Content uploaded by Jen Walsh
Author content
All content in this area was uploaded by Jen Walsh on Aug 27, 2016
Content may be subject to copyright.
SLEEP, Vol. 31, No. 11, 2008 1543
OBSTRUCTIVE SLEEP APNEA (OSA) IS EXACERBAT-
ED BY THE SUPINE POSTURE IN THE MAJORITY OF
PATIENTS AND APPROXIMATELY 60% OF PATIENTS
have positional sleep apnea, dened as a supine apnea hypo-
pnea index (AHI) twice that observed when in the lateral re-
cumbent posture.1,2 Apnea severity (apnea duration, minimum
oxygen desaturation, arousal length and frequency) is increased
when supine.3 The optimal level of continuous positive airway
pressure,4 and the critical closing pressure, an objective mea-
sure of airway collapsibility, are higher when a subject is su-
pine than when in the lateral recumbent posture.5,6 Similarly,
the pressure required to reestablish airow is higher in the su-
pine than in the lateral recumbent posture.7 In some individuals,
merely avoiding the supine posture during sleep is sufcient to
resolve sleep apnea.1
The mechanism responsible for the worsening of sleep dis-
ordered breathing in the supine posture is not clear but most
likely relates to the effect of gravity on upper airway size or
shape. Gravitational effects could act directly on the upper
airway by displacing anterior pharyngeal structures and the
pharynx,6 or indirectly by displacing the abdominal contents
into the thorax and decreasing lung volume,8 and thereby de-
creasing the tension within the walls of the upper airway9 and
increasing its susceptibility to collapse. In the lateral recum-
bent posture, these compressive gravitational effects are re-
duced.
It is commonly thought that these effects result in a smaller
pharyngeal airway in the supine than in the lateral recumbent
posture, making it more vulnerable to collapse.10 However, re-
ports are inconsistent in this regard, with some studies reporting
the pharynx to be smaller in the supine than in the lateral re-
cumbent posture6,11,12 and others reporting a similar pharyngeal
size in the 2 postures.13-15 It is possible that airway shape may
also contribute to its propensity to collapse, as several studies
have suggested that orientation of the elliptically shaped up-
per airway differs between individuals with and without OSA.
Specically, they suggest that in individuals with OSA, the long
axis of the ellipse is oriented anteroposteriorally, making the
lateral pharyngeal walls more susceptible to collapse, whereas
in subjects without OSA, the long axis of the ellipse is oriented
transversely.16-19 However, this observation has not been consis-
tent; numerous other studies report airway shape to be similar
in apneics and non-apneic controls.14,20-26 The effect of posture
on pharyngeal shape is unknown.
The aim of the current study was to address these questions
by measuring upper airway shape and size, and the effect on
them of change in body position, in awake individuals with and
without OSA. We used anatomical optical coherence tomog-
raphy (aOCT), a quantitative imaging technique particularly
suited to repeated measurements in the same individual.27-29
POSTURNAL ASSOCIATION WITH SLEEP APNEA
Effect of Body Posture on Pharyngeal Shape and Size in Adults With and
Without Obstructive Sleep Apnea
Jennifer H. Walsh, PhD1; Matthew S. Leigh, BTech2; Alexandre Paduch, MSc2; Kathleen J. Maddison, BSc (Hons)1; Julian J. Armstrong, PhD2;
David D. Sampson, PhD2; David R. Hillman, MD1; Peter R. Eastwood, PhD1,3
1West Australian Sleep Disorders Research Institute, Sir Charles Gairdner Hospital, Nedlands, Western Australia; 2Optical & Biomedical
Engineering Laboratory, School of Electrical, Electronic and Computer Engineering and 3School of Anatomy and Human Biology, University of
Western Australia, Crawley, Western Austrailia
Study Objectives: In patients with obstructive sleep apnea (OSA), the
severity and frequency of respiratory events is increased in the supine
body posture compared with the lateral recumbent posture. The mecha-
nism responsible is not clear but may relate to the effect of posture on
upper airway shape and size. This study compared the effect of body
posture on upper airway shape and size in individuals with OSA with
control subjects matched for age, BMI, and gender.
Participants: 11 males with OSA and 11 age- and BMI-matched male
control subjects.
Results: Anatomical optical coherence tomography was used to scan
the upper airway of all subjects while awake and breathing quietly, ini-
tially when supine, and then in the lateral recumbent posture. A stan-
dard head, neck, and tongue position was maintained during scanning.
Airway cross-sectional area (CSA) and anteroposterior (A-P) and lat-
eral diameters were obtained in the oropharyngeal and velopharyngeal
regions in both postures. A-P to lateral diameter ratios provided an in-
dex of regional airway shape. In equivalent postures, the ratio of A-P
to lateral diameter in the velopharynx was similar in OSA and control
subjects. In both groups, this ratio was signicantly less for the supine
than for the lateral recumbent posture. CSA was smaller in OSA sub-
jects than in controls but was unaffected by posture.
