Int. J. Oral Maxillofac. Surg.

2008; 37: 228–231

doi:10.1016/j.ijom.2007.06.020, available online at http://www.sciencedirect.com

Clinical Paper
Orthognathic Surgery

A cephalometric evaluation of T. Muto1, A. Yamazaki2,

S. Takeda1
1
Department of Oral & Maxillofacial Surgery,

the pharyngeal airway space in School of Dentistry, Health Sciences

University of Hokkaido, Japan; 2Department
of Orthodontics, School of Dentistry, Health
Sciences University of Hokkaido, Hokkaido,

patients with mandibular Japan

retrognathia and prognathia,

and normal subjects
T. Muto, A. Yamazaki, S. Takeda: A cephalometric evaluation of the pharyngeal
airway space in patients with mandibular retrognathia and prognathia, and normal
subjects. Int. J. Oral Maxillofac. Surg. 2008; 37: 228–231. # 2007 International
Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights
reserved.

Abstract. The antero-posterior diameter of the pharyngeal airway space (PAS) at the
level of the soft palate and base of the tongue was assessed in age-matched females
with a normal mandible (n = 31), mandibular retrognathism (n = 30) or mandibular
prognathism (n = 38). All subjects were examined by lateral cephalometry.
Measured variables were corrected with the use of appropriate regression equations
to eliminate the effects of head posture on the PAS. The corrected data showed more
Key words: pharyngeal airway space; lateral
clear-cut differences in the PAS among the three groups than did the measured data.
cephalogram; craniofacial morphology; man-
Pharyngeal airway diameter was largest in the group with mandibular prognathism, dibular retrognathism; mandibular prognath-
followed by the normal mandible and mandibular retrognathism groups. These ism.
results indicate that the antero-posterior dimension of the PAS is affected by
different skeletal patterns of the mandible. Accepted for publication 6 June 2007

Advancement and setback operations are
standard procedures for the correction of
mandibular retrognathism and prognath-
ism. Surgery for mandibular deformity
alters skeletal and soft-tissue components,
including the pharyngeal airway space
(PAS)1,2. The possibility of obstructive
sleep apnoea (OSA) after setback surgery
has been reported and discussed2.
Evaluation of the PAS thus has a very
important role in diagnosis and treatment
planning in patients with OSA and dento-
facial deformity. Cephalometric radiogra-
phy, computed tomography (CT) and
magnetic resonance imaging (MRI) have
been used to study the PAS. Although CT
and MRI can provide a three-dimensional
assessment, the results of different studies
are difficult to compare owing to a lack of
standardized protocols defining the thick-
ness, direction and precise location of sec-
tions14. Cephalometry cannot be applied at
a constant head posture, as is also the case
for the above two methods, and does not
provide information on the transverse
dimension of the airway. Cephalometric
analysis of the airway does permit precise
measurements in a sagittal plane, and has
the advantages of convenience, low cost
and minimal exposure to radiation10,11.
Changes in head posture can affect the
size of the PAS4,10,11. A significant rela-
tion (r = 0.807) between head posture and

0901-5027/030228 + 04 $30.00/0 # 2007 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.

