See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/265861932

Mechanics of Circular Breathing in Wind Musicians Using Cine Magnetic

Resonance Imaging Techniques

Article  in  The Laryngoscope · February 2015

DOI: 10.1002/lary.24928 · Source: PubMed

CITATIONS READS

0 63

4 authors, including:

Clifford Douglas Phillips Jonathan Dyke

Weill Cornell Medical College Weill Cornell Medical College
130 PUBLICATIONS   3,969 CITATIONS    145 PUBLICATIONS   3,384 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Measurement of Bone Perfusion View project

MRI and Spectroscopy in Essential Tremor View project

All content following this page was uploaded by Jonathan Dyke on 10 July 2018.

The user has requested enhancement of the downloaded file.

The Laryngoscope
C 2014 The American Laryngological,
V
Rhinological and Otological Society, Inc.

Mechanics of Circular Breathing in Wind Musicians Using Cine

Magnetic Resonance Imaging Techniques

Tiffany Peng, MD; C. Douglas Phillips, MD; Jonathan P. Dyke, PhD; Michael G. Stewart, MD, MPH

Objectives/Hypothesis: To investigate the mechanics of circular breathing using cine magnetic resonance imaging
(cMRI).
Study Design: Pilot study.
Methods: Eight musicians were asked to sustain a note by circular breathing while being studied inside a Siemens Trio
3T magnetic resonance imaging scanner. Subjects were imaged in the midsagittal plane using an innovative T1-weighted
cMRI. Our study population included six professionals and two experienced amateurs (two saxophonists, two trombonists,
and four trumpeters). Five predetermined oropharyngeal distances were measured frame by frame. Data were analyzed for
displacement and percentage change over time. Oral airway area was measured using automated bounding box techniques.
Results: All subjects were observed to complete the same series of steps characterized by 1) superior/posterior tongue
displacement, 2) inferior/anterior soft palate displacement, 3) anterior tongue/soft palate displacement, 4) superior/posterior
soft palate displacement, and 5) return of the tongue to baseline. Posterior oropharyngeal occlusion was demonstrated in
bounding box data. Relative occlusion time was significantly longer among experienced amateurs than professionals (P 5.01),
with a similar trend in absolute occlusion time (P 5.08). A trend toward a longer circular breathing cycle time was observed
among saxophonists and trumpeters (P 5.2), although the difference among trumpeters disappeared when amateurs were
excluded.
Conclusions: Circular breathing has been investigated as therapy for obstructive sleep apnea. This study defines the
mechanics of circular breathing for use in training—and potentially in therapy—if proven useful. We demonstrated similar
mechanics across subgroups, and increased efficiency among professionals. Our findings demonstrate the feasibility and reli-
ability of cMRI in the dynamic assessment of oropharyngeal motion.
Key Words: Circular breathing, cine magnetic resonance imaging, cine magnetic resonance, obstructive sleep apnea.
Level of Evidence: NA
Laryngoscope, 00:000–000, 2014

INTRODUCTION they are circular breathing, which allows them to play

Circular breathing is a musical performance tech-
nique in which wind instrumentalists play a seemingly
impossibly long note or note progression without taking
a breath. This is not because they have increased lung
capacity and are playing from a single breath. Rather,
Additional Supporting Information may be found in the online
version of this article.
From the Department of Otolaryngology–Head and Neck Surgery
(T.P., M.G.S.), and the Department of Radiology (C.D.P., J.P.D.), Weill Cornell
Medical College, New York, New York, U.S.A.
Editor’s Note: This Manuscript was accepted for publication
August 20, 2014.
Presented at the 47th Annual Meeting of the American Society of
Head and Neck Radiology, Milwaukee, Wisconsin, U.S.A., September 29,
2013.
Funding for this study was provided by the Music and Medicine
Initiative of Weill Cornell Medical College and the Departments of
Radiology and Otolaryngology of NewYork–Presbyterian Hospital/Weill
Cornell Medical Center.
The authors have no funding, financial relationships, or conflicts
of interest to disclose.
Send correspondence to Michael G. Stewart, MD, Professor and
Chairman, Otolaryngology Medical Director, Center for the Performing
Artist, Vice Dean, Weill Cornell Medical College Department of Otolar-
yngology–Head and Neck Surgery, NewYork–Presbyterian Hospital/Weill
Cornell Medical Center, 1305 York Avenue, 5th Floor, New York, NY
10021. E-mail: mgs2002@med.cornell.edu
DOI: 10.1002/lary.24928
their instrument through the mouth continuously while
occasionally breathing in through their nose to simulta-
neously refill the lungs. This is accomplished by expel-
ling air via cheek compression in the anterior
oropharynx during nasal inhalation.1
Circular breathing is employed when playing an
obscure instrument called the didgeridoo. There is some
evidence that didgeridoo practice may be therapeutic in
obstructive sleep apnea (OSA).2,3 Puhan et al. demon-
strated in a cohort of 25 adult patients with moderate
OSA that 3 months of didgeridoo practice led to signifi-
cant improvements in daytime sleepiness, apnea-
hypopnea index scores, and partner-reported sleep dis-
turbance.2 Although studies have investigated didgeri-
doo practice in medical therapy, and one study
visualized glottic activity during didgeridoo performance
using transnasal endoscopy, no evidence exists directly
characterizing the oropharyngeal mechanics of circular
breathing.4,5 Consequently, there are no data in the lit-
erature to demonstrate exactly how circular breathing is
accomplished, whether different circular breathers uti-
lize the same techniques, the impact of different instru-
ments on the technique, and whether amateurs perform
this technique similarly to trained professionals. To
assess circular breathing as therapy and begin to