Conclusions: The upper airway changes from a more transversely
oriented elliptical shape when supine to a more circular shape when
in the lateral recumbent posture but without altering CSA. Increased
circularity decreases propensity to tube collapse and may account for
the postural dependency of OSA.
Keywords: Pharyngeal anatomy, pharyngeal size, pharyngeal shape,
body posture, anatomical optical coherence tomography
Citation: Walsh JH; Leigh MS; Paduch A; Maddison KJ; Armstrong JJ;
Sampson DD; Hillman DR; Eastwood PR. Effect of body posture on
pharyngeal shape and size in adults with and without obstructive sleep
apnea. SLEEP 2008;31(11):1543-1549.
Submitted for publication January, 2008
Accepted for publication June, 2008
Address correspondence to: Dr Jennifer Walsh, Internal Mail Box 201,
Queen Elizabeth Medical Centre, Hospital Avenue, Nedlands, WA 6009;
Tel: +61 8 9346 2888; Fax: +61 8 9346 2034; E-mail: Jennifer.Walsh@
health.wa.gov.au
Effect of Body Posture on Airway Shape and Size—Walsh et al
SLEEP, Vol. 31, No. 11, 2008 1544
METHODS
Subjects
Eleven male volunteers with a BMI < 30 kg/m2 were recruit-
ed from patients with recently diagnosed OSA (AHI > 10/h) on
a laboratory-based polysomnogram.30,31 They were not selected
on the basis of presence or absence of positional OSA (dened
as supine AHI > 2 times lateral AHI and a total AHI of > 12.5/h,
having slept ≥ 30 min in each posture).2,32 They had not pre-
viously received treatment for OSA, including upper airway
surgery, and were otherwise healthy. Eleven healthy BMI- and
age-matched male control subjects without a history of habitual
snoring were recruited from local service clubs. OSA was ex-
cluded (AHI < 10/h) by a full night of laboratory-based poly-
somnography. The Human Research Ethics Committee of Sir
Charles Gairdner Hospital approved the project, and informed
written consent was obtained from all participants.
Protocol
Measurements of velopharyngeal and oropharyngeal shape
and size were obtained in each subject using anatomical opti-
cal coherence tomography (aOCT).27-29 Briey, aOCT requires
a sealed, transparent catheter (3.0 mm outside diameter) to be
inserted via the nares to mid-esophageal level. An optical probe
is moved systematically within the catheter, which is xed in po-
sition. The distance between the head of the optical probe and the
air-tissue interface of the airway wall is determined from reect-
ed light, using a low-coherence optical interferometer. A software
program controls the longitudinal translation and rotation (1.25
Hz) of the probe head, enabling collection of quantitative cross-
sectional images at regions of interest within the pharynx.
All aOCT scans were performed while the subject was awake;
initially supine, then repeated in the lateral recumbent posture.
Because head and body position have been shown to inu-
ence airway size,12 measurements were obtained with the head
and neck in a controlled neutral posture. Specically, when su-
pine, the head was supported with a Shea headrest (Gyrus ENT,
Memphis, TN, USA) and a goniometer was used to position the
Frankfort plane (line from infraorbital rim to tragus of the ear)
perpendicular to the bed. When lateral recumbent, the body was
perpendicular to the axis of the bed and the head supported with
pillows and foam pads to eliminate rotation or lateral exion/ex-
tension of the head and/or neck. A goniometer was used to align
the Frankfort plane perpendicular to the long axis of the body.
The subject was instructed to breathe quietly, not speak, and to
maintain a constant head and tongue position during all scans.
Rib cage and abdominal motion were continuously moni-
tored at 1000Hz (Powerlab model 16s; ADInstruments, Syd-
ney, NSW, Australia) by respiratory inductance pneumography
(Respitrace, Ambulatory Monitoring, Ardsley, NY, USA).
Airway Imaging
A “pullback” scan was performed in each subject, in each
posture, by systematically retracting the aOCT probe from the
upper esophagus to the nasal cavity at a constant speed (0.2
mm/sec). Each pullback scan took between 9 and 12 min, dur-
ing which time approximately 900 images were obtained. Each
image displayed airway cross-sectional dimensions for the pre-
vious 0.8 sec. Images were time-synchronized with the summed
pneumography signal and reconstructed to provide a video with
each frame providing a single quantitative cross-sectional im-
age of the pharynx.
Two regions of interest were dened from the reconstructed
video: the oropharynx (tip of epiglottis to base of uvula); and
the velopharynx (distal portion of the nasopharynx immediately
proximal to the base of the uvula). The precise locations of the
selected images within each region were determined a priori ac-
cording to the following: oropharyngeal cross-sectional images
were obtained from the mid-oropharynx or, where the uvula
was visible in the mid-oropharynx, just distal to the tip of the
uvula; velopharyngeal cross-sectional images were obtained
approximately 7 mm craniad to the base of the uvula.