Table 1. Characteristics of subjects
Group
Normal mandible Mandibular retrognathism
Subjects (n) 31
Age (years) 20–30
SNB (8) 76  SNB  82
SNB: angle between SN and NB. See Fig. 1 for cephalometric variables.
the PAS has previously been demon-
strated10. Since craniofacial morphology
is related to the PAS in non-obese patients
with OSA, cephalometric radiographs have
been used to measure the craniofacial and
upper airway structures5–9. Previous stu-
dies attempting to measure the PAS did
not take head posture into account, so the
precise relationship between the craniofa-
cial morphology and the PAS remains
unclear. The measured values were cor-
rected in the authors’ previous study11 to
avoid the effects of head posture, and this
demonstrated that craniofacial morphology
was related to the PAS in normal subjects.
To extend these findings, the present study
examined the relationship between cranio-
facial morphology and the PAS in patients
with mandibular deformities (mandibular
retrognathism and mandibular prognath-
ism), as compared to normal subjects.
Materials and methods
All subjects were selected from the files of
patients registered at the Department of
Fig. 1. Reference points and lines on the lateral cephalogram.
Evaluation of the pharyngeal airway space
Mandibular prognathism
30 38
17–30 18–32
SNB < 76 SNB > 82
Orthodontics at the Health Sciences Uni-
versity Hospital (Table 1). Female sub-
jects were divided into three groups:
mandibular retrognathism (n = 30; mean
age 22.3 years, range 17–30; defined by
angle SNB < 768), normal mandible
(n = 31; mean age 23.5 years, range 20–
30; 768  SNB  828) and mandibular
prognathism (n = 38; mean age 21.7 years,
range 18–32; SNB > 828).
Lateral cephalograms were taken with
the orbital-auricular plane parallel to the
floor and the teeth in centric occlusion or
centric relation at the end of the expiratory
phase. All subjects received the same
instructions for radiographic positioning
and were told to place the tongue in a
relaxed position and to breathe through
their nose after swallowing.
The reference points, lines (Fig. 1) and
variables in the cephalometric analysis
have been reported elsewhere11. Head pos-
ture (OPT/NSL) was defined as the cranio-
cervical angulation at the uppermost part of
the cervical spine, and PAS was defined as
the minimal sagittal linear distance
229
between the uvula and the posterior phar-
yngeal airway space (PAS–UP), and the
back of the tongue and the posterior phar-
yngeal airway space (PAS–TP). To mini-
mize the effect of head posture on the PAS,
measurements were corrected using regres-
sion equations10: y = 14.063 + 0.258x,
r = 0.628 for PAS–UP (y: PAS–UP, x:
head posture) and y = 27.177 + 0.39x,
r = 0.807 for PAS–TP (y: PAS–TP, x: head
posture). Correlations of cephalometric
variables with the measured PAS and the
corrected PAS were tested by linear regres-
sion analysis, performed with StatView for
Macintosh. The same statistical package
was used for descriptive statistics and sim-
ple linear regression. Statistically signifi-
cant differences were noted as P < 0.01 or
P < 0.05.
Twenty lateral cephalograms obtained
from randomly selected subjects were
retraced and reanalysed on another day
by the same investigator, and the method
error for each variable was calculated to
determine measurement reproducibility.
The method error P was calculated as
1=2
method error = ð d 2 =2nÞ , where d
represents the difference between the first
and second measurements, and n denotes
the sample size.
Results
Method errors for each variable were
within 1 mm and similar to those in pre-
vious studies10,11. No statistically signifi-
cant differences were noted and the results
were considered highly reproducible.
The craniofacial morphologic variables
obtained in the three groups are shown in
Table 2. The size of the mandible (SNA,
SNP, Go–Gn) and the developmental direc-
tion of the mandible (facial axis) character-
istically differed among the three groups.
The length and angulation of the uvula
significantly differed among the three
groups (34.7 mm and 126.68 in the normal
mandible, 36.9 mm and 132.18 in mandib-
ular retrognathism, and 31.7 mm and
118.88 in mandibular prognathism). Both
of these variables were large when the
mandible was small.
A decreased OPT/NSL ratio correlated
with a large mandible. As compared with
the measured PAS–UP, PAS–TP, the cor-
rected data were greater in the normal
mandible and mandibular prognathism
and smaller in mandibular retrognathism.
The corrected data showed more clear-cut
differences in the PAS among groups than
did the measured data.
The relationship between the PAS and
craniofacial morphology was examined
with the corrected data because the PAS
230 Muto et al.