Laryngoscope 00: Month 2014 Peng et al.: Circular Breathing Mechanics With Cine MRI
1
 TABLE I.
Demographics of Study Participants.
Professional Age, Years of Hours Playing
Subject Status Instrument yr Experience per Week
1 Professional Trumpet 26 14
2 Professional Trumpet 48 38
3 Professional Trumpet 60 48
4 Professional Trombone 25 15
5 Professional Saxophone 25 14
6 Professional Saxophone 31 21
7 Amateur Trumpet 25 15
8 Amateur Trombone 21 11
understand the physiologic basis for any incurred benefit
from its practice, further knowledge of circular breath-
ing is required.
Cine magnetic resonance imaging (cMRI) is a tech-
nique that has been used for the assessment of cardiac
and bowel motility.6–8 It has also been utilized in the study
of upper airway dynamics of the M€ uller maneuver in
healthy subjects.9 This imaging modality allows for real-
time analysis and high-quality visualization of oropharyn-
geal structures. Using an innovative, fast, temporally
resolved T1-weighted cMRI sequence, we sought to explore
the feasibility of cMRI for the assessment of oropharyngeal
motion, and then, if feasible, to utilize this technology to
characterize the mechanics of circular breathing in profes-
sional and experienced amateur wind musicians.
MATERIALS AND METHODS
Subject Recruitment
After obtaining institutional review (IRB) board approval,
eight volunteer subjects were recruited from the New York/New
Jersey area, primarily through The Juilliard School. Our study
population included six professionals and two experienced ama-
teurs (two saxophonists, two trombonists, and four trumpeters)
(Table I). The mean age of participants was 32.6 years (range,
21–60 years). All participants were male. Professionals played
their instruments for an average of 25 years (range, 14–48
years) and reported playing their instrument for an average of
36 hours per week (range, 27–46 hours) over the preceding 6
months. Experienced amateurs, referred to as “amateurs” in the
remainder of this article, played their instruments for an aver-
age of 13 years (range, 11–15 years) and reported playing their
instrument for an average of 6 hours per week (range, 1–11
hours) over the preceding six months.
Professionals were defined as full-time musicians whose
sole source of income was instrument performance. Amateurs
were defined as those who do not receive payment for instru-
ment performance. Musicians receiving part-time or occasional
pay were excluded from the study. Trumpeters, trombonists,
and saxophonists were studied because circular breathing is
commonly employed among these musicians, and because it
would be feasible for subjects to play these instrument mouth-
pieces while inside the magnetic resonance imaging (MRI) scan-
ner and head coil. The selected instruments allow for
comparison of brass versus reed instruments, and differences in
native instrument resistance, including high resistance (trum-
pet), intermediate resistance (trombone), and low resistance
(saxophone). Because circular breathers partially rely on
Laryngoscope 00: Month 2014
2
pushing against the back pressure, or resistance, of the instru-
ment, native instrument resistance may impact the oropharyn-
geal dynamics of circular breathing and overall task efficiency.
Exclusion criteria included age <18 years, contraindication
to MRI, significant imaging artifact, and inability to sustain a
note while circular breathing in a supine position. All subjects
35
were consented per IRB regulations prior to participation.
30
46
46 Imaging Technique and Study Protocol
27 Subjects were studied in a 12-channel head coil in a Sie-
mens Trio 3T (Siemens Healthcare, Erlangen, Germany) MRI
33
scanner, and imaged in the midsagittal plane using a fast radial
1 k-space sampled, T1-weighted gradient echo cMRI technique
11 (LiveView; Siemens Healthcare).10 The sequence acquired a sin-
gle 8-mm slice with a 25.6-cm field of view and a 128 3 128
matrix, yielding an in-plane voxel size of 2.0 3 2.0 mm. A flip
angle of 5 was used, with repetition and echo times of 2.1 ms
and 1.3 ms, respectively, producing a frame speed of 19 fps.
Subjects were supplied with MRI-compatible instrument
equipment. Trumpet players were given a Kelly plastic trumpet
mouthpiece, size 1-1/2C, and trombonists were given a Kelly
small-shank plastic trombone mouthpiece, size 51D. Saxophon-
ists were permitted to use their own plastic mouthpieces. Metal
ligatures (part of the mouthpiece securing the reed to the plas-
tic body of the mouthpiece) were replaced with modified Velcro
ligatures. Saxophonists provided their own reeds. No image dis-
tortion was evident as a result of this equipment. Mouthpieces
were cleaned with Roche-Thomas Mi-T mouthpiece cleaner
(Roche-Thomas Corp., Dubuque, IA) and alcohol swabs.
Once inside the MRI scanner, subjects were instructed to
breathe quietly through the mouth and imaged with cMRI for 5
seconds to assess baseline dimensions. Oral breathing was
selected as a baseline state, because the action of circular
breathing consists of sustained oral exhalation with superim-
posed nasal inhalation. After completion of quiet breathing, sub-
jects were asked to sustain a note by circular breathing on their
mouthpiece while modeling the usual technique utilized to per-
form their tuning note. Emphasis was placed on modeling their
usual technique, because sustaining the actual pitch of their
tuning note on the mouthpiece alone would not reflect their
true playing technique on the complete instrument. Subjects
were observed to complete a minimum of two complete cycles of
circular breathing before terminating the study, a task achieved
by all subjects on the first attempt. The first two consecutive
cycles of circular breathing were analyzed.
Measurements and Data Analysis
Five predetermined oropharyngeal distances were meas-
ured frame by frame across two complete cycles of circular
breathing per subject: 1) lip to tongue distance (the anteroposte-
rior distance from the posterior border of the lip aperture to the
anterior tip of the tongue), 2) midtongue to palatal height (the
vertical distance from the midpoint of the superior border of
tongue to the inferior border of the hard palate), 3) uvula tip to
tongue distance (the distance from the inferior tip of the soft
palate to the posterior border of the tongue), 4) base of tongue
to first vertebra distance (the anteroposterior distance from the
base of tongue to the superior border of the first vertebra),
and 5) roof of nasopharynx to soft palate height (the vertical
distance from the midpoint of the roof of the nasopharynx to
the soft palate) (Fig. 1). Measurements were recorded for two
consecutive cycles, yielding two complete datasets per subject.
Because circular breathing cycle time varied between subjects,
averages were calculated by normalizing the time as a percent-
age of trial completed.
Peng et al.: Circular Breathing Mechanics With Cine MRI
Fig. 1. Five predetermined oropharyngeal distances. Measurements were conducted frame by frame across two complete cycles of circular
breathing for each subject. (A) Lip to tongue distance. (B) Midtongue to palatal height. (C) Uvula tip to tongue distance. (D) Base of tongue
to first vertebra distance. (E) Roof of nasopharynx to soft palate height.