Analysis
Oropharyngeal and velopharyngeal images were selected
by the same experienced investigator at the point of maximum
and minimum cross-sectional area (CSA) during multiple suc-
cessive respiratory cycles. In instances where images from 3
respiratory cycles were not available for analysis, 2 successive
cycles or 1 cycle was used if, by inspection of the video, they
were judged to be representative of that region. In instances
where a complete airway prole was not visible, images were
either excluded from analysis or, if at least 75% of the prole
was visible, a straight line connected the visible portions of the
airway. In cases where images for successive respiratory cycles
at a given location were analyzed, each was performed inde-
pendently and mean values used for statistical analyses.
No assumptions were made as to the relationship between
maximum and minimum CSA and phase of respiration; how-
ever, for all measurements, the phase of respiration in which
maximum and minimum CSA occurred was documented.
Analyses of aOCT images were performed using ImageJ soft-
ware (National Institutes of Health, Bethesda, MD). For each
image, the mucosa-lumen interface was manually traced by the
same experienced investigator and airway CSA calculated. A-P
diameter was calculated at the widest point in the parasagittal
plane and lateral diameter was measured at the widest point in
the coronal plane, perpendicular to the A-P diameter.14,17 The in-
traclass correlation coefcient for repeat measurements of air-
way CSA and diameters was 0.99 (P < 0.0001) and is reported
in more detail elsewhere.28
Statistical Analysis
Student unpaired t-tests were used to compare anthropomet-
ric and polysomnographic measurements between control and
OSA groups. Two-way repeated-measures ANOVA (SigmaStat,
San Jose, CA, USA) was used to compare differences in region-
al pharyngeal dimensions and locations, at both maximum and
minimum CSA, between OSA and control groups in the supine
and lateral recumbent postures. A Holm-Sidak test was applied
for all post hoc comparisons. Unless stated, all data are reported
as mean ± SD. Signicance was assumed at P < 0.05.
Effect of Body Posture on Airway Shape and Size—Walsh et al
SLEEP, Vol. 31, No. 11, 2008 1545
RESULTS
Anthropometric and polysomnographic measurements in the
11 OSA and 11 control subjects are presented in Table 1. The 2
groups were well-matched for age and BMI. AHI ranged from
15.0 to 76.8 events/h in the OSA group. In all OSA subjects,
AHI was less in the lateral posture than in the supine. Seven
OSA subjects met the criteria for positional sleep apnea.2,32 In
all but one control subject, AHI was less in the lateral posture
than in the supine. Although total AHI was < 10/h in the con-
trol subjects, 7 had a supine AHI greater than twice that in the
lateral posture.
Image Location and Analysis
In some individuals, at some sites, it was not possible to vi-
sualize the complete circumference of the airway. In the 22 sub-
jects examined in the present study, axial images with ≥ 75%
of the airway circumference (including lateral extents) visible
were obtained in 91% and 95% of scans performed in the su-
pine and lateral postures, respectively. Of the images analyzed,
56% were complete proles, with the remainder requiring mod-
est straight-line extrapolation to connect the visible portions of
the airway prole with similar frequency of extrapolation in
subjects with and without OSA.
The locations of velopharyngeal and oropharyngeal images
relative to anatomical landmarks were similar in the OSA and
control subjects and in the lateral and supine posture (ANOVA,
P = 0.27). For example, velopharyngeal images in the supine
posture were obtained 6.8 ± 1.4 and 6.4 ± 3.4 mm craniad to the
base of the uvula in the OSA and control groups, respectively.
Oropharyngeal images were 22.0 ± 4.8 and 17.2 ± 3.5 mm cau-
dad to the base of the uvula in the OSA and control groups,
respectively.
Effect of Posture on Airway Size and Dimensions
The effects of posture on wakeful velopharyngeal and
oropharyngeal shape and size are shown in the images in Figure
1. The tissue-air boundaries in each gure appear fuzzy because
aOCT also detects subsurface reections, but the interface be-
tween the airway and airway wall is sharp, allowing accurate
surface location. Most notable features from these images are:
(i) the smaller velopharynx in the individual with OSA than the
matched control subject; and (ii) the marked effect of body pos-
ture on airway shape in both individuals in the velopharynx and
oropharynx. These changes were representative of the group
data, which are presented below.
Cross-Sectional Area
OSA vs Control
Velopharyngeal maximum and minimum CSA were signi-
cantly less in OSA subjects than in control subjects in supine
(P < 0.05 and P < 0.01, respectively) but not lateral recumbent
posture (P = 0.12 and 0.11, respectively) (Figure 2). Oropha-
ryngeal maximum and minimum CSA were similar in OSA and
control groups for each posture (P > 0.2 for all comparisons).