Table 2. Cephalometric data in different skeletal patterns

Variable Normal mandible Mandibular retrognathism Mandibular prognathism
SNA (8) 82.8  3.3 79.2  3.5** 83.1  3.2++
SNB (8) 79.6  3.1 73.0  2.0** 87.1  2.8**,++
SNP (8) 79.3  3.3 72.2  1.9** 86.9  2.9**,++
Go–Gn (mm) 76.1  3.6 72.4  4.2** 83.7  4.0**,++
Facial axis (8) 85.0  4.1 78.1  4.2** 89.4  4.8**,++
PNS–Ba (mm) 44.9  3.3 44.6  3.4 45.2  3.3
Uvula length (mm) 34.7  3.7 36.9  4.0* 31.7  4.2**,++
Uvula thickness (mm) 9.7  1.8 8.9  1.1* 9.3  1.5
Uvula angulation (8) 126.6  6.9 132.1  5.7** 118.8  6.0**,++
OPT/NSL (8) 100.0  9.3 110.1  8.4** 93.0  7.4**,++
PAS–UP (mm) (corrected data) 10.2  2.8 (11.0  3.0) 7.4  2.3** (5.7  2.3**) 12.5  3.1**,++ (15.1  3.5**,++)
PAS–TP (mm) (corrected data) 11.0  3.6 (12.2  3.5) 8.3  3.5** (5.6  3.2**) 12.8  3.7*,++ (16.8  3.4**,++)
SNA: angle between SN and NA; SNB: angle between SN and NB; SNP: angle between SN and NP; Go–Gn: distance between Go and Gn; facial
axis: angle between PtGn (line through Pt and Gn) and BaN (line through Ba and N); PNS–Ba: distance between PNS and Ba: uvula length:
distance between PNS and UT; uvula thickness: thickest part of the uvula; uvula angulation: angle between ANS–PNS line and PNS–UT line;
OPT/NSL: angle between OPT and NSL; PAS–UP: minimal PAS between the uvula and the posterior pharyngeal wall; PAS–TP: minimal PAS
between the back of the tongue and the posterior pharyngeal wall. See Fig. 1 for cephalometric variables. Data are expressed as means  SD.
The following regression equations were used to correct the data: y = 14.063 + 0.25x (y = PAS–UP, x = OPT/NSL, r = 0.628); and
y = 27.177 + 0.39x (y = PAS–TP, x = OPT/NSL, r = 0.807). The measured data were corrected at an OPT/NSL of 1038 as follows: in a
patient with an OPT/NSL of 908, PAS–TP was calculated as the measured value + 138(103–90)  0.39.
*
P < 0.05 vs. normal mandible (control), **P < 0.01 vs. normal mandible (control), +P < 0.05 vs. mandibular retrognathism, ++P < 0.01 vs.
mandibular retrognathism.

Table 3. Variables correlating with PAS–UP

Normal mandible Mandibular retrognathism Mandibular prognathism Whole group
**
PNS–Ba 0.071 0.207 0.457 0.226
Uvula length 0.522 ** 0.488 ** 0.281 0.588 **
Facial axis 0.063 0.020 0.358* 0.640 **
PNS–Ba: distance between PNS and Ba; uvula length: distance between PNS and UT; facial axis: angle between PtGn (line through Pt and Gn) and
BaN (line through Ba and N). See Fig. 1 for cephalometric variables.
Values are correlation coefficients. *P < 0.05, **P < 0.01.