The volume of the oral airway was measured using a semi- 5) return of the tongue to baseline (Fig. 2) (see Support-
automated bounding box technique written with IDL 8.1 (Excelis ing Video 1 in the online version of this article). These
Visual, Boulder, CO). A rectangular box was defined, including steps were clearly noted on imaging and by graphical
the entire tongue and oral cavity. Airway volume was calculated representation of oropharyngeal measurements over
using a region-growing algorithm to include all neighboring vox-
time (Fig. 3). Occlusion of the posterior oropharynx from
els within a specific intensity range. This threshold was defined
the anterior oral cavity was achieved during step 2,
for each subject as the mean 6 three standard deviations of a 5
3 5-voxel kernel placed just posterior to the lips. Neighboring when the soft palate swings inferiorly and anteriorly to
voxels were smoothed, dilated, and eroded to produce a contigu- meet the posterior tongue, creating a seal.
ous mask of the airway at each point in the cine loop. The total Relative occlusion time was significantly longer
number of voxels per frame was multiplied by 4.0E-3 cc/voxel to among amateurs than professionals (P 5.01). A similar
produce a measurement of the volume. trend was noted in absolute occlusion time (P 5.08), sug-
Data were analyzed for displacement and percentage gesting that experienced amateurs required more time to
change from baseline with ImageJ (National Institutes of Health, complete adequate nasal inhalation (Table II). A trend
Bethesda, MD), and assessed for changes in spatial relationships toward longer cycle time was observed among saxophonists
over time. Circular breathing cycle time (time required to com- and trumpeters (P 5.2), although the difference among
plete one five-step cycle of circular breathing, or “cycle time”)
trumpeters disappeared when amateurs were excluded.
and posterior oropharyngeal occlusion time (duration in which
There was no difference in occlusion time among instru-
the posterior oropharynx was fully occluded from the anterior
oropharynx, or “occlusion time”) were recorded for each subject. ment subgroups (Table III). Mean, range, and standard
Cycle time was measured to gauge each subject’s overall task deviations for each oropharyngeal distance are reported
efficiency. Posterior oropharyngeal occlusion was regarded as (Table IV).
critical to circular breathing technique for nasal inhalation and Bounding box data could not be acquired in one ama-
for generation of intraoral pressures high enough to sustain a teur subject due to signal dropout, but we have nonethe-
continuous tone. Occlusion time was measured to gauge each less reported our data in professionals to demonstrate the
subject’s efficiency in nasal inhalation. Because occlusion time measurement technique (Fig. 4). Measurements of airway
may be confounded by total cycle time for each subject, compari-
sons were made with both absolute values (“absolute occlusion
time”) and relative values (“relative occlusion time”). Relative
occlusion time reports the absolute occlusion time as a percent-
age of the subject’s total cycle time, thus accounting for differen-
ces in total time required to complete the task.
Comparisons were made between amateurs and professio-
nals, and between instrument types. In light of the small sam-
ple size of this study, analysis of statistical significance was
conducted using the nonparametric Mann-Whitney U test. Sta-
tistical significance was set at a P value < .05.