Table 1—Anthropometric and Polysomnographic Measurements
in OSA and Healthy Control Subjects
OSA Control
(n = 11) (n = 11)
Age (y) 56 ± 13 59 ± 9
BMI (kg/m2) 27.9 ± 1.0 25.9 ± 1.7
AHI (events/h) 39.6 ± 19.1* 3.3 ± 2.5
Supine AHI (events/h) 53.7 ± 22.2* 9.2 ± 9.2
NREM (events/h) 55.1 ± 23.4* 9.2 ± 9.2
REM (events/h) 52.9 ± 21.6* 4.1 ± 10.3
Lateral Recumbent AHI (events/h) 23.8 ± 21.6*† 1.0 ± 1.0†
NREM (events/h) 25.8 ± 22.0*† 0.5 ± 0.6†
REM (events/h) 38.0 ± 13.9*† 2.4 ± 2.8†
Values are mean ± SD. * P < 0.05 vs control; † P < 0.01 vs equiva-
lent when supine; BMI, body mass index; AHI, apnea hypopnea
index.
Figure 1—Representative aOCT images of the velopharynx and
oropharynx from one healthy control subject and one OSA subject
in the supine and lateral recumbent postures. All scans were ob-
tained when the airway was at its minimum cross-sectional area
during the respiratory cycle. The inner and outer walls of the imag-
ing catheter are visible within each airway. All images have been ro-
tated to align the anterior pharyngeal wall with the top of the page.
Figure 2—Maximum (closed symbols) and minimum (open sym-
bols) cross-sectional area (CSA) in the oropharynx (left panel)
and velopharynx (right panel) in healthy control (circles) and
OSA subjects (triangles) in the supine and lateral recumbent pos-
tures. n = 11 per group; mean ± SE; * signicantly different from
control group; P < 0.05.
Effect of Body Posture on Airway Shape and Size—Walsh et al
SLEEP, Vol. 31, No. 11, 2008 1546
eter in either group when at maximum CSA (P > 0.2 for both)
(Figure 3).
Shape
The ratio of A-P to lateral diameter provides an index of the
circularity of the airway, with a ratio of 1.0 representing a circle,
a ratio <1.0 representing an ellipse with its long axis oriented
laterally, and a ratio >1.0 representing an ellipse with its long
axis oriented in the A-P dimension.
OSA vs Control
The shape of the airway changed minimally with respiration,
as seen by the lack of change in this ratio between minimum
and maximum CSA (Figure 4). The ratio at maximum and mini-
mum CSA was similar in the OSA and control groups in both
the supine and lateral recumbent postures, indicating a similarly
shaped airway in both groups in both body postures.
Supine vs Lateral
Except for the oropharynx of the OSA group at maximum
CSA, this ratio was less than 1.0 and increased with change
from the supine to lateral recumbent posture in both groups and
in both pharyngeal regions, both at minimum and maximum
CSA (P < 0.05 for all comparisons) (Figure 4). In the orophar-
ynx of the OSA group at maximum CSA, the ratio increased,
but this change did not reach statistical signicance (P = 0.10).
DISCUSSION
The mechanism underlying aggravation of sleep disordered
breathing in the supine posture remains to be dened, although
gravity-related changes in pharyngeal dimensions have been
Supine vs Lateral
Moving from the supine to lateral recumbent posture had
no effect on velopharyngeal CSA (P > 0.2 for all comparisons)
or oropharyngeal CSA (P > 0.3 for all comparisons) in either
group (Figure 2). Maximum velopharyngeal CSA occurred dur-
ing expiration in 72% of control subjects and in 67% of OSA
subjects, whereas maximum oropharyngeal CSA occurred dur-
ing expiration in 62% of controls and in 72% of OSA subjects
(pooled data from both postures).
Lateral Diameter
OSA vs Control
Velopharyngeal lateral diameter at minimum velopharyn-
geal CSA was less in OSA subjects than controls in supine (P <
0.05), but not lateral recumbent posture (P > 0.4; Figure 3). Lat-
eral velopharyngeal diameter at maximum CSA was similar in
both groups in both postures. Oropharyngeal lateral diameter
was similar in OSA and in control subjects in both postures
when measured at maximum and minimum CSA (Figure 3).
Supine vs Lateral
Moving from the supine to the lateral recumbent posture (i)
decreased velopharyngeal lateral diameter at both maximum
and minimum velopharyngeal CSA in control subjects (P < 0.01
for both) but not in OSA subjects (P > 0.2 for both); and (ii)
decreased oropharyngeal lateral diameter in control and OSA
groups at both minimum and maximum CSA (P < 0.04 for all
comparisons) (Figure 3).
Anteroposterior (A-P) Diameter
OSA vs Control
There was a signicant group and posture effect on velopha-
ryngeal A-P diameter at both maximum and minimum CSA.
However, because of a large increase in A-P diameter in one
control subject when moving from supine to lateral (10.7 to
19.0 mm at maximum CSA) and the associated increase in vari-
ability, post hoc analyses did not identify the differences. When
this subject was excluded from the analysis, it was revealed that
velopharyngeal A-P diameter was smaller in OSA than control
subjects in supine, but not in lateral recumbent posture when
measured at maximum and minimum CSA (P < 0.05 for both)
(Figure 3). Oropharyngeal A-P diameter was similar in OSA
and control subjects in both postures when measured at maxi-
mum and minimum CSA.
