 boys, 17 girls;
ages, 12.9 6 2.1 years) who had RME as part of compre-
hensive orthodontic treatment at the University of
Minnesota School of Dentistry were included in this
retrospective cohort study. Patients were excluded if
they had a history of periodontal disease, previous or-
thodontic treatment, congenital malformations, or a
time period greater than 6 months between the pretreat-
ment CBCT scan and the beginning of treatment.
February 2017  Vol 151  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Gr€
unheid, Larson, and Larson
Table I. Descriptive summary of treatment variables
Variable Mean 6 SD Range
Age at expansion (y) 12.9 6 2.1 7.9-16.6
Prescribed amount of 8.6 6 2.2 3.3-13.0
expansion (mm)
Postexpansion retention 15.0 6 14.1 2.0-77.0
time (wk)
Total treatment time (mo) 28.7 6 9.4 12.0-49.0
In each patient, RME was carried out using a Hyrax-
type expander, which transmitted the force to the first
molars and both first and second premolars or, when
applicable, the deciduous molars. The prescribed
amount of expansion was determined before treatment
from measurements taken on digital casts and recalcu-
lated based on intraoral measurements taken at each pa-
tient's first RME-check appointment. The patients were
asked to activate the expander once daily, for expansion
of 0.25 mm per day, until adequate correction of dental
arch width—ie, the predetermined amount of expan-
sion—was achieved as determined by the clinical judg-
ment of the treating clinician. Patients who did not
comply with the instructions or who had difficulties acti-
vating the expander were excluded from the study. After
completion of activation, the expander was kept
passively in place for an average of 15 weeks. After
this postexpansion retention period, the expander was
removed, and preadjusted edgewise appliances were
placed to complete comprehensive orthodontic treat-
ment. A descriptive summary of treatment variables is
provided in Table I.
Full field-of-view (17 3 23 cm) CBCT scans were ob-
tained before (T1) and after (T2) comprehensive ortho-
dontic treatment using an i-CAT Next Generation unit
(Imaging Sciences International, Hatfield, Pa) at
120 kV and 18.54 mAs, with a pulsed scan time of
8.9 seconds as part of the diagnostic records for compre-
hensive orthodontic treatment per the protocol used at
the Division of Orthodontics at the University of Minne-
sota. The scan data were reconstructed with a voxel size
of 0.3 mm3. CBCT scans obtained at T1 and T2 were as-
signed random numeric identifiers to achieve blinding to
subject and time point during data collection and anal-
ysis.
Data collection was performed using digital imaging
and communications in medicine (DICOM) volumes.
Each DICOM volume was first oriented using a translu-
cent lateral view to achieve superimposition of the left
and right inferior orbital rims and the zygomatic pro-
cesses of the maxilla (Fig 1). Solid right and left lateral
views were then used to orient the Frankfort horizontal
parallel to the true horizontal.
Gr€
unheid, Larson, and Larson 269

Fig 1. Orientation of the 3-dimensional reconstructions in Dolphin Imaging software: A, a translucent

view was used to superimpose the orbital rims and the zygomatic processes of the maxilla; B, solid left
and right lateral views were used to orient the Frankfort horizontal parallel to the true horizontal.

Fig 2. Linear skeletal measurements. Distances were measured between A, the lateral margins of the
greater palatine foramina; B, the lateral walls of the nasal cavity; and C, the lateral margins of the in-
fraorbital foramina.