was affected by head posture (Tables 3 and The correlation between PAS–TP and C3– Besides the authors’ previous study11,
4). PAS–UP correlated with PNS–Ba in Me was particularly strong (0.725). In gen- one other investigation has taken head
mandibular prognathism (r = 0.457), uvula eral, a large mandible correlated with a posture into account when measuring
length in the normal mandible (r = 0.522) large PAS. the PAS15: cephalograms measured at
and mandibular retrognathism (0.488), the Frankfort horizontal plane more than
and facial axis in mandibular prognathism 108 above the true horizontal plane were
(r = 0.358). When the whole study group Discussion deleted to avoid bias. Another previous
was analysed, PAS–UP correlated with This study shows that the antero-posterior study10 demonstrated that a change in
uvula length (0.588) and facial axis size of the PAS clearly differed among craniocervical inclination of 108 alters
(0.640). PAS–TP correlated with Ar–Gn subjects with a normal mandible, man- the PAS by about 4 mm. Efforts should
in the normal mandible, SNP in mandibular dibular retrognathism and mandibular therefore be made to measure the PAS
prognathism, C3–Me in all three groups prognathism. The analysis, based on more accurately. Previous studies have
and facial axis in mandibular prognathism. cephalometric radiographs and taking reported that the PAS in OSA syndrome
When all subjects were analysed together, head posture into account, confirms the does not differ from that in the absence of
PAS–TP correlated with five variables (Ar– hypothesis that craniofacial morphology is this syndrome5,17. Perhaps this was
Gn, SNP, Go–Gn, C3–Me and facial axis). related to the PAS. because the data were obtained in different
head postures. Similar results have been
Table 4. Variables correlating with PAS–TP obtained in studies using MRI12,13.
Normal Mandibular Mandibular Whole Although CT and MRI can accurately
mandible retrognathism prognathism group determine upper airway cross-sectional
Ar–Gn 0.407 * 0.002 0.230 0.529 ** area and volume, head posture is not
SNP 0.266 0.134 0.379 * 0.572 ** maintained in a constant position. Unlike
Go–Gn 0.202 0.071 0.141 0.456 ** cephalographic data10, data obtained by
C3–Me 0.580 ** 0.653** 0.723 ** 0.725 ** CT and MRI cannot be corrected because
Facial axis 0.016 0.122 0.386 * 0.494 ** validated techniques are lacking.When the
Ar–Gn: distance between Ar and Gn; SNP: angle between SN and NP; Go–Gn: distance between raw data for the PAS were compared with
Go and Gn; C3–Me: distance between C3 and Me; facial axis: angle between PtGn (line through the corrected data,, —
Pt and Gn) and BaN (line through Ba and N). See Fig. 1 for cephalometric variables. When the raw data for the PAS were
Values are correlation coefficients. *P < 0.05, **P < 0.01. corrected, differences in PAS were clearly