Supine vs Lateral
Moving from the supine to lateral recumbent posture (i)
increased the A-P diameter at both maximum and minimum
CSA in the OSA and control subjects (P < 0.05 for all com-
parisons); and (ii) increased oropharyngeal A-P diameter in
both the OSA and control groups (P < 0.05 for both) when
at minimum CSA, but did not alter oropharyngeal A-P diam-
Figure 3—Lateral (upper) and anteroposterior (A-P) (lower) diam-
eter at maximum (closed symbols) and minimum (open symbols)
cross-sectional area (CSA) in the oropharynx (left panel) and ve-
lopharynx (right panel) in healthy control (circles) and OSA sub-
jects (triangles) in the supine and lateral recumbent postures. n = 11
per group; mean ± SE; * signicantly different from control group;
P < 0.05; †signicantly different from supine posture; P < 0.05.
Effect of Body Posture on Airway Shape and Size—Walsh et al
SLEEP, Vol. 31, No. 11, 2008 1547
nologies.33 Failure to obtain orthogonal images would introduce
substantial error into assessments of size and shape.
Measurements obtained in the present study indicate a lack
of posture inuence on upper airway caliber. This nding is
also in agreement with some,13-15 but not all,6,11,12 previous im-
aging studies. Two studies utilizing acoustic reection found
no difference in total airway volume or area at the level of the
oropharyngeal junction between supine and lateral recumbent
postures in non-snoring, snoring, and sleep apneic subjects dur-
ing wakefulness.13,15 Similarly, awake CT studies in positional
and non-positional OSA patients showed no difference in mini-
mum or mean CSA of the entire airway between the supine and
lateral recumbent postures.14 In contrast, Isono et al. used vid-
eoendoscopy to show that CSA was larger in the supine than
lateral recumbent posture in anesthetized and paralyzed OSA
patients at a range of static airway pressures.6 Magnetic reso-
nance imaging (MRI) studies in healthy sedated children11 and
awake young adults12 have also shown decreased retroglossal
airway volumes and CSA in supine than lateral recumbent pos-
ture. The reasons for the discrepancies between studies are not
entirely clear, but may relate to differences in gender or age of
study participants, conscious state, disease severity, or use of
image gating with phase of respiration.
The ndings of the present study provide several insights
into the positional dependence of OSA. The combined effects
of posture, gravity and upper airway anatomy can be consid-
ered in terms of the “bony enclosure’”model described by Isono
et al.,6 which suggests that non-uniform distribution of soft tis-
sue around the pharyngeal airway may result in regional differ-
ences in the extraluminal forces acting on the airway (Figure 6).
Based upon an airway with similar lateral and A-P dimensions
in both postures (i.e., a circular airway), Isono’s model propos-
thought to play a major role. Although consideration has been
given to posture-related changes in airway caliber, the effect of
posture on pharyngeal shape has not previously been examined.
The present study utilized a novel imaging technique suit-
able for repetitive, quantitative imaging of the upper airway,
aOCT, to demonstrate that moving from the supine to the
lateral recumbent posture alters the shape, but not the size of
the velopharyngeal and oropharyngeal airways in individuals
with and without OSA. Specically, the airway changes from a
transversely oriented elliptical shape when supine to a more cir-
cular shape when in the lateral recumbent posture. This change
in shape may be an important factor underlying the decreased
propensity of the upper airway to collapse when in a lateral re-
cumbent posture. Laplace’s Law states that at equilibrium, the
transmural pressure across a concave surface is directly propor-
tional to wall tension and inversely proportional to its radius
of curvature. It follows that the transmural pressure gradient
required to collapse the airway varies inversely with its radius
of curvature. Hence, as the transverse elliptical airway assumes
a more circular shape with change to the lateral posture, its pro-
pensity to collapse decreases as a function of the reduction in
radius of curvature of its anterior and posterior walls. This pro-
pensity can be expressed dimensionally as proportional to the
ratio of the lengths of the major and minor axes of the elliptical
cross-section.
Our nding that, when supine, individuals with OSA and
BMI- and age-matched control subjects have a similarly shaped
airway (an ellipse with its long axis oriented laterally) in both
the velopharyngeal and oropharyngeal regions contrasts with
some,16-19 but not all14,20-26 previous reports of pharyngeal shape.