RESULTS
All subjects successfully accomplished circular
breathing using their mouthpiece while in the MRI scan-
ner. We were able to obtain clear, interpretable images—
both dynamic and static—of the oral cavity mechanics
during circular breathing. Fig. 2. Graphical representation of a single professional trumpet
All subjects were observed to complete the same player during one complete cycle of circular breathing. Five oro-
series of steps characterized by 1) superior/posterior pharyngeal distances shown here as a percentage of baseline val-
ues over time. Similar findings were elicited from all subjects. (A)
tongue displacement, 2) inferior/anterior soft palate dis- Lip to tongue distance. (B) Midtongue to palatal height. (C) Uvula
placement, 3) anterior tongue and soft palate displace- tip to tongue distance. (D) Base of tongue to first vertebra dis-
ment, 4) superior/posterior soft palate displacement, and tance. (E) Roof of nasopharynx to soft palate height.

Laryngoscope 00: Month 2014 Peng et al.: Circular Breathing Mechanics With Cine MRI
3
Fig. 3. Still-frame radiographs (A) and graphical representation (B) of a single professional trumpet player during circular breathing. Each of
the five steps are shown: step 1, superior/posterior tongue displacement; step 2, inferior/anterior soft palate displacement; step 3, anterior
tongue and soft palate displacement; step 4, superior/posterior soft palate displacement; step 5, return of the tongue to baseline. A 5 lip to
tongue distance; B 5 midtongue to palatal height; C 5 uvula tip to tongue distance; D 5 base of tongue to first vertebra distance; E 5 roof
of nasopharynx to soft palate height.

volume showed a reduction in the airway of 55.4% 6 10.9%
for professional musicians (n 5 6).
DISCUSSION
Breathing exercises have been investigated for clini-
cal utility in disease, including OSA and asthma.
Numerous breathing exercises have been explored in the
literature, including yoga breathing techniques, wind
instrument performance, and circular breathing via
didgeridoo performance.2–4,11–15
Our findings delineate a uniform mechanism for cir-
cular breathing among participants playing different
instruments and with different degrees of playing