seen between mandibular prognathism
and mandibular retrognathism. The PAS
in the mandibular prognathism group was
considerably larger than that in the man-
dibular retrognathism group. The antero-
posterior position of the mandible most
likely affects the position of the base of
the tongue because the tongue is con-
nected to the inner surface of the anterior
mandible by the genioglossus muscle.
Comminuted fractures of the anterior
mandible are associated with glossopto-
sis, leading to pathologic sequelae. This
condition is caused by the fracture-
induced posterior shift of the attached
area of the genioglossus muscle. Pierre
Robin syndrome is also often associated
with glossoptosis (a retrognathic mand-
ible holding the tongue back), thus
restricting the PAS. Oral appliances are
a viable option for treating patients with
OSA syndrome, especially because a ret-
rognathic mandible is often involved3,6.
Treatment with oral appliances is
designed to reposition the mandible and
tongue anteriorly, thereby enlarging the
airway. From the present results, demon-
strating a direct relationship between the
position of the mandible (mandibular ret-
rognathism, normal subjects and mandib-
ular prognathism) and the PAS, the use of
oral appliances may be acceptable.
Interestingly, the results show that the
PAS–UP (PAS at the level of soft palate)
differed among the three groups (Table 2),
whereas the position of the maxilla (PNS–
Ba) did not. The PAS was largest in
mandibular prognathism, followed by a
normal mandible and mandibular retro-
gnathism. The length and angle of the soft
palate also differed among the groups.
These two variables were largest in man-
dibular retrognathism, followed by the
normal mandible and mandibular prog-
nathism. From these results, it is suggested
that the PAS–UP was affected by the
length and angle of the soft palate, but
not by maxillary size and position.
The results indicate that the PAS–UP
and PAS–TP (PAS at the base of the
tongue) are more narrow and collapsible
in mandibular retrognathism than in sub-
jects with a normal mandible. Many
patients with retrognathism show contact
Evaluation of the pharyngeal airway space
between the soft palate and tongue
because of downward displacement of
the tongue. In supine position the tongue
is situated ventrally to the soft palate and
can push the soft palate dorsally, changing
the PAS7. During sleep, airway narrowing
may occur as a result of posterior displa-
cement of the soft palate16. Mandibular
retrognathism may be the most important
risk factor for upper airway obstruction.
References
1. Achilleos A, Krogstad O, Lyberg T.
Surgical mandibular advancement and
changes in uvuloglossopharyngeal mor-
phology and head posture: a short- and
long-term cephalometric study in males.
Eur J Orthod 2000: 22: 367–381.
2. Achilleos A, Krogstad O, Lyberg T.
Surgical mandibular setback and changes
in uvuloglossopharyngeal morphology
and head posture: a short- and long-term
cephalometric study in males. Eur J
Orthod 2000: 22: 383–394.
3. Conway WA, Bower GC, Barnes ME.
Hypersomnolence and intermittent
upperairway obstruction. Occurrence
caused by micrognathia. J Am Med Assoc
1977: 237: 2740–2742.
4. Hellsing E. Changes in the pharyngeal
airway in relation to extension of the
head. Eur J Orthod 1989: 11: 359–365.
5. Hui DSC, Ko FWS, Chu ASY, Fok JPC,
Chan MCH, Li TST, Choy DKL, Lai
CKW, Ahuja A, Ching ASC. Cephalo-
metric assessment of craniofacial mor-
phology in Chinese patients with
obstructive sleep apnoea. Respir Med
2003: 97: 640–646.
6. Imes NK, Orr WC, Smith RO, Rogers
RM. Retrognathia and sleep apnea. A life-
threatening condition masquerading as
narcolepsy. J Am Med Assoc 1977:
237: 1596–1597.
7. Isono S, Remmers JE, Tanaka A, Sho
Y, Sato J, Nishino T. Anatomy of phar-
ynx in patients with obstructive sleep
apnea and in normal subjects. J Appl
Physiol 1997: 82: 1319–1326.
8. Li KK, Powell NB, Kushida C, Riley
RW, Adornato B, Guilleminault C.
A comparison of Asian and white patients
with obstructive sleep apnea syndrome.
Laryngoscope 1999: 109: 1937–1940.
9. Liu Y, Lowe AA, Zeng X, Fu M, Fleet-
ham JA. Cephalometric comparison
between Chinese and Caucasian patients
231
with obstructive sleep apnea. Am J
Orthod Dentofacial Orthop 2000: 117:
479–485.
10. Muto T, Takeda S, Kanazawa M,
Yamazaki A, Fujiwara Y, Mizoguchi
I. The effect of head posture on the phar-
yngeal airway space (PAS). Int J Oral
Maxillofac Surg 2002: 31: 579–583.
11. Muto T, Yamazaki A, Takeda S,
Kawakami J, Tsuji Y, Shibata T,
Mizoguchi I. Relationship between the
pharyngeal airway space and craniofacial
morphology, taking into account head
posture. Int J Oral Maxillofac Surg
2006: 35: 132–136.
12. Schwab RJ, Gupta KB, Gefter WB,
Metzger LJ, Hoffman EA, Pack AI.
Upper airway and soft tissue anatomy in
normal subjects and patients with sleep-
disordered breathing. Am J Respir Crit
Care Med 1995: 152: 1673–1689.
13. Schwab RJ, Pasirstein M, Pierson R,
Mackley A, Hachadoorian R, Arens
R, Maislin G, Pack AI. Identification of
upper airway anatomic risk factors for
obstructive sleep apnea with volumetric
magnetic resonance imaging. Am J
Respir Crit Care Med 2003: 168: 522–
530.
14. Solow B, Skov S, Ovesen J, Norup P,
Wildschiodtz G. Airway dimension and
head posture in obstructive sleep apnoea.
Eur J Orthod 1996: 18: 571–579.
15. Trenouth MJ, Timms DJ. Relationship
of the functional oropharynx to craniofa-
cial morphology. Angle Orthod 1999: 69:
419–423.
16. Trudo FJ, Gefter WB, Welch KC,
Gupta KB, Maislin G, Schwab RJ.
State-related changes in upper airway
caliber and surrounding soft-tissue struc-
tures in normal subjects. Am J Respir Crit
Care Med 1998: 158: 1259–1270.
17. Yu X, Fujimoto K, Urushibata K,
Matsuzawa Y, Kubo K. Cephalometric
analysis in obese and nonobese patients
with obstructive sleep apnea syndrome.
Chest 2003: 124: 212–218.
Address:
Toshitaka Muto
Department of Oral & Maxillofacial Surgery
School of Dentistry
Health Sciences University of Hokkaido
Ishikari-Tobetsu 1757
Hokkaido 061-0293
Japan
Fax: +81 1332 3 1429.
E-mail: muto@hoku-iryo-u.ac.jp