The different ndings could be attributable to a number of fac-
tors, including differences in location and orientation of images,
variable head or neck exion/extension,16 averaging of images
over several breaths16 versus breath holding,18 or to the presence
of adenotonsillar hypertrophy on the lateral airway walls.19 Our
study addresses these potential confounding factors through
careful matching of age and BMI in our all-male subjects;
control of head posture; and the use of an imaging technique
(aOCT) that produces quantitative, breath-by-breath images
orthogonal to the airway wall. The accuracy of the orthogonal
plane alignment is a central issue that has proved difcult to
control with older computed tomography (CT) scanning tech-
Figure 4—Ratio of anteroposterior (A-P):lateral diameter at maxi-
mum (closed symbols) and minimum (open symbols) cross-section-
al area (CSA) in the oropharynx (left panel) and velopharynx (right
panel) in healthy control (circles) and OSA subjects (triangles) in
the supine and lateral recumbent postures. n = 11 per group; mean ±
SE; † signicantly different from supine posture; P < 0.05
Figure 5—Schematic representation of the compartmental tissue
arrangement surrounding the pharyngeal airway when in the su-
pine and lateral recumbent postures. BE, bony enclosure; PA, pha-
ryngeal airway; A, anterior soft tissue mass; P, posterior soft tissue
mass; L, lateral soft tissue mass. Note (i) the increased circularity
of the airway in the lateral recumbent posture, (ii) greater radius
of curvature of the anterior and posterior airway walls in the su-
pine posture, and (iii) the relatively greater mass on the anterior
pharyngeal airway when supine (shaded region, A) than the mass
on the lateral pharyngeal airway when lateral recumbent (shaded
region, L). Modied from Isono’s bony enclosure model.6
Effect of Body Posture on Airway Shape and Size—Walsh et al
SLEEP, Vol. 31, No. 11, 2008 1548
(aOCT) anatomical optical coherence tomography
CT computed tomography
MRI magnetic resonance imaging
ACKNOWLEDGMENTS
Sources of Funding: The study has been supported by the
National Health and Medical Research Council Australia (Proj-
ect Grant No. 403953; Development Grant No. 303319). PRE
was supported by a NHMRC Senior Research Fellowship (No.
513704).
DISCLOSURE STATEMENT
This was not an industry supported study. Dr. Hillman has
received research support from ResMed and has consulted for
ResMed and Inspiration Medical. Dr. Eastwood has consulted
for Inspiration Medical. The other authors have indicated no
nancial conicts of interest.
REFERENCES
1. Cartwright RD. Effect of sleep position on sleep apnea severity.
Sleep 1984;7:110-4.
2. Oksenberg A, Silverberg DS, Arons E, Radwan H. Positional vs
nonpositional obstructive sleep apnea patients: anthropomorphic,
nocturnal polysomnographic, and multiple sleep latency test data.
Chest 1997;112:629-39.
3. Oksenberg A, Khamaysi I, Silverberg DS, Tarasiuk A. Asso-
ciation of body position with severity of apneic events in pa-
tients with severe nonpositional obstructive sleep apnea. Chest
2000;118:1018-24.
4. Oksenberg A, Silverberg DS, Arons E, Radwan H. The sleep su-
pine position has a major effect on optimal nasal continuous posi-
tive airway pressure: relationship with rapid eye movements and
non-rapid eye movements sleep, body mass index, respiratory
disturbance index, and age. Chest 1999;116:1000-6.
5. Boudewyns A, Punjabi N, Van de Heyning PH, et al. Abbreviated
method for assessing upper airway function in obstructive sleep
apnea. Chest 2000;118:1031-41.
6. Isono S, Tanaka A, Nishino T. Lateral position decreases collaps-
ibility of the passive pharynx in patients with obstructive sleep
apnea. Anesthesiology 2002;97:780-5.
7. Neill AM, Angus SM, Sajkov D, McEvoy RD. Effects of sleep
posture on upper airway stability in patients with obstructive
sleep apnea. Am J Respir Crit Care Med 1997;155:199-204.
8. Fouke JM, Strohl KP. Effect of position and lung volume on up-
per airway geometry. J Appl Physiol 1987;63:375-80.
9. Kairaitis K, Byth K, Parikh R, Stavrinou R, Wheatley JR, Amis
TC. Tracheal traction effects on upper airway patency in rabbits:
the role of tissue pressure. Sleep 2007;30:179-86.
10. Oksenberg A, Silverberg DS. The effect of body posture on sleep-
related breathing disorders: facts and therapeutic implications.
Sleep Med Rev 1998;2:139-62.
11. Litman RS, Wake N, Chan LM, et al. Effect of lateral positioning
on upper airway size and morphology in sedated children. Anes-
thesiology 2005;103:484-8.
12. Ono T, Otsuka R, Kuroda T, Honda E, Sasaki T. Effects of head
and body position on two- and three-dimensional congurations
of the upper airway. J Dent Res 2000;79:1879-84.
13. Jan MA, Marshall I, Douglas NJ. Effect of posture on upper air-
way dimensions in normal human. Am J Respir Crit Care Med
1994;149:145-8.
es that the larger soft tissue mass in the anterior pharyngeal air-
way compartment exerts greater pressure on the anterior airway
when in the supine posture than the relatively smaller tissue
mass in the lateral compartment exerts on the lateral airway
wall when in the lateral recumbent posture. In other words, the
increased extraluminal tissue pressure on the anterior pharyn-
geal airway when supine increases its susceptibility to collapse
in this posture.
Our ndings suggest airway shape as an alternate or addition-
al inuence. The greater radius of curvature of the anterior and
posterior walls of the transverse elliptical airway in the supine
posture, relative to the circular airway in the lateral recumbent
posture, increases its propensity to collapse, as outlined earlier.
