AM ER IC AN JOUR NA L OF OTOLARY NG OLOG Y H EA D A N D N E CK ME D I CI NE AN D SUR G E RY 3 4 ( 2 0 13 ) 62 66 3 0

Available online at www.sciencedirect.com

ScienceDirect
www.elsevier.com/locate/amjoto

Analysis of obstruction site in obstructive sleep apnea

syndrome patients by drug induced sleep endoscopy,
Soo Kweon Koo, MD, PhD, Jang Won Choi, MD, Nam Suk Myung, MD, Hyoung Ju Lee, MD,
Yang Jae Kim, MD, Young Joong Kim, MD
Department of Otorhinolaryngology-Head and Neck Surgery, Busan Saint Mary's Hospital, Busan, Korea

ARTI CLE I NFO

A BS TRACT

Article history:

Purpose: We analyzed site, pattern and degree of obstruction in Korean male obstructive

Received 18 January 2013

sleep apnea syndrome (OSAS) patients by drug-induced sleep endoscopy (DISE). We also
investigated possible links between BMI, AHI and DISE findings.
Materials and methods: Sixty-nine male patients underwent DISE. DISE findings were
reported using our classification system in which modified VOTE classification
obstruction type, site of obstruction, degree of obstruction and anatomical site
contributing obstruction was reported. Associations were analyzed among the results of
the polysomnography, patients' characteristics and DISE finding.
Results: Multilevel airway obstruction was found in 84.06% of patients and 15.94% had a
unilevel obstruction. Among those with unilevel obstruction, 90.90% had retropalatal level
obstruction and 9.10% had retrolingual level obstruction. Palate with lateral pharyngeal wall
obstruction (49.28%) is the most common obstruction type of the retropalatal level and
tongue with lateral pharyngeal wall (37.68%) is the most common obstruction type of the
retrolingual level. Examining the relation between obstruction site according to body mass
index (BMI) and severity of OSAS (apnea hypopnea index, AHI), the lateral pharyngeal wall
had an increasing tendency associated with higher BMI and higher AHI. But the lateral
pharyngeal wall of both levels was statistically significant associated with higher AHI.
Conclusion: The majority of the Korean male OSAS patients have multilevel obstruction and
according to BMI and AHI, the DISE findings indicate that the lateral pharyngeal wall is the
most important anatomical site contributing to obstruction regardless of the level at which
the obstruction lies.
2013 Elsevier Inc. All rights reserved.

1.

Introduction

Obstructive sleep apnea syndrome (OSAS) is a growing

problem associated with excessive daytime sleepiness and
an increased risk of cardiovascular and cerebrovascular
complication [13]. Continuous positive airway pressure
(CPAP) therapy remains the first line medical therapy in the

management of OSAS. However, long term compliance to

CPAP treatment is suboptimal and many patients with OSAS
seek alternative treatment, including surgical treatment [4,5].
For surgical treatment planning of snoring and OSAS,
identification and determination of the site, pattern and
engaging anatomical structure are important [6]. Upper
airway anatomical assessments such as CT, MRI and cepha-

This research protocol was reviewed and approved after deliberation by the Busan Saint Mary's Hospital Institutional Review Board (IRB).
This manuscript was presented at the 24th ERS & 31th ISIAN, Toulouse, France, June 1721, 2012.
Corresponding author. 538-41 Yongho-Dong, Nam-Gu, Busan 608-838, Korea. Tel: +82 51 933 7214; fax: + 82 51 956 1956.
E-mail address: koosookweon@naver.com (S.K. Koo).