Laryngoscope 00: Month 2014 Peng et al.: Circular Breathing Mechanics With Cine MRI
4
 TABLE II.
Results Comparison of Circular Breathing Cycle Time and Posterior Oropharynx Occlusion
Time (Absolute and Relative) by Professional Status.
Professionals, Mean (SD), Range, s Amateurs, Mean (SD), Range, s P

Circular breathing cycle time 1.18 (0.421), 0.630–1.83 2.18 (1.6), 1.2–3.6 .5
Absolute occlusion time 0.480 (0.232), 0.210–0.780 1.5 (1.3), 0.470–2.7 .08
Relative occlusion time 42% (15%), 20%–75% 67% (9%), 56%–75% .01

Relative posterior oropharyngeal occlusion time (relative occlusion time) compares posterior oropharyngeal occlusion time for each subject as a percent-
age of his respective circular breathing cycle time for the same trial.
SD 5 standard deviation.

experience. This anatomical information may aid future
investigations in identifying specific populations of
patients who may benefit from the intervention, and
assist in generating theories for other potential applica-
tions of the therapy, if proven useful. In this study, expe-
rienced amateurs and professionals performed the task
similarly. Our results suggest that experienced ama-
teurs, and potentially nonmusicians, have the ability to
acquire this skill in a manner that is mechanistically
comparable to professionals. Professionals performed the
task in less time than experienced amateurs, suggesting
that their movements were more efficient.
Anatomy and Physiology of Circular Breathing
Our results demonstrate that circular breathing pri-
marily involves the posterior tongue and the velophar-
ynx. Because circular breathing consists of nasal
inhalation superimposed on sustained oral exhalation,
the most critical movements appear to involve the com-
ponent of nasal inhalation. Previous research on the
velopharyngeal system states that the movement of
these muscles in humans relies largely on changes in
airflow and air pressure,16 raising questions about which
muscles, if any, are specifically strengthened by circular
breathing. Current literature also suggests that humans
have limited voluntary control of velopharyngeal move-
ments, and that kinesthesia and proprioception in the
velar anatomy is minimal.16
The pharyngeal constrictors may play a role in modi-
fying airway diameter (and thus airflow and air pressure)
through the oropharynx. However, these changes may be
insufficient to explain circular breathing. In contrast with
the observed anterior/inferior movement of the velum,
increased airflow in the posterior oropharynx during
nasal inhalation would draw the velum posteriorly via
Bernoulli’s principle. Increased pressure in the anterior
oral cavity (from external cheek compression) may also
exert force posteriorly. Thus, the action of the palatoglos-
sus muscle likely plays an important role in opposing
these forces, causing both the downward and forward
motion of the velum and the elevation of the posterior
tongue. Further studies are required to determine
whether the palatoglossus is a “trainable” muscle.
Learnability of Circular Breathing
The learnability of circular breathing in
instrument-na€ıve users remains unknown; a learnability
assessment in this population is needed to delineate the
clinical feasibility and cost-effectiveness of teaching this
skill as therapy. Furthermore, the instrument utilized
may impact learnability and efficacy. Ward et al. demon-
strated in a nationwide survey study that rates of OSA
were lower in double reed instrumentalists, but not in
wind instrumentalists, compared to the general popula-
tion.17 Although our findings suggest that native instru-
ment resistance does not alter the oropharyngeal
dynamics of the technique, there may be differences in
therapeutic efficacy depending upon the instrument
used. Evaluation with different instruments would be
valuable in the study of circular breathing as therapy.
Circular Breathing and OSA
Puhan et al. postulated that the perceived benefit of
circular breathing during didgeridoo performance in OSA
was related to strengthening of upper airway muscula-
ture.2 Decreased pharyngeal tone has been reported as a
potential contributor to the pathophysiology of OSA.18,19
Although the role of circular breathing in the treatment
of OSA must be further elucidated, our findings demon-
strate utilization of velopharyngeal structures in a way
that may strengthen the involved muscles.