This may be particularly relevant when airway CSA is reduced,
as has been observed in OSA patients in the present study and
by others.17,19,22,24,34 A smaller CSA requires less change in air-
way caliber for collapse to occur and is also associated with
more negative intraluminal pressure for a given inspiratory
ow, which also increases the tendency to collapse.
Other factors may also contribute to the positional depen-
dency of OSA, such as posture-related changes in lung volume
or activity of pharyngeal muscles. Previous studies have dem-
onstrated that functional residual capacity is less in the supine
than in the lateral recumbent posture.35-37 A lower lung volume
could act to decrease pharyngeal patency and increase pharyn-
geal collapsibility through a decrease in longitudinal airway
tension.9,38-40 It is possible that such volume-related changes in
airway stability could occur without accompanying changes in
pharyngeal dimensions.41 Changes in upper airway muscle ac-
tivity may also contribute to posture-related changes in airway
shape. Several studies have demonstrated increased genioglos-
sus muscle activity when supine compared to the upright42-44 and
lateral recumbent postures.44,45 However, others have reported
decreased suprahyoid muscle activity in the supine compared
to lateral recumbent posture.46,47 Thus, a consistent relationship
between upper airway shape and size and muscle activity is yet
to be dened. It is important to note, however, that the ndings
of the present study were obtained in wakeful subjects. Thus,
conrmation of the role of changes in pharyngeal shape and
size on the postural worsening of OSA will require studies of
sleeping subjects, which represents a natural extension of the
present study.
In conclusion, aOCT-derived measurements when supine
indicate that healthy control subjects have a larger velopharyn-
geal airway than subjects with OSA, despite having a similarly
shaped airway: elliptical with the long axis in the lateral dimen-
sion. Change to the lateral recumbent posture makes the airway
more circular in both groups, but does not alter its CSA. This
change in shape provides a cogent explanation for the reduced
propensity for pharyngeal collapse in the lateral recumbent rel-
ative to supine posture.
ABBREVIATIONS
OSA obstructive sleep apnea
BMI body mass index
CSA cross sectional area
A-P anteroposterior
AHI apnea hypopnea index
Effect of Body Posture on Airway Shape and Size—Walsh et al
SLEEP, Vol. 31, No. 11, 2008 1549
30. Rechtschaffen A, Kales A. A manual of standardized terminology,
technique and scoring system for sleep stages of human sleep:
National Institutes of Health, Washington DC; 1968.
31. American Academy of Sleep Medicine Task Force. Sleep-related
breathing disorders in adults: recommendations for syndrome
denition and measurement techniques in clinical research. Sleep
1999;22:667-89.
32. Cartwright RD, Diaz F, Lloyd S. The effects of sleep posture and
sleep stage on apnea frequency. Sleep 1991;14:351-3.
33. Kumar V, Ludlow JB, Mol A, Cevidanes L. Comparison of con-
ventional and cone beam CT synthesized cephalograms. Den-
tomaxillofac Radiol 2007;36:263-9.
34. Haponik EF, Smith PL, Bohlman ME, Allen RP, Goldman SM,
Bleecker ER. Computerized tomography in obstructive sleep ap-
nea. Correlation of airway size with physiology during sleep and
wakefulness. Am Rev Respir Dis 1983;127:221-6.
35. Rehder K, Hatch DJ, Sessler AD, Fowler WS. The function of
each lung of anesthetized and paralysed man during mechanical
ventilation. Anesthesiology 1972;37:16-26.
36. Marini JJ, Tyler ML, Hudson LD, Davis BS, Huseby JS. Inu-
ence of head-dependent positions on lung volume and oxygen
saturation in chronic air-ow obstruction. Am Rev Respir Dis
1984;129:101-5.
37. Washko GR, O’Donnell CR, Loring SH. Volume-related and
volume-independent effects of posture on esophageal and
transpulmonary pressures in healthy subjects. J Appl Physiol
2006;100:753-8.
38. Hoffstein V, Zamel N, Phillipson EA. Lung volume dependence
of pharyngeal cross-sectional area in patients with obstructive
sleep apnea. Am Rev Respir Dis 1984;130:175-8.
39. Van de Graaff WB. Thoracic inuence on upper airway patency. J
Appl Physiol 1988;65:2124-31.
40. Thut DC, Schwartz AR, Roach D, Wise RA, Permutt S, Smith PL.
Tracheal and neck position inuence upper airway airow dynam-
ics by altering airway length. J Appl Physiol 1993;75:2084-90.
41. Tagaito Y, Isono S, Remmers JE, Tanaka A, Nishino T. Lung
volume and collapsibility of the passive pharynx in patients with
sleep-disordered breathing. J Appl Physiol 2007;103:1379-85.
42. Sauerland EK, Mitchell SP. Electromyographic activity of intrin-
sic and extrinsic muscles of the human tongue. Tex Rep Biol Med
1975;33:444-55.