To quantify the skeletal effects of RME, linear mea-
surements were made between bilateral skeletal struc-
tures on slices from the T1 and T2 images using
Dolphin Imaging software (version 11.7; Dolphin Imag-
ing & Management Solutions, Chatsworth, Calif). All
measurements were made by 1 examiner (C.E.L.) and
repeated after a 6-week washout period for 10 randomly
chosen subjects to assess intraexaminer reliability. The
linear measurements included the distance between
the greater palatine foramina, which was measured be-
tween the lateral margins of the foramina on an axial
slice through the center of the hard palate; the maximum
width of the nasal cavity, which was measured on a cor-
onal slice through the center of the incisive foramen; and
the distance between the infraorbital foramina, which
was measured between the lateral margins of the
foramina on an axial slice (Fig 2). The distance between
the greater palatine foramina was selected because it al-
lows measurement of skeletal expansion at the posterior
hard palate, and the greater palatine foramen is an easily
and reproducibly identifiable landmark. The maximum
width of the nasal cavity and the distance between the
infraorbial foramina were selected because the nasal
walls and infraorbital foramina are reproducible
ascending skeletal landmarks and the effects of RME
have been reported to be pyramidal.5 Landmarks for
skeletal measurements were chosen lingually or apically
to the dentition to ensure that they remained unaffected
by treatment with preadjusted edgewise appliances. To
account for the individualized amount of expansion,
each distance was converted to a proportion of the pre-
scribed expansion by dividing the difference in distances
between T1 and T2 by the amount of prescribed expan-
sion (Table II).

American Journal of Orthodontics and Dentofacial Orthopedics February 2017  Vol 151  Issue 2
270
Table II. Skeletal effect measures
Measure Definition
Greater palatine foramina D Distance between the greater
proportion (GPFp) palatine foramina (GPFd)/
Prescribed amount of
expansion
Nasal cavity width D Maximum width of the nasal
proportion (NWp) cavity (NWd)/
Prescribed amount of
expansion
Infraorbital foramina D Distance between the
proportion (IOFp) infraorbital foramina (IOFd)/
Prescribed amount of
expansion
To determine the midpalatal suture density ratio, the
T1 images were oriented to the palatal plane in the
sagittal and frontal views to yield an axial slice through
the center and parallel to the hard palate (Fig 3). Gray
density measurements were then made on 0.3-mm slices
using Invivo5 software (version 5.5.2; Anatomage, San
Jose, Calif). The Invivo5 software assigns each voxel a
gray density value on a scale specific to the machine
and exposure settings. Average gray density values
were determined for defined regions of the suture
(GDs), soft palate (GDsp), and palatal process of the
maxilla (GDppm) (Fig 4). The defined region of the suture
was always determined on the most central axial slice
through the hard palate. The average gray density values
were used to calculate the midpalatal suture density
(MPSD) ratio by the following equation.
GDs GDsp
MPSD ratio 5
GDppm GDsp
This ratio ranges from 0 to 1, with lower values indi-
cating that the suture region is closer in density to the
soft palate and hence less calcified. Conversely, values
close to 1 indicate that the suture region is closer in den-
sity to the palatal process of the maxilla and hence more
highly calcified.
The CVM stage was assessed on a cephalometric
radiograph extracted from the T1 DICOM volume as
described by Franchi et al.22 The midpalatal suture
maturation stage was assessed on the T1 DICOM volume
using the 5-stage classification of midpalatal suture
development described by Angelieri et al.19
Statistical analysis
The categorical variables were described by the
number and percentage of subjects in each category.
For the continuous variables, mean values, median
values, standard deviations, and ranges were
calculated. Linear regression analysis was used to
February 2017  Vol 151  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Gr€
unheid, Larson, and Larson
determine the correlations between the independent
variables—age, CVM, midpalatal suture maturation,
and midpalatal suture density—and the dependent
variables—greater palatine foramina proportion
(GPFp), nasal cavity width proportion (NWp), and in-
fraorbital foramina proportion (IOFp). The P value of
the slope was used to determine whether a significant
correlation existed. Linear regressions with adjust-
ments for potentially confounding variables were
conducted to assess whether they affected the
P value. The potentially confounding variables that
were tested were sex, age at the beginning of RME,
retention time, and total treatment time. The Pearson
correlation coefficients and coefficients of determina-
tion were calculated for each relationship. Intraexa-
miner reliability was assessed using intraclass
correlation coefficients (ICC values) and Bland-
Altman plots for the linear and gray density measure-
ments to better visualize errors.23 Statistical analyses
were performed using SAS software (version 9.4; SAS
Institute, Cary, NC). P values of less than 0.05 were
considered statistically significant.
RESULTS
ICC values were greater than 0.95 for all measure-
ments, indicating excellent intraexaminer reliability.
The linear measurements (distance between the greater
palatine foramina. maximum width of the nasal cavity,
and distance between the infraorbital foramina) had a
mean ICC of 0.988. The gray density value measure-
ments had a mean ICC of 0.998. The CVM and midpala-
tal suture maturation had ICC values of 0.985 and 0.977,
respectively. Bland-Altman comparison of linear mea-
surements at 2 time points yielded a mean difference
of 0.02 mm with limits of agreement of 0.79 and
0.84 mm at 95% confidence; comparison of gray density
value measurements at 2 time points yielded a mean dif-
ference of 4.2 with limits of agreement of 76.3 and
67.9 at 95% confidence (Fig 5).
The distributions of CVM and midpalatal suture
maturation stages are shown in Table III. Mean values,
standard deviations, and ranges of the continuous inde-
pendent and dependent variables measured are shown in
Table IV.
The correlation between the prescribed amount of
expansion and the skeletal expansion achieved at the
level of the greater palatine foramina is shown in
Figure 6. The scatter plot exemplifies the variability in
the amount of skeletal expansion achieved with RME.
A scatter plot matrix of the proportional skeletal ef-
fect measures (GPFp, NWp, and IOFp) vs midpalatal su-
ture density ratio, age, CVM, and stage of midpalatal
Gr€
unheid, Larson, and Larson 271