TABLE III.
Results Comparison of Circular Breathing Cycle Time and Posterior Oropharynx Occlusion
Time (Absolute and Relative) by Instrument Type.
Trumpets, Mean (SD), Range, s Trombones, Mean (SD), Range, s Saxophones, Mean (SD), Range, s

Circular breathing cycle time 1.65 (1.3), 0.630–3.6 0.880 (0.137), 0.680–1.05 1.52 (0.237), 1.3–1.8
Absolute occlusion time 0.980 (1.1), 0.210–2.7 0.520 (0.184), 0.370–0.780 0.510 (0.258), 0.260–0.730
Relative occlusion time 50% (16%), 31%–75% 56% (15%), 39%–75% 34% (17%), 20%–54%

Relative posterior oropharyngeal occlusion time (relative occlusion time) compares posterior oropharyngeal occlusion time for each subject as a percent-
age of his respective circular breathing cycle time for the same trial.
SD 5 standard deviation.

Laryngoscope 00: Month 2014 Peng et al.: Circular Breathing Mechanics With Cine MRI
5
 TABLE IV.
Mean, Standard Deviation, and Range of All Measured Oropharyngeal Distances in All Subjects.
Lip to Tongue Midtongue-Palatal Uvula Tip to Tongue Base of Tongue Roof of Nasopharynx to
Performer Distance, Height, Mean (SD), Distance, to First Vertebra Distance, Soft Palate Height,
Subject Type Mean (SD), Range Range Mean (SD), Range Mean (SD), Range Mean (SD), Range

1 Professional, 1.0 (0.21), 0.50–1.5 0.80 (0.28), 0.03–1 1.1 (0.75), 0–3.7 1.3 (0.33), 0.82–3.3 1.0 (0.61), 0.95–2.8
trumpet
2 Professional, 0.77 (0.22), 0.41–1.4 1.0 (0.27), 0.25–1.3 0.89 (0.32), 0–1.1 1.2 (0.33), 1.0–1.8 1.0 (0.24), 0.96–1.8
trumpet
3 Professional, 0.70 (0.33), 0.38–1 0.53 (0.33), 0.03–1 0.81 (0.23), 0–1 1.1 (0.26), 0.78–2.0 0.96 (0.27), 0.95–1.6
trumpet
4 Professional, 0.91 (0.21), 0.36–1.1 0.91 (0.29), 0.02–1 1.3 (0.48), 0–1.8 1.1 (0.53), 0.8–2.8 1.0 (0.44),0.85–2.5
trombone
5 Professional, 0.97 (0.40), 0.76–2.3 0.93 (0.50), 0.03–1.2 1.0 (0.43), 0–1.2 0.92 (0.28), 0.12–1.1 0.96 (0.80), 0.92–3.5
saxophone
6 Professional, 1.08 (0.33), 0.86–2.2 0.81 (0.30), 0.11–1 0.91 (0.34), 0–1.0 1.0 (0.13), 0.66–1.2 0.99 (0.62), 0.88–2.6
saxophone
7 Amateur, 0.99 (0.24), 0.61–1.4 0.98 (0.39), 0.12–1.2 1.0 (0.43), 0–1.5 1.1 (0.81), 0.50–2.7 0.99 (0.97), 0.94–3.3
trumpet
8 Amateur, 1.0 (0.21), 0.53–1.4 0.99 (0.39), 0.02–1.1 1.6 (0.92), 0–3.3 1.5 (2.1), 0.75–7.8 0.99 (0.54), 0.93–2.7
trombone

All values are reported as ratios of baseline. Because the circular breathing cycle time varied significantly between subjects, average values were calcu-
lated by normalizing the time as a percentage of trial completed. Subject numbers correlate with the demographics listed in Table I.

Exercises derived from speech therapy by Guimaraes
et al. were effective in reducing symptoms, such as snor-
ing and daytime sleepiness, among adult patients with
moderate OSA.20 These exercises were designed for soft
palate elevation and tongue repositioning, with specific
targeting of upper airway musculature, including the ten-
sor and levator veli palatini, palatopharyngeus, and the
palatoglossus. Previous studies have demonstrated that 8
weeks of twice-daily 20-minute tongue muscle training
did not reduce apnea-hypopnea index scores despite a
reduction in snoring.21 Our results suggest that strength-
ening in circular breathing may be related to velophar-
yngeal movement against resistance. Because circular
breathing utilizes both the velopharynx and tongue, how-
ever, it is not clear if one or both of these elements would
be responsible for therapeutic benefit.
Further investigation is required to compare circular
breathing to other techniques, such as the Guimaraes
exercises and typical wind instrument performance.22
The possibility that circular breathing may specifically
and reliably strengthen the upper pharyngeal muscula-
ture is supported by our observations in this study,
including the observed soft palate movement against pre-
sumably higher pressures in the oral cavity anteriorly
during performance, the ability of experienced amateurs
and professionals to apply similar mechanics during the
exercise, and the repetitive nature of the exercise.
Clinical Applications for cMRI
cMRI has numerous applications for the dynamic
oral cavity, including evaluation of speech pathology,