43. Douglas NJ, Jan MA, Yildirim N, Warren PM, Drummond GB.
Effect of posture and breathing route on genioglossal electromyo-
gram activity in normal subjects and in patients with the sleep ap-
nea/hypopnea syndrome. Am Rev Respir Dis 1993;148:1341-5.
44. Otsuka R, Ono T, Ishiwata Y, Kuroda T. Respiratory-related ge-
nioglossus electromyographic activity in response to head rota-
tion and changes in body position. Angle Orthod 2000;70:63-9.
45. Malhotra A, Trinder J, Fogel R, et al. Postural effects on pharyn-
geal protective reex mechanisms. Sleep 2004;27:1105-12.
46. Ormeno G, Miralles R, Loyola R, et al. Body position effects on
EMG activity of the temporal and suprahyoid muscles in healthy
subjects and in patients with myogenic cranio-cervical-mandibu-
lar dysfunction. Cranio 1999;17:132-42.
47. De Mayo T, Miralles R, Barrero D, et al. Breathing type and body
position effects on sternocleidomastoid and suprahyoid EMG ac-
tivity. J Oral Rehabil 2005;32:487-94.
14. Pevernagie DA, Stanson AW, Sheedy PF 2nd, Daniels BK, Shep-
ard JW Jr. Effects of body position on the upper airway of pa-
tients with obstructive sleep apnea. Am J Respir Crit Care Med
1995;152:179-85.
15. Martin SE, Marshall I, Douglas NJ. The effect of posture on air-
way caliber with the sleep-apnea/hypopnea syndrome. Am J Re-
spir Crit Care Med 1995;152:721-4.
16. Rodenstein DO, Dooms G, Thomas Y, et al. Pharyngeal shape
and dimensions in healthy subjects, snorers, and patients with ob-
structive sleep apnoea. Thorax 1990;45:722-7.
17. Schwab RJ, Gefter WB, Hoffman EA, Gupta KB, Pack AI. Dy-
namic upper airway imaging during awake respiration in normal
subjects and patients with sleep disordered breathing. Am Rev
Respir Dis 1993;148:1385-400.
18. Fogel RB, Malhotra A, Dalagiorgou G, et al. Anatomic and physi-
ologic predictors of apnea severity in morbidly obese subjects.
Sleep 2003;26:150-5.
19. Arens R, Sin S, McDonough JM, et al. Changes in upper airway
size during tidal breathing in children with obstructive sleep ap-
nea syndrome. Am J Respir Crit Care Med 2005;171:1298-304.
20. Kuna ST, Bedi DG, Ryckman C. Effect of nasal airway positive
pressure on upper airway size and conguration. Am Rev Respir
Dis 1988;138:969-75.
21. Stanford W, Galvin J, Rooholamini M. Effects of awake tidal
breathing, swallowing, nasal breathing, oral breathing and the
Muller and Valsalva maneuvers on the dimensions of the upper
airway. Evaluation by ultrafast computerized tomography. Chest
1988;94:149-54.
22. Schwab RJ, Gupta KB, Gefter WB, Metzger LJ, Hoffman EA,
Pack AI. Upper airway and soft tissue anatomy in normal sub-
jects and patients with sleep-disordered breathing. Signicance
of the lateral pharyngeal walls. Am J Respir Crit Care Med
1995;152:1673-89.
23. Ryan CF, Love LL. Mechanical properties of the velopharynx
in obese patients with obstructive sleep apnea. Am J Respir Crit
Care Med 1996;154:806-12.
24. Ciscar MA, Juan G, Martinez V, et al. Magnetic resonance im-
aging of the pharynx in OSA patients and healthy subjects. Eur
Respir J 2001;17:79-86.
25. Schwab RJ, Pasirstein M, Pierson R, et al. Identication of up-
per airway anatomic risk factors for obstructive sleep apnea with
volumetric magnetic resonance imaging. Am J Respir Crit Care
Med 2003;168:522-30.
26. Hsu PP, Tan BY, Chan YH, Tay HN, Lu PK, Blair RL. Clinical
predictors in obstructive sleep apnea patients with computer-as-
sisted quantitative videoendoscopic upper airway analysis. Lar-
yngoscope 2004;114:791-9.
27. Armstrong JJ, Leigh MS, Walton ID, et al. In vivo size and shape
measurement of the human upper airway using endoscopic long-
range optical coherence tomography. Opt Exp 2003;11:1817-26.
28. Armstrong JJ, Leigh MS, Sampson DD, Walsh JH, Hillman DR,
Eastwood PR. Quantitative upper airway imaging with anatom-
ic optical coherence tomography. Am J Respir Crit Care Med
2006;173:226-33.
29. Leigh MS, Armstrong JJ, Paduch A, et al. Anatomical optical co-
herence tomography for long-term, portable, quantitative endos-
copy. IEEE Trans Biomed Eng 2008;55:1438-46.
Effect of Body Posture on Airway Shape and Size—Walsh et al