Fig 3. Orientation of the 3-dimensional reconstructions in Invivo5 software: A, sagittal and B, frontal
views were used to orient the hard palate parallel to the true horizontal; C, central slice used for mea-
surements.

Fig 4. Regions used to determine average gray density values: A, the gray density of the suture was
determined in a 6 mm-wide rectangle centered on the midpalatal suture from the distal aspect of the
incisive foramen to the distal aspect of the first molar crown; B, the gray density of the palatal process
of the maxilla was determined in a 4 3 4-mm cortical portion of the palatal process of the maxilla, and
the gray density of the soft palate was determined in a 4 3 4-mm portion representing the soft palate.

suture maturation is shown in Figure 7. There was a
strong negative linear correlation (r 5 0.79) between
the midpalatal suture density ratio and GPFp as well as
a fair negative linear correlation (r 5 –0.36) between
the midpalatal suture density ratio and both NWp and
IOFp. Age, CVM, and midpalatal suture maturation
had weak linear correlations to all skeletal effect mea-
sures ( 0.25 \r \0).
The coefficients of determination (R2) for each corre-
lation are shown in Table V. Whereas 62% of the variation
in GPFp was explained by the midpalatal suture density
ratio, 13% of the variations of both NWp and IOFp
were explained by the midpalatal suture density ratio.
For each relationship, the null hypothesis that there
was no relationship (slope 5 b1 5 0) between variables
was tested. The P values for these tests are shown in
Table VI. For the relationships between the midpalatal
suture density ratio and both GPFp and IOFp, there
were statistically significant negative correlations
(P \0.0001 and P 5 0.0475, respectively). Therefore,

American Journal of Orthodontics and Dentofacial Orthopedics February 2017  Vol 151  Issue 2
272 Gr€
unheid, Larson, and Larson

Fig 5. Intraexaminer reliability assessment. Bland-Altman plots of agreement of A, linear measure-

ments and B, gray density measurements. In each plot, the data points represent the differences be-
tween the original measurements and the repeated measurements.

