/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97076-en.cckm

201606168

page

100

UWHC,UWMF,

Tools,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

High STOP-Bang Score Indicates a High Probability of Obstructive Sleep Apnea

High STOP-Bang Score Indicates a High Probability of Obstructive Sleep Apnea - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


High STOP-Bang score indicates a high probability
of obstructive sleep apnoea
F. Chung
1
*
, R. Subramanyam
1
, P. Liao
1
, E. Sasaki
1
, C. Shapiro
2
and Y. Sun
1
1
Department of Anaesthesia and
2
Department of Psychiatry, Toronto Western Hospital, University Health Network, University of Toronto,
399 Bathurst Street, Toronto, ON, Canada M5T 2S8
*Corresponding author. E-mail: frances.chung@uhn.ca
Editor’s key points
† The authors investigated
the value of STOP-Bang
score in predicting
obstructive sleep apnoea
(OSA) in surgical patients.
† Results from 746 patients
were analysed.
† The odds ratio of
moderate-to-severe OSA
increased with the
increase in the score.
† Importantly, the study
shows the usefulness of
STOP-Bang score in
predicting OSA.
Background. The STOP-Bang questionnaire is used to screen patients for obstructive sleep
apnoea (OSA). We evaluated the association between STOP-Bang scores and the
probability of OSA.
Methods. After Institutional Review Board approval, patients who visited the preoperative
clinics for a scheduled inpatient surgery were approached for informed consent. Patients
answered STOP questionnaire and underwent either laboratory or portable
polysomnography (PSG). PSG recordings were scored manually. The BMI, age, neck
circumference, and gender (Bang) were documented. Over 4 yr, 6369 patients were
approached and 1312 (20.6%) consented. Of them, 930 completed PSG, and 746 patients
with complete data on PSG and STOP-Bang questionnaire were included for data analysis.
Results. The median age of 746 patients was 60 yr, 49% males, BMI 30 kg m
22
, and neck
circumference 39 cm. OSA was present in 68.4% with 29.9% mild, 20.5% moderate, and
18.0% severe OSA. For a STOP-Bang score of 5, the odds ratio (OR) for moderate/severe
and severe OSA was 4.8 and 10.4, respectively. For STOP-Bang 6, the OR for moderate/
severe and severe OSA was 6.3 and 11.6, respectively. For STOP-Bang 7 and 8, the OR for
moderate/severe and severe OSA was 6.9 and 14.9, respectively. The predicted
probabilities for moderate/severe OSA increased from 0.36 to 0.60 as the STOP-Bang
score increased from 3 to 7 and 8.
Conclusions. In the surgical population, a STOP-Bang score of 5–8 identified patients with
high probability of moderate/severe OSA. The STOP-Bang score can help the healthcare
team to stratify patients for unrecognized OSA, practice perioperative precautions, or
triage patients for diagnosis and treatment.
Keywords: mass screening; obstructive/ep (epidemiology); polysomnography; prospective
studies; questionnaires; sleep apnoea; snoring/di (diagnosis); snoring/ep (epidemiology)
Accepted for publication: 23 December 2011
Obstructive sleep apnoea (OSA) is a common medical condi-
tion affecting 2–26% of the general population
1
and can
occur in all age groups.
2
Studies have shown that even asymp-
tomatic OSA is independently associated with an increased
morbidity and mortality.
34
Patients with OSA were found to
have an increase in postoperative complications.
5 – 9
It is, there-
fore, imperative to have an early diagnosis of OSA. However, it is
estimated that 82% of men and 92% of women with
moderate-to-severe sleep apnoea have not been diagnosed.
10
The use of preoperative screening instruments will help to
identify the patients with undiagnosed OSA.
11– 13
The STOP-Bang questionnaire is a scoring model consist-
ing of eight easily administered questions starting with
the acronym STOP-Bang (Appendix) and is scored based on
Yes/No answers (score: 1/0). Thus, the scores range from a
value of 0 to 8. A score of ≥3 has shown a high sensitivity
for detecting OSA: 93% and 100% for moderate and severe
OSA, respectively.
11
Owing to its high sensitivity at a score of ≥3, the STOP-
Bang questionnaire is considered very helpful to rule out
patients having moderate and severe OSA.
11
However, the
specificity at the same cut-off is low: 47% and 37% for mod-
erate and severe OSA, respectively, resulting in fairly high
false-positive rates. The objective of this study is to evaluate
the predictive probabilities for OSA at different scores on the
STOP-Bang questionnaire. We hypothesize that a high STOP-
Bang score indicates a high probability of moderate/severe
OSA.
& The Author [2012]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://
creativecommons.org/licenses/by-nc/2.5/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium,
provided the original work is properly cited.
British Journal of Anaesthesia 108 (5): 768–75 (2012)
Advance Access publication 8 March 2012
.
doi:10.1093/bja/aes022