Fig. 4. Graphical representation of

airway volume during three cycles of
circular breathing in a single profes-
sional saxophonist. This clearly dem-
onstrates three episodes of greatly
decreased airway volume coinciding
with occlusion of the posterior oro-
pharynx. [Color figure can be viewed
in the online issue, which is available
at www.laryngoscope.com.]

Laryngoscope 00: Month 2014 Peng et al.: Circular Breathing Mechanics With Cine MRI
6
dysphagia, and obstructive sleep apnea.23,24 We demon-
strated one method by which real-time analysis of oro-
pharyngeal motion could be objectively and reproducibly
conducted to ascertain a greater anatomical understand-
ing of a potential therapeutic intervention. The appropri-
ate measurements and analysis will likely vary
considerably based upon the reason for investigation.
Nonetheless, cMRI may prove to be an invaluable imag-
ing modality in the assessment of oropharyngeal move-
ment, and further applications must be investigated to
explore the full potential of this technology.
Limitations
This investigation is limited by a small sample size
and few instruments studied. This was unavoidable,
because time on the cMRI machine is expensive, and
this was designed as a pilot study. Although we are
aware of the resultant limitations of our analysis and
affirm that larger groups are needed to make statisti-
cally significant comparisons, we feel it is important to
report our data to provide a foundation for future inves-
tigations into the mechanics and therapeutic potential of
circular breathing.
The exact impact of the modified instrument equip-
ment and supine position on circular breathing tech-
nique is not known, but is not felt to be significant. The
plastic trumpet and trombone mouthpieces used in this
study are commercial models used by musicians for play-
ing in cold weather conditions. Saxophone-playing sub-
jects utilized normal saxophone mouthpieces, and the
modification applied (Velcro ligatures) served only to
secure the reed to the mouthpieces. Supine position
likely caused the task of circular breathing to be more
challenging, but study subjects were able to accomplish
the task in this position. Nonetheless, the potential
impact on ease of task performance and on task effi-
ciency is worth noting.
CONCLUSION
Circular breathing has been studied as potential
therapy in obstructive sleep apnea. This study
delineated the velopharyngeal mechanics of circular
breathing, and demonstrated similar mechanics across
subgroups, with increased efficiency among professionals
compared to experienced amateurs. Our data represent
a real-time anatomical study of circular breathing and
provide a foundation for future studies. cMRI is a reli-
able technique for imaging dynamic motion in the oral
airway, and has many potential applications for the eval-
uation of oropharyngeal pathology and therapy.
Acknowledgments
The authors would like to acknowledge the Music and
Medicine Initiative of Weill Cornell Medical College and
the Department of Radiology of NewYork-Presbyterian
Laryngoscope 00: Month 2014
View publication stats
Hospital/Weill Cornell Medical Center for their generous
contributions in funding this investigation. In addition,
the authors would like to acknowledge the Departments of
Otolaryngology and Radiology of NewYork-Presbyterian
Hospital/Weill Cornell Medical Center, the Citigroup Bio-
medical Imaging Center, The Juilliard School, Dr. David
Shapiro, and Dr. Gary S. Dorfman for their professional
advisement and support for this study.
BIBLIOGRAPHY
1. Kynaston T, Kynaston TP. Circular Breathing: For the Wind Performer.
Los Angeles, CA: Alfred Music; 1978.
2. Puhan MA, Suarez A, Cascio CL, Zahn A, Heitz M, Braendli O. Didgeridoo
playing as alternative treatment for obstructive sleep apnoea syndrome:
randomised controlled trial. BMJ 2006;332:266–270.
3. Eley R, Gorman D. Didgeridoo playing and singing to support asthma
management in Aboriginal Australians. J Rural Health 2010;26:100–
104.
4. Eley R, Gorman D. Music therapy to manage asthma. Aboriginal Islander
Health Worker J 2008;32:9–10.
5. Izdebski K, Hyde L, Ward RR, Ross JC. Visual observations of glottic activ-
ity during didgeridoo performance. Proc. SPIE 8207. Photonic Therapeu-
tics and Diagnostics VIII, 82072U (February 3, 2012).Available at:
http://dx.doi.org/10.1117/12.925215. Accessed on July 17, 2013.
6. Heitner J, Bhumireddy GP, Crowley AL, et al. Clinical application of cine-
MRI in the visual assessment of mitral regurgitation compared to echo-
cardiography and cardiac catheterization. PloS One 2012;7:e40491.
7. Kirchhoff S, Nicolaus M, Schirra J, Reiser MF, G€oke B, Lienemann A.
Assessment of colon motility using simultaneous manometric and func-
tional cine-MRI analysis: preliminary results. Abdom Imaging 2011;36:
24–30.
8. Wakamiya M, Furukawa A, Kanasaki S, Murata K. Assessment of small
bowel motility function with cine MRI using balanced steady-state free
precession sequence. J Magn Reson Imaging 2011;33:1235–1240.
9. Fukushi I, Okada Y. Dynamics of upper airways during the M€ uller maneu-
ver in healthy subjects: A cine MRI study. Adv Exp Med Biol 2013;788:
189–195.
10. Niebergall A, Zhang S, Kunay E, et al. Real-time MRI of speaking at a
resolution of 33 ms: undersampled radial FLASH with nonlinear inverse
reconstruction. Magn Reson Med 2013;69:477–485.
11. Singh V, Wisniewski A, Britton J, Tattersfield A. Effect of yoga breathing
exercises (pranayama) on airway reactivity in subjects with asthma.
Lancet 1990;335:1381–1383.
12. Cooper S, Oborne J, Newton S, et al. Effect of two breathing exercises
(Buteyko and pranayama) in asthma: a randomised controlled trial.
Thorax 2003;58:674–679.
13. Holloway E, Ram FSF. Breathing exercises for asthma. Cochrane Database
Syst Rev 2004;(1):CD001277.
14. Lucia, R. Effects of playing a musical wind instrument in asthmatic teen-
agers. J Asthma 1994;31:375–385.
15. Deniz O, Savci S, Tozkoparan E, Ince DI, Ucar M, Ciftci F. Reduced pul-
monary function in wind instrument players. Arch Med Res 2006;37:
506–510.
16. Hixon TJ, Weismer J, Hoit JD. Velopharyngeal-nasal function and speech
production. In: Preclinical Speech Science: Anatomy Physiology Acous-
tics Perception. 2nd ed. San Diego, CA: Plural Publishing; 2014:171–231.
17. Ward CP, York KM, McCoy JG. Risk of obstructive sleep apnea lower in
double reed wind musicians. J Clin Sleep Med 2012;8:251–255.
18. Horner, RL. Pathophysiology of obstructive sleep apnea. J Cardiopulm
Rehabil Prev 2008;28:289–298.
19. Ayappa I, Rapoport DM. The upper airway in sleep: physiology of the
pharynx. Sleep Med Rev 2003;7:9–33.
20. Guimaraes KC, Drager LF, Genta PR, Marcondes BF, Lorenzi-Filho G.
Effects of oropharyngeal exercises on patients with moderate obstructive
sleep apnea syndrome. Am J Respir Crit Care Med 2009;179:962–966.
21. Randerath WJ, Galetke W, Domanski U, Weitkunat R, Ruhle KH. Tongue-
muscle training by intraoral electrical neurostimulation in patients with
obstructive sleep apnea. Sleep 2004;27:254–258.
22. Brown DL, Zahuranec DB, Majersik JJ, et al. Risk of sleep apnea in
orchestra members. Sleep Med 2009;10:657–660.
23. Kitano H, Asada Y, Hayashi K, Inoue H, Kitajima K. The evaluation of
dysphagia following radical surgery for oral and pharyngeal carcinomas
by cine-magnetic resonance imaging (cine-MRI). Dysphagia 2002;17:
187–191.
24. Takemoto H, Honda K, Masaki S, Shimada Y, Fujimoto I. Measurement of
temporal changes in vocal tract area function from 3D cine-MRI data.
J Acoust Soc Am 2006;119:1037.
Peng et al.: Circular Breathing Mechanics With Cine MRI
7