Table III. Descriptive statistics of categorical indepen- Table IV. Descriptive statistics of continuous inde-
dent variables pendent and dependent variables
Variable Category Occurrence Percentage Variable Mean 6 SD Range
Cervical vertebral Stage 1 4 13.3 Midpalatal suture density 0.54 6 0.14 0.28-0.80
maturation (CVM) (MPSD) ratio
Stage 2 4 13.3 Greater palatine foramina 0.18 6 0.10 0.00-0.36
Stage 3 4 13.3 proportion (GPFp)
Stage 4 8 26.7 Nasal cavity width 0.18 6 0.13 0.03 to 0.40
Stage 5 6 20.0 proportion (NWp)
Stage 6 4 13.3 Infraorbital foramina 0.25 6 0.13 0.00-0.55
Midpalatal suture Stage A 3 10.0 proportion (IOFp)
maturation (MPSM)
Stage B 9 30.0
Stage C 7 23.3 of RME success. In this study, we determined whether a
Stage D 5 16.7 novel measurement of midpalatal suture maturation,
Stage E 6 20.0
the midpalatal suture density ratio, is a valid predictor
of the amount of skeletal response to RME treatment.
the null hypotheses for these variables were rejected. The In contrast to chronologic age, CVM, and stage of midpa-
null hypotheses were not rejected for the remaining re- latal suture maturation, the midpalatal suture density ra-
lationships. Linear associations with adjustments for tio was found to have a significant correlation with
potentially confounding variables—sex, age at RME, measures of skeletal response to RME treatment.
retention time, and total treatment time—did not signif- A retrospective study design was chosen to ensure
icantly influence the P value of any association tested. that patients were not exposed to radiation solely for
The statistically significant negative linear associa- study purposes. As a consequence, there were a few lim-
tion between GPFp and midpalatal suture density itations. First, the study sample was a convenience sam-
(MPSD) ratio can be represented by the following ple. Generalizations about the total population from this
least-squares linear regression equation. sample must be made with due caution. Second, the
GPFp 5 0:60 3 MPSD ratio 1 0:50 amount of expansion was tailored to each patient and,
therefore, not standardized. Ideally, each patient would
have had the same amount of expansion because the
proportion of dentoalveolar to skeletal effects may
DISCUSSION vary at different magnitudes of RME. Although stan-
Skeletal maturation from adolescence to young dardizing the amount of expansion was not deemed
adulthood involves progressive closure of the midpalatal realistic to correct each patient's unique clinical presen-
suture, which causes increasing impedance and tation, the amount of expansion was accounted for by
decreasing skeletal response to RME, and eventual failure reporting the skeletal effect measures as proportions.
to separate the hemimaxillae. Therefore, a personalized Furthermore, the activation protocol was standardized
assessment of a patient's stage of skeletal maturation for all subjects. Third, the true amount of expansion
in the area of interest would be an important predictor was not recorded, only the prescribed amount of