a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


a
t

U
n
i
v
e
r
s
i
t
y

o
f

W
i
s
c
o
n
s
i
n
-
M
a
d
i
s
o
n

o
n

O
c
t
o
b
e
r

2
5
,

2
0
1
3
h
t
t
p
:
/
/
b
j
a
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m


Methods
The study was conducted in the preoperative clinics of
Toronto Western Hospital and Mount Sinai Hospital,
Toronto, Ontario, Canada. Institutional Review Board
approvals were obtained from both institutions (MSH:
06-0143-E and 07-0183-E; UHN: 06-0135-AE and
07-0515-AE). Patients aged 18 yr or older, who were ASA
I–IV, and were undergoing elective procedures in general
surgery, gynaecology, orthopaedics, urology, plastic surgery,
ophthalmology, or spinal surgery were included in the
screening process and approached for consent by the re-
search assistants for the preoperative polysomnograpy
(PSG). Patients who were unwilling or unable to give informed
consent or patients who were expected to have abnormal
EEG findings (e.g. brain tumour, epilepsy surgery, patients
with deep brain stimulator) were excluded.
All the patients were asked to complete the STOP ques-
tionnaire.
11
Information concerning BMI, age, neck circum-
ference, and gender (Bang) were collected by a research
assistant. In the initial 2 yr period of the study, the patients
were invited to undergo a laboratory PSG. During the subse-
quent 2 yr of the study, the patients underwent a portable
PSG study at home. The results of the PSG were used to
evaluate the various scores of the STOP-Bang questionnaire.
The portable PSG was performed with a level 2 portable
sleep device (Embletta X100) which is shown to be a reliable
alternative for standard PSG in surgical patients.
14
The PSG
recordings were performed at the patients’ home. The
recording montage consisted of two EEG channels (C3 and
C4), electrooculogram (left or right), and chin muscle EMGs.
Thoracic and abdominal respiratory effort bands, body pos-
ition sensors, and pulse oximeter were also used.
The device was attached to patients by a well-trained PSG
technician at their home and the overnight recordings were
unattended. The patients were advised on how to remove
the device which was picked up the next morning from the
patients’ home by the same sleep technician. A certified
PSG technologist who was blinded to the study information
analysed the PSG. The manual scoring was performed using
Somnologia Studio 5.0 as the scoring platform. Manual
scoring was performed according to the Manual of the
American Academy of Sleep Medicine.
15
The laboratory PSG was performed overnight and patients
went to bed at their usual bedtime. A standard EEG montage
consisting of EEG, electrooculogram, submental EMG, and
ECG obtained with surface electrodes were used to collect
the sleep architectural data. A pulse oximeter measured
the oxygen saturation. Additional recordings included the re-
spiratory effort by thoraco-abdominal excursion, respiratory
inductive plethysmography, and oronasal airflow.
A certified polysomnographic technologist scored the
polysomnographic recordings under the supervision of a
sleep physician who assessed and approved the reports.
The technologist was blinded to the results of the STOP-Bang
questionnaire and other clinical information about the
patients. The sleep stages and apnoea–hypopnea index
(AHI) were scored according to the American Academy of
Sleep Medicine Task Force recommendations.
16
The diagnosis of OSA was based on an AHI .5 with frag-
mented sleep and daytime sleepiness. The severity of OSA
with both laboratory and portable PSG was classified based
on the AHI values: .5–15 as mild OSA, .15–30 as moder-