0196-0709/$ see front matter 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.amjoto.2013.07.013

AM ER IC AN JOUR NA L OF OTOLARY NG OLOG Y H EA D A N D N E CK ME D I CI N E AN D SUR G E RY 3 4 ( 2 0 13 ) 62 66 3 0

lometric measurement are often static and performed during

wakefulness, which may not represent dynamic upper airway
behavior during sleep. Croft and Pringle [7] described "sleep
nasoendoscopy" in 1991 as an endoscopic evaluation of the
upper airway during sedation. It was developed in multiple
centers throughout Europe and is now widely performed.
Kezirian proposed Drug induced sleep endoscopy (DISE), and
designed the VOTE classification that focuses on the specific
structures that contribute to obstruction [8,9]. Although DISE
has some limitations such as a limited period of observation
and multiple classification systems, it approaches the natural
physiologic state of sleep more than currently available
diagnostic tools. Multiple classification systems prevent the
comparison of results across studies and centers. VOTE
classification is a method of characterizing DISE findings
focusing on its core feature, the specific structure that
contributes to obstruction. This system may be an oversimplification of upper airway structure and it has some
limitations for site specific surgical treatment [8]. Although
many studies for DISE were performed with Westerners, a
study of Asians is rare. In some Asian countries, life patterns
such as diet tend to westernization, but the way of life of
Asians differs from that of westerners. We based our
classification system on a modification of the VOTE system
and analyzed site, pattern and degree of obstruction in Korean
male OSAS patients by DISE. We also investigate possible links
between BMI, AHI and DISE findings.

2.

Patients and methods

2.1.

General setting of study

In participants consenting for this research, Korean male

patients who underwent DISE by one examiner at Busan Saint
Marys Hospital, ENT department were enrolled. To exclude
gender differences, only male patients were enrolled. Data
were collected prospectively. This research protocol was
reviewed and approved after deliberation by the Busan Saint
Mary's Hospital Institutional Review Board (IRB).
Before DISE, all patients received a thorough ear, nose, and
throat examination and a medical history was taken. All
patients underwent full night PSG (Polysomnography, WEE1000K, Nihon Kohden, Japan), and sixty-nine patients were
diagnosed by OSAS. Patients with obvious retrognathia,

627

mandibular dysplasia, or who had undergone prior surgery

of the soft palate or tongue were excluded.

2.2.

Drug induced sleep endoscopy

The DISE technique is as follows. DISE was performed after

application of local anesthetic spray in the nasal cavity. Sleep
was induced, with patients in supine position, by intravenous
administration of midazolam (0.07 mg/kg). In our study, DISE
was performed under respiratory monitoring and through the
help of an anesthesiologist in the operating room. Once
patients were asleep, a 4 mm flexible videolaryngoscope was
introduced gently through each patients nose. The video
images of recorded DISE procedures were later evaluated by
one otolaryngologist.

2.3.

Classification system

We divided the pharynx into two portions: the retropalatal

level (the region of posterior to the soft palate) and the
retrolingual level (the region of the pharynx posterior to the
vertical portion of the tongue). On this basis, we made our
classification system, a modified VOTE classification. Our
classification system included the following about the obstruction: site, degree, and the anatomical structure contributing the
most. The retropalatal level was subdivided into the palate
(antero-posterior diameter), lateral pharyngeal wall (lateral
diameter), tonsil (specific structure contributing to obstruction). The retrolingual level was divided into the tongue base
(antero-posterior diameter), the lateral pharyngeal wall (lateral
diameter), and the epiglottis (specific structure contributing to
obstruction). Degree of airway obstruction was categorized as
no obstruction (0), partial obstruction (1, 50%75%), and
complete obstruction (2, >75%) (Table 1).

2.4.

Relation between BMI, AHI and DISE finding

BMI was defined as body weight(kg)/height(m)2. We classify

patients into lower BMI group (BMI < 25) and higher BMI group
(BMI 25), and compared DISE findings. The severity of OSAS
is expressed in the apnea hypopnea index (AHI). Apnea was
defined as a cessation of airflow for at least 10 s and hypopnea
was defined as period of reduction equal to 30% airflow for at
least 10 s and 4% decreased O2 saturation. The AHI was
calculated as sum of total apnea and hypopnea events.

Table 1 Classification of DISE findings

Level

Degree of obstruction a

Retropalatal

0/1/2

Retrolingual

0/1/2

Configuration b
AP diameter

Lat. diameter

Contributing structure

Palate
+/
Tongue base
+/

LPW
+/
LPW
+/

Tonsil
+/
Epiglottis
+/

DISE: Drug-induced sleep endoscopy, AP: Antero-posterior, Lat.: Lateral, LPW: Lateral pharyngeal wall.
a
Degree of obstruction has one number for each structure: 0 = no obstruction (no vibration), 1 = partial obstruction (vibration, 50%75%), 2 =
complete obstruction (collapse,>75%).
b
Dichotomous configuration noted for structures with degree of obstruction greater than 0.