February 2017  Vol 151  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Gr€
unheid, Larson, and Larson
Fig 6. Correlation between the prescribed amount of expansion and the skeletal expansion achieved
at the level of the greater palatine foramina (GPF). A best-fit line is shown with the Pearson correlation
coefficient.
expansion; this added patient compliance as a potential
confounding factor. However, as noted above, uncom-
pliant patients were excluded from the study sample,
and ultimately the predetermined amount of expansion
was achieved for all patients.
The midpalatal suture density ratio was determined
on CBCT scans as a measure of midpalatal suture matu-
ration. In the infantile period, the suture is a broad gap
between the maxillary bones and is filled with connective
tissue.24 Since this tissue is not calcified and, therefore,
radiolucent, the suture area is equivalent in gray scale
value to the tissue of the soft palate, and the midpalatal
suture density ratio is close to 0. As maturation pro-
gresses into the juvenile stage, bony spicules begin to
form on the margins of the suture, resulting in a mixture
of noncalcified connective tissue and calcified bone. As a
consequence, the midpalatal suture density ratio in-
creases throughout the maturation process. In the
adolescent period, the bony spicules become increas-
ingly interdigitated, and the suture area becomes calci-
fied to a degree similar to cortical bone, resulting in a
midpalatal suture density ratio near 1.
With CBCT, it is not possible to directly compare the
gray density value from 1 scan with another.25 In
contrast to medical computed tomography images,
where 1000 Hounsfield units always indicate air and
0 Hounsfield units always indicate water, attenuation
coefficients in CBCT images are not standardized. Be-
tween CBCT scanners, the dynamic contrast and gray
density values are influenced by technical factors such
as x-ray beam hardening, scatter radiation, and projec-
tion data discontinuity related effect.25 However, ac-
cording to Cassetta et al,26 a linear relationship exists
in reconstructed CBCT images between gray density
values and the true attenuation coefficients. If this equa-
tion is rearranged and plugged into the midpalatal su-
ture density ratio for each measured gray density
American Journal of Orthodontics and Dentofacial Orthopedics
273
value, the parameters affected by machine and condi-
tions cancel out, and only the ratio of attenuation coef-
ficients is left. It follows from this that the midpalatal
suture density ratio is equivalent to a ratio of absolute
values, allowing it to be compared between different
scanners and scanning conditions.
This study did not show the pyramidal maxillary skel-
etal response to RME that has been reported by
others.7,27,28 The most superior measurement, IOFp,
achieved a larger mean proportion of prescribed
expansion than did the more inferior measurements
GPFp and NWp. The different expansion pattern is
most likely related to the different time points at
which effects were measured. Although we measured
the long-term skeletal effects of RME at an average of
28.7 months after initiation of expansion, other authors
have assessed the immediate skeletal effects of RME,
either directly after the active phase of RME7,28 or
upon expander removal27 at approximately 3 weeks or
6 months later, respectively. It can be argued that for
both clinicians and patients the long-term effects of
RME as assessed in this study are more relevant than
the immediate effects.
However, the long-term effects are also more
affected by growth and remodeling. It must be assumed
that growth and remodeling of the nasomaxillary com-
plex occurred in our study population during the 28.7-
month postexpansion period. During nasomaxillary
growth, the maxilla is displaced anteriorly and inferiorly
relative to the anterior cranial base,29 and bone deposi-
tion occurs at the circummaxillary sutures.30,31 This
displacement has been reported to occur at about 45
to the anterior cranial base and at average rates of 0.6
to 0.7 mm per year.29,32 At the same time, active
resorption takes place on the surface of the maxilla,
including the orbital rims and infraorbital foramina.31
The net effect is anterior, inferior, and lateral movement
February 2017  Vol 151  Issue 2
274 Gr€
unheid, Larson, and Larson

Fig 7. Correlations between the skeletal effect measures (GPFp, NWp, and IOFp) and midpalatal su-
ture density (MPSD) ratio, chronologic age, CVM, and stage of midpalatal suture maturation (MPSM).
Best-fit lines are shown on each plot with Pearson correlation coefficients.

of the infraorbital foramina. This process may account
for the greater increase in width noted at the infraorbital
foramina. Remodeling also occurs on the inside of the
nasal cavity that may contribute to the increase in nasal
width noted. In contrast, remodeling of the hard palate
occurs mainly as bone resorption at the superior surface
and apposition at the inferior surface causing movement
mostly in the vertical dimension.31 This remodeling is
paired with transverse maxillary growth including su-
tural separation of the hemimaxillae that is greater pos-
teriorly than anteriorly and has been reported to occur at
average rates of 0.12 to 0.48 mm per year.32-35 It must
be assumed that this sutural growth affected maxillary
width at the level of the hard palate, where GPFp was
measured.
The proportion of skeletal expansion we measured at
the hard palate, an average of 18% of the prescribed
expansion, was generally lower than the proportions re-
ported elsewhere.4,7,14,36 Ballanti et al36 reported in a
CBCT study opening of the midpalatal suture of 26.6%
at the midpoint between the anterior and posterior nasal
spines. Using a similar approach, Podesser et al37 re-
ported opening of the midpalatal suture as 22.9% of
the total expansion on a coronal section through the
maxillary first molar, whereas Weissheimer et al7 re-
ported 36%. However, it is virtually impossible to directly
compare the values because different landmarks were
used in each study, and each study population had a
different age distribution. Even more importantly, the
values reported elsewhere were immediate RME effects,
whereas we assessed long-term effects. It is reasonable
to believe that in the long term the proportions reported
elsewhere may decrease close to or even below the pro-
portion in our study.

February 2017  Vol 151  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Gr€
unheid, Larson, and Larson 275

2 not significantly correlated with any measures of skeletal

Table V. Coefficients of determination (R )
response to RME treatment and can therefore not be
Greater palatine Nasal cavity Infraorbital considered a clinically useful predictor. These findings
foramina width foramina corroborate previous assumptions that, because of large
proportion proportion proportion
Variable (GPFp) (NWp) (IOFp) interindividual variations, age is not well associated with
MPSD ratio 0.6204 0.1303 0.1330 midpalatal suture maturation.15,38
Age 0.0048 0.0174 0.0010 Similarly, CVM did not have a significant correlation
CVM 0.0000 0.0013 0.0488 with any measures of skeletal response to RME treat-
MPSM 0.0002 0.0069 0.0002 ment used in this study. At first glance, this appears to
MPSD, Midpalatal suture density; CVM, cervical vertebral matura- contrast with the findings reported by Baccetti et al,18
tion; MPSM, midpalatal suture maturation. who found skeletal changes as measured on posteroan-
terior cephalograms to be significantly larger in an early
treatment group characterized by CVM stages 1 to 3
Table VI. P values for the statistical test of linear rela- than in a late treatment group characterized by CVM
tionship stages 4 to 6. However, these groupings of CVM stages
Greater palatine Nasal cavity Infraorbital
(CVM 1-3 vs 4-6) simply discriminate between the times
foramina width foramina before and after the peak in skeletal maturity and have
proportion proportion proportion little predictive value for quantitative changes.
Variable (GPFp) (NWp) (IOFp)
Finally, the stage of midpalatal suture maturation as
MPSD ratio \0.0001* 0.0500 0.0475*
Age 0.7153 0.4865 0.8671
proposed by Angelieri et al19 did not have a significant
CVM 0.9886 0.8526 0.2407 correlation with any measures of skeletal response to
MPSM 0.9413 0.6631 0.9361 RME treatment. Although this method of classification
MPSD, Midpalatal suture density; CVM, cervical vertebral matura- by visual assessment of sutural morphology on CBCT
tion; MPSM, midpalatal suture maturation. scans was found to be reliable, its usefulness for the indi-
*Statistically significant. vidualized assessment of midpalatal suture morphology
before RME seems to be limited. If a pretreatment CBCT
image is available, determining the midpalatal suture
The midpalatal suture density ratio has several density ratio should be preferred because it is a more
potentially valuable clinical applications. The first appli- valid predictor of the amount of skeletal response to
cation is to determine whether conventional or surgically RME treatment.
assisted expansion is indicated in patients where the
RME response was previously unpredictable, such as ad- CONCLUSIONS
olescents and young adults. The second application is to
estimate the proportion of skeletal effects before expan- 1. The results suggest that the midpalatal suture den-
sion. For instance, to get adequate skeletal correction, sity ratio, a novel measure of midpalatal suture
patients with a lower proportion of skeletal effects will maturation, has a significant negative correlation
need more total expansion since they will exhibit more (r 5 0.7877; P \0.05) with the amount of
tipping of the anchor teeth. Uprighting these teeth typi- long-term maxillary skeletal expansion achieved
cally leads to partial loss of the increased intermolar dis- from RME at the level of the palate.
tance. Conversely, patients with a higher portion of 2. The amount of long-term skeletal response to RME
skeletal effects will need less expansion to achieve the can be estimated on a pretreatment CBCT image by
same skeletal correction because the anchor teeth tip the equation GPFp 5 0.60 3 midpalatal suture
less and can be maintained close to their immediate density ratio10.50:
postexpansion position. These applications of the mid- 3. The midpalatal suture density ratio has the potential
palatal suture density ratio make it a powerful clinical to become a useful clinical predictor of the skeletal
diagnostic parameter that will help tailor RME treatment response to RME and aid in clinical decisions on
for precise and efficient treatment while minimizing un- whether conventional RME therapy will be a suc-
wanted effects. cessful treatment. Moreover, the midpalatal suture
In addition to the midpalatal suture density ratio, we density ratio was found to be a more useful predic-
assessed other previously suggested potential predictors tor than chronologic age, CVM, or midpalatal suture
of skeletal response to RME treatment such as age, CVM, maturation stage, which had no significant correla-
and stage of midpalatal suture maturation to determine tions (r range, 0.2209 to 0.0831; P .0.05) to the
their relative clinical usefulness. Chronologic age was amount of long-term skeletal expansion.