628

AM ER IC AN JOUR NA L OF OTOLARY NG OLOG Y H EA D A N D N E CK ME D I CI NE AN D SUR G E RY 3 4 ( 2 0 13 ) 62 66 3 0

Patients who were diagnosed as having OSAS were classified

into mild to moderate group (5 AHI < 30) and severe group
(AHI 30). We also compared the AHI and DISE findings.
Each obstruction level and the anatomical structure
contributing the most were calculated separately, with
percentages expressed as a fraction of the total number of
each subgroup. The sum of the percentages is greater than
100% because it is possible to have more than one most
contributing anatomical structure to airway obstruction in
each patient. To exclude gender differences, only male
patients were included.

2.5.

Statistical analysis

All statistical analyses were performed using SPSS (Statistical

Package for the Social Sciences, version 17.0, SPSS, Inc., an IBM
Company, Chicago, Illinois, USA). The unpaired t test was used
to evaluate associations of absence or presence of various
obstruction sites and its configurations with each subgroup
according to the AHI and BMI. Null hypotheses of no
difference were rejected if p-values were less than 0.05.
Results were presented as mean and standard deviation.

3.

Results

The study sample consisted of 69 male patients, with mean

( 2SD) age = 41.85 2.95 years (range = 1870 years), mean
weight = 82.82 9.23 kg (range = 50122 kg), mean height =
172.91 12.31 cm (range = 156192 cm) and mean BMI =
25.68 0.62. The mean AHI was 21.96 4.70 (median = 14.28).
The mean score of Epworth Sleepiness Scale(ESS) was 9.91
1.12 (median = 9) (Table 2).
The majority of the patients had multilevel obstruction
(84.06%); the remainder had unilevel obstruction (15.94%). The
retropalatal level obstructions were 90.90% and retrolingual
level obstructions were 9.10% of the unilevel obstructions. The
anatomical site that contributed most was the palate with
lateral pharyngeal obstruction (49.28%), then the palate with
tonsil and lateral pharyngeal wall obstruction (23.19%),
followed by obstruction of the palate alone (20.29%) at the
retropalatal level obstruction (Fig. 1). In retrolingual level
obstruction, 37.68% were tongue base with lateral pharyngeal
wall obstruction, next was obstruction of the lateral pharyngeal wall alone (18.84%), and obstruction of the tongue base

Fig. 1 Classification and contributiveness of the single and

multiple obstructive structures according to the retropalatal
level (LPW: Lateral pharyngeal wall on the retropalatal level).

alone (11.59%) followed by tongue, epiglottis with lateral

pharyngeal wall(5.80%) and epiglottis alone(5.80%) (Fig. 2).
Examining the relation between obstruction site and BMI,
the lateral pharyngeal wall of retropalatal level had an
increasing tendency associated with higher BMI. But the
correlation, though not negligible, was insufficient to establish statistical significance. (p = 0.0974) (Table 3). Examining
the relation between obstruction site and severity of OSAS(AHI), the lateral pharyngeal wall of both (retropalatal and
retrolingual) levels was statistically significantly associated
with higher AHI (retropalatal level: p = 0.0169, retrolingual
level: p = 0.0434) (Table 4).

4.

Discussion

DISE offers a unique structure-based assessment of the

airway, compared with other commonly used evaluation
techniques and it is used in the clinical practice not only to
locate obstructions, but as an aid in surgical decision-making.

Table 2 Patient characteristics (number of patients = 69).

Patient characteristics
Age, years
BMI, kg/m2
ESS
AHI, per hours
Average SaO2, %
Minimal SaO2, %

Mean 2SD
41.85
25.68
9.91
21.96
95.66
80.27

2.95
0.62
1.12
4.70
0.97
2.77

SD: standard deviation, BMI: body mass index, ESS: Epworth

Sleepiness Scale, PSG: polysomnography, AHI: apnea hypopnea
index, SaO2: oxygen saturation.

Fig. 2 Classification and contributiveness of the single and

multiple obstructive structures according to the retrolingual
level (LPW: Lateral pharyngeal wall on the retrolingual level).

AM ER IC AN JOUR NA L OF OTOLARY NG OLOG Y H EA D A N D N E CK ME D I CI N E AN D SUR G E RY 3 4 ( 2 0 13 ) 62 66 3 0

Table 3 Correlation between contributing obstruction

site of each obstruction level and BMI (Total patients
No. = 69, *: p value < 0.05).
25 (No. = 40)

< 25 (No. = 29)

No.

No. a

Obstruction structure of Retropalatal level

Palate
27
93.10%
LPW
19
65.52%
Tonsils
10
34.48%

38
32
9

95.00%
80.00%
22.50%

Obstruction structure of Retrolingual level

Tongue
19
65.52%
LPW
19
65.52%
Epiglottis
6
20.69%

22
25
6

55.00%
62.50%
15.00%

No.: Number, BMI: Body mass index, LPW: Lateral pharyngeal wall.
a
Each level and structure are considered separately, with percentages
expressed as a fraction of total number of each subgroup. Percentages
sum to greater than 100% because it was possible for a patient to have
more than one structure contributing to airway obstruction.

For example, UPPP (Uvulopalatopharyngoplasty) is today's

most frequently used surgical technique to treat OSAS at the
retropalatal level. When UPPP is carried out without preoperative assessment of the site of upper airway obstruction by
means of DISE, its success rate as described in a meta-analysis
of a large series of patients was only 41% [10]. However, when
UPPP is only carried out in the patients in whom obstruction of
the airways at the level of the retropalatal level is diagnosed
by DISE, its success rate almost doubles [11]. DISE has good
testretest reliability, especially in its evaluation of hypopharyngeal airway [12], but has some limitations. Drug induced
sleep is not equivalent to natural physiologic sleep. Ideally,
sleep endoscopy would be performed during natural sleep, but
it is rarely done [13,14]. Rabelo et al. recently reported that the

Table 4 Correlation between contributing obstruction site

of each obstruction level and AHI (Total patients No. = 69, *:
p value < 0.05).
Mildmoderate
(5 AHI < 30)
(No. = 50)
No. a

Severe
(AHI 30)
(No. = 19)
No. a

Obstruction structure of Retropalatal level

Palate
49
98.00%
LPW*
42
84.00%
Tonsils
41
82.00%

19
17
17

100.00%
89.47%
89.47%

Obstruction structure of Retrolingual level

Tongue
47
94.00%
LPW*
34
68.00%
Epiglottis
11
22.00%

18
17
8

94.74%
89.47%
42.11%

No.: Number, AHI: Apneahypopnea index, LPW: Lateral

pharyngeal wall.
a
Each level and structure are considered separately, with percentages
expressed as a fraction of total number of each subgroup. Percentages
sum to greater than 100% because it was possible for a patient to have
more than one structure contributing to airway obstruction.