American Journal of Orthodontics and Dentofacial Orthopedics February 2017  Vol 151  Issue 2
276
REFERENCES
1. Bishara SE, Staley RN. Maxillary expansion: clinical implications.
Am J Orthod Dentofacial Orthop 1987;91:3-14.
2. Zuccati G, Casci S, Doldo T, Clauser C. Expansion of maxillary
arches with crossbite: a systematic review of RCTs in the last 12
years. Eur J Orthod 2013;35:29-37.
3. Lagravere MO, Heo G, Major PW, Flores-Mir C. Meta-analysis of
immediate changes with rapid maxillary expansion treatment. J
Am Dent Assoc 2006;137:44-53.
4. Liu S, Xu T, Zou W. Effects of rapid maxillary expansion on the
midpalatal suture: a systematic review. Eur J Orthod 2015;37:
651-5.
5. Garrett BJ, Caruso JM, Rungcharassaeng K, Farrage JR, Kim JS,
Taylor GD. Skeletal effects to the maxilla after rapid maxillary
expansion assessed with cone-beam computed tomography. Am
J Orthod Dentofacial Orthop 2008;134:8-9.
6. Lione R, Ballanti F, Franchi L, Baccetti T, Cozza P. Treatment and
posttreatment skeletal effects of rapid maxillary expansion studied
with low-dose computed tomography in growing subjects. Am J
Orthod Dentofacial Orthop 2008;134:389-92.
7. Weissheimer A, de Menezes LM, Mezomo M, Dias DM, de Lima EM,
Rizzatto SM. Immediate effects of rapid maxillary expansion with
Haas-type and hyrax-type expanders: a randomized clinical trial.
Am J Orthod Dentofacial Orthop 2011;140:366-76.
8. Northway WM, Meade JB. Surgically assisted rapid maxillary
expansion: a comparison of technique, response, and stability.
Angle Orthod 1997;67:309-20.
9. Rungcharassaeng K, Caruso JM, Kan JY, Kim J, Taylor G. Factors
affecting buccal bone changes of maxillary posterior teeth after
rapid maxillary expansion. Am J Orthod Dentofacial Orthop
2007;132:428.e1-8.
10. Suri L, Taneja P. Surgically assisted rapid palatal expansion: a
literature review. Am J Orthod Dentofacial Orthop 2008;133:
290-302.
11. Baysal A, Karadede I, Hekimoglu S, Ucar F, Ozer T, Veli I, et al. Eval-
uation of root resorption following rapid maxillary expansion us-
ing cone-beam computed tomography. Angle Orthod 2012;82:
488-94.
12. Chrcanovic BR, Custodio AL. Orthodontic or surgically assisted
rapid maxillary expansion. Oral Maxillofac Surg 2009;13:123-37.
13. Yao W, Bekmezian S, Hardy D, Kushner HW, Miller AJ, Huang JC,
et al. Cone-beam computed tomographic comparison of surgically
assisted rapid palatal expansion and multipiece Le Fort I osteot-
omy. J Oral Maxillofac Surg 2014;73:499-508.
14. Lagravere MO, Major PW, Flores-Mir C. Long-term skeletal
changes with rapid maxillary expansion: a systematic review. Angle
Orthod 2005;75:1046-52.
15. Wehrbein H, Yildizhan F. The mid-palatal suture in young adults. A
radiological-histological investigation. Eur J Orthod 2001;23:
105-14.
16. Revelo B, Fishman LS. Maturational evaluation of ossification of
the midpalatal suture. Am J Orthod Dentofacial Orthop 1994;
105:288-92.
17. Fishman LS. Radiographic evaluation of skeletal maturation. A
clinically oriented method based on hand-wrist films. Angle Or-
thod 1982;52:88-112.
18. Baccetti T, Franchi L, Cameron CG, McNamara JA Jr. Treatment
timing for rapid maxillary expansion. Angle Orthod 2001;71:
343-50.
19. Angelieri F, Cevidanes LH, Franchi L, Goncalves JR, Benavides E,
McNamara JA Jr. Midpalatal suture maturation: classification
February 2017  Vol 151  Issue 2 American Journal of Orthodontics and Dentofacial Orthopedics