629

use of propofol-induced REM sleep, which alters sleep

architecture slightly, does not influence the respiratory
pattern or significantly influence AHI [15]. The same conclusions were drawn in a study examining the effects of
diazepam [16]. To prevent awakening when introducing the
endoscope, we applied topical anesthesia to widen the nostril.
It is important not to over-anesthetize, because of aspiration
and coughing risks. According to White et al. [17], OSAS
increases following application of topical anesthesia to both
nostrils because nasal receptors sensitive to airflow are
maintaining breathing rhythmicity during sleep. Thus, we
applied topical anesthesia to one nostril only and avoided
aspiration into the pharynx. When using a sedative, the
degree of anesthetic depth is very important. Over-sedation
can cause a decrease in upper airway muscle tone and can
increase OSAS [18]. Thus far, there has been no standard
protocol, although it appears that ideal concentration of
sedative differs by individual. We performed tailored patient
specific procedures with the collaboration of an anesthesiologist. We induced sleep using low-dose midazolam, acting
only as a sleep inducer and not as a sedative agent. Midazolam
is inexpensive and readily available. In our study, DISE was
performed under respiratory monitoring and with the help of
an anesthesiologist in the operating room. Although DISE has
some limitations, it approaches the more natural physiologic
state of sleep than currently available diagnostic tools.
The evolution of sleep endoscopic technique has led to the
formulation of various grading systems. However none provides a comprehensive and accurate classification reflecting
the endoscopic findings in patients with OSAS and there is no
uniform nomenclature. Ideally, the system should cover the
entire upper airway, be both simple and practical, and provide
a means to quantify the severity of the obstruction. Its use
should also make it possible for physicians to establish clinical
diagnosis and standardized treatment. Kezirian proposed a
VOTE classification that focuses on the specific structures that
contribute to obstruction [9]. In surgical treatment of OSAS,
tonsillectomy, UPPP and lateral pharyngoplasty are very
important site-specific surgical procedures, but it is difficult
to plan such surgical procedure only by VOTE classification
because it may oversimplify upper airway structures. For
example, velum by VOTE classification includes the soft
palate, uvula, and lateral pharyngeal wall of velopharynx.
Oropharyngeal lateral wall by VOTE classification includes
palatine tonsil and lateral pharyngeal wall tissue, so we
cannot decide individual surgical technique by VOTE classification. On this basis and a modification of VOTE classification, we made our classification. Fujita et al. [19] divides the
upper airway functionally into two portions: the retropalatal
pharynx (the region of the pharynx posterior to the soft palate)
and the retrolingual pharynx (the region of the pharynx
posterior to the vertical portion of the tongue). The division of
the pharynx into these entities does not represent a formal
anatomic classification but, rather, a descriptive paradigm
that appears to have significance in terms of functional and
surgical consideration.
We analyzed video images in three aspects. The first
represents the region of obstruction, the second represents
the pattern and severity of obstruction, and the third
represents the structure that engages the obstruction.

630

AM ER IC AN JOUR NA L OF OTOLARY NG OLOG Y H EA D A N D N E CK ME D I CI NE AN D SUR G E RY 3 4 ( 2 0 13 ) 62 66 3 0

Thus, our method serves two major purposes of OSAS

surgical upper airway evaluation. One is to characterize the
pattern of the airway obstruction and the second is to select
the surgical treatment.
The obstruction of the upper airway during sleep of OSAS
patients may be isolated or in multiple segments. In our study,
about 84% of OSAS patients have multilevel obstruction.
According to BMI and AHI in Korean male patients, the DISE
findings indicate that the lateral pharyngeal wall is the most
different anatomical site contributing to obstruction regardless
of the level at which the obstruction lies. This finding suggests
that surgeons should consider treating the lateral pharyngeal
wall obstruction in patients with a higher BMI and AHI.
The role of lateral pharyngeal wall collapse as a significant
contributor to airway obstruction in OSAS patients has been
speculated about for decades but remains poorly understood.
Rodenstein et al. first noticed that the magnetic resonance
imaging scanned upper airway of patients OSAS was narrowed
in the lateral dimension when compared to the airways of
normal subjects [20]. Ciscar et al. found that the lateral
pharyngeal walls of OSAS patients were also more collapsible
than those of normal subjects [21]. Schellenberg et al. found
lateral pharyngeal wall narrowing to be the only statistically
significant factor for OSAS in males [22]. Our study found the
same result, Bachar et al. found that the severity of OSAS was
significantly associated with the AHI and not associated with
BMI. By site, the tongue base and hypopharynx were significantly related to obstruction severity [23]. Our study found no
relation between BMI, AHI and obstruction of other sites except
lateral pharyngeal wall. Our guess is that such differences are
due to ethnic differences and small study samples. A larger,
more uniform study sample should be planned to assess
surgical results with the postoperative endoscopic examination.
We conclude that in greater and lesser degrees, the lateral
pharyngeal wall is the most important anatomical site
contributing to obstruction regardless of the level, associated
with higher BMI and higher AHI in Korean male patients.
REFERENCES

[1] Peppard PE, Young T, Palta M, et al. Prospective study of the

association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:137884.
[2] Javaheri S, Wexler L. Prevalence and treatment of breathing
disorders during sleep in patients with heart failure. Curr
Treat Options Cardiovasc Med 2005;7:295306.
[3] Yaggi HK, Concato J, Kernan WN, et al. Obstructive sleep
apnea as a risk factor for stroke and death. N Engl J Med
2005;353:203441.
[4] Kohler M, Smith D, Trippet V, et al. Predictor long-term
compliance with continous positive airway pressure. Thorax
2010;65:82932.
[5] Caples SM, Rowley JA, Prinsell JR, et al. Surgical modifications
of the upper airway for obstructive sleep apnea in adults: a

[6]
[7]

[8]
[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

systematic review and meta-analysis. Sleep 2010;33:

1396407.
Sundaram S, Bridgman SA, Lim J, et al. Surgery for obstructive
sleep apnoea. Cochrane Database Syst Rev 2005;19 CD001004.
Croft CB, Pringle M. Sleep nasoendocopy: a technique of
assessment in snoring and obstructive sleep apnea. Clin
Otolaryngol Allied Sci 1991;16:5049.
Kezirian FJ. Drug-induced sleep endoscopy. Op Tech Otolaryngol 2006;17:2302.
Kezirian EJ, Hohenhorst W, de Vries N. Drug-induced sleep
endoscopy: the VOTE classification. Eur Arch Otorhinolaryngol 2011;268:12336.
Sher AE, Schechtman KB, Piccirillo JF. The efficacy of surgical
modifications of the upper airway in adults with obstructive
sleep apnea syndrome. Sleep 1996;19:15677.
Hessel NS, de Vries N. Results of uvulopalatopharyngoplasty
after diagnostic workup with polysomnography and sleep
endoscopy: a report of 136 snoring patients. Eur Arch
Otorhinolaryngol 2003;260:915 Epub 2002:4.
Rodriguez-Bruno K, Goldberg AN, McCulloch CE, et al. Testretest reliability of drug-induced sleep endoscopy. Otolaryngol
Head Neck Surg 2009;140:64651.
Borowiecki B, Pollak CP, Weitzman ED, et al. Fibro-optic study
of pharyngeal airway during sleep in patients with hypersomnia obstructive sleep-apnea syndrome. Laryngoscope
1978;88:13103.
Maurer JT, Stuck BA, Hein G, et al. Videoendoscopic assessment of uncommon sites of upper airway obstruction during
sleep. Sleep Breath 2004;4:1316.
Rabelo FA, Braga A, Kpper DS, et al. Propofol-induced sleep:
polysomnographic evaluation of patients with obstructive
sleep apnea and controls. Otolaryngol Head Neck Surg
2010;142:21824.
Sadaoka T, Kakitsuba N, Fujiwara Y, et al. The value of sleep
nasoendoscopy in the evaluation of patients with suspected
sleep-related breathing disorders. Clin Otolaryngol Allied Sci
1996;21:4859.
White DP, Cadieux RJ, Lombard RM, et al. The effects of nasal
anesthesia on breathing during sleep. Am Rev Respir Dis
1985;132:9725.
Hillman DR, Walsh JH, Maddison KJ, et al. Evolution of
changes in the upper airway collapsibillity during slow
induction of anesthesia with propofol. Anesthesiology
2009;111:6371.
Fujita S, Conway W, Zorick F, et al. Surgical correction of
anatomic azbnormalities in obstructive sleep apnea syndrome: uvulopalatopharyngoplasty. Otolaryngol Head Neck
Surg 1981;89:92334.
Rodenstein DO, Dooms G, Thomas Y, et al. Pharyngeal shape
and dimensions in healthy subjects, snorers, and patients
with obstructive sleep apnoea. Thorax 1990;45:7227.
Ciscar MA, Juan G, Martnez V, et al. Magnetic resonance
imaging of the pharynx in OSAS patients and healthy
subjects. Eur Respir J 2001;17:7986.
Schellenberg JB, Maislin G, Schwab RJ. Physical findings and
the risk for obstructive sleep apnea. The importance of
oropharyngeal structures. Am J Respir Crit Care Med
2000;162(2 pt 1):7408.
Bachar G, Nageris B, Feinmesser R, et al. Novel grading system
for quantifying upper-airway obstruction on sleep endoscopy.
Lung 2012;190:3138.