Gr€
unheid, Larson, and Larson
method for individual assessment before rapid maxillary expan-
sion. Am J Orthod Dentofacial Orthop 2013;144:759-69.
20. Noar JH, Pabari S. Cone beam computed tomography—current un-
derstanding and evidence for its orthodontic applications? J Or-
thod 2013;40:5-13.
21. De Vos W, Casselman J, Swennen GR. Cone-beam computerized
tomography (CBCT) imaging of the oral and maxillofacial region:
a systematic review of the literature. Int J Oral Maxillofac Surg
2009;38:609-25.
22. Franchi L, Baccetti T, McNamara JA Jr. Mandibular growth as
related to cervical vertebral maturation and body height. Am J Or-
thod Dentofacial Orthop 2000;118:335-40.
23. Bland JM, Altman DG. Statistical methods for assessing agreement
between two methods of clinical measurement. Lancet 1986;1:
307-10.
24. Melsen B. Palatal growth studied on human autopsy material. A
histologic microradiographic study. Am J Orthod 1975;68:42-54.
25. Arisan V, Karabuda ZC, Avsever H, Ozdemir T. Conventional multi-
slice computed tomography (CT) and cone-beam CT (CBCT) for
computer-assisted implant placement. Part I: relationship of
radiographic gray density and implant stability. Clin Implant
Dent Relat Res 2013;15:893-906.
26. Cassetta M, Stefanelli LV, Pacifici A, Pacifici L, Barbato E. How ac-
curate is CBCT in measuring bone density? A comparative CBCT-
CT in vitro study. Clin Implant Dent Relat Res 2014;16:471-8.
27. Kanomi R, Deguchi T, Kakuno E, Takano-Yamamoto T,
Roberts WE. CBCT of skeletal changes following rapid maxillary
expansion to increase arch-length with a development-
dependent bonded or banded appliance. Angle Orthod 2013;83:
851-7.
28. Leonardi R, Sicurezza E, Cutrera A, Barbato E. Early post-treatment
changes of circumaxillary sutures in young patients treated with
rapid maxillary expansion. Angle Orthod 2011;81:36-41.
29. Iseri H, Solow B. Growth displacement of the maxilla in girls stud-
ied by the implant method. Eur J Orthod 1990;12:389-98.
30. Enlow DH, Bang S. Growth and remodeling of the human maxilla.
Am J Orthod 1965;51:446-64.
31. Enlow D, Hans M. Essentials of facial growth. 1st ed. Philadelphia:
W. B. Saunders; 1996.
32. Bj€ork A, Skieller V. Growth of the maxilla in three dimensions as
revealed radiographically by the implant method. Br J Orthod
1977;4:53-64.
33. Bj€ork A, Skieller V. Growth in width of the maxilla studied by the
implant method. Scand J Plast Reconstr Surg 1974;8:26-33.
34. Hesby RM, Marshall SD, Dawson DV, Southard KA, Casko JS,
Franciscus RG, et al. Transverse skeletal and dentoalveolar changes
during growth. Am J Orthod Dentofacial Orthop 2006;130:
721-31.
35. Korn EL, Baumrind S. Transverse development of the human jaws
between the ages of 8.5 and 15.5 years, studied longitudinally with
use of implants. J Dent Res 1990;69:1298-306.
36. Ballanti F, Lione R, Baccetti T, Franchi L, Cozza P. Treatment and
posttreatment skeletal effects of rapid maxillary expansion inves-
tigated with low-dose computed tomography in growing subjects.
Am J Orthod Dentofacial Orthop 2010;138:311-7.
37. Podesser B, Williams S, Crismani AG, Bantleon HP. Evaluation of
the effects of rapid maxillary expansion in growing children using
computer tomography scanning: a pilot study. Eur J Orthod 2007;
29:37-44.
38. Persson M, Thilander B. Palatal suture closure in man from 15 to
35 years of age. Am J Orthod 1977;72:42-52.