ate OSA, and .30 as severe OSA.
15 16
Statistical analysis
Statistical analyses were performed using SAS version 9.2.
The patient characteristic data are presented with descriptive
statistics; median and inter-quartile range were used for
non-normally distributed continuous data, and frequency
and percentage were used for categorical data. Predicted
probabilities for each score at cut-off points of all OSA
(AHI.5), moderate/severe OSA (AHI.15), and severe OSA
(AHI.30) were calculated using logistic regression, and
plotted. The probability and its 95% confidence interval
(95% CI) were calculated for each score. The STOP-Bang
scores of 7 and 8 were combined due to the small number
of patients with either score. A similar strategy was followed
with scores 0, 1, and 2.
To assess the performance of the STOP-Bang question-
naire, multiple 2×2 contingency tables were used to calcu-
late sensitivity, specificity, positive predictive values (PPVs),
and negative predictive values (NPVs) for each score. The re-
sponse was dichotomized using all OSA (AHI.5), moderate/
severe OSA (AHI.15), and severe OSA (AHI.30) as the
cut-offs. The area under the receiver operating curves was
calculated using logistic regression to assess the diagnostic
ability of the STOP-Bang questionnaire.
Multinomial logistic regression was used to compare the
severity of the AHI with the STOP-Bang questionnaire score.
For the dependent variable, an AHI ≤5 was classified as
non-OSA and was used as the reference. For the independent
variable, patients who scored 0, 1, or 2 were grouped as the
reference. Odds ratios (ORs) and 95% confidence intervals of
each STOP-Bang score group (3, 4, 5, 6, 7, and 8) at different
AHI cut-offs were calculated.
Results
A total of 6369 patients were approached for consent and
screened for OSA by the STOP-Bang questionnaire. Of the
2870 patients screened and invited for laboratory PSG, 414
(14.4%) patients gave consent. Of the 3499 patients
screened and invited for portable PSG, 898 (25.7%) patients
gave consent. Laboratory PSG was completed by 219
patients, and 711 patients completed portable PSG. Of the
930 patients who completed the PSG, 212 patients with a la-
boratory PSG and 534 patients with a portable PSG answered
all of the items in the STOP questionnaire and had complete
documentation of BMI, age, gender, and neck circumference.
These 746 patients were used for the analysis (Fig. 1).
The summary of age, gender, BMI, and neck circumfer-
ence of the different patient groups is shown in Table 1.
The patient characteristics were similar between the 930
STOP-Bang score for predicting OSA BJA
769

patients who underwent a PSG and the 5439 patients who
did not undergo a PSG due to the reasons of no consent or
no show. The 184 patients, who underwent a PSG but did
not complete all the elements of the STOP-Bang question-
naire, were excluded from the analysis set. Patient character-
istics other than the neck circumference were similar
between the 184 patients excluded from the analysis set
and the 746 patients used for the analysis.
Of the 746 patients used for analysis, there were 510
(68.4%), 287 (38.5%), and 134 (18.0%) patients who had
OSA (AHI.5), moderate/severe OSA (AHI.15), and severe
OSA (AHI.30), respectively. The distribution of each of the
STOP-Bang scores is detailed in Figure 2. Most patients had
a STOP-Bang score of 3 (22.9%) and 4 (22.3%).
The area under the receiver operating curves was 0.65 (95%
CI: 0.61–0.70), 0.67 (95% CI: 0.63–0.70), and 0.71 (95% CI:
0.66–0.75) for all OSA, moderate/severe OSA, and severe OSA,
respectively. Although the areas under the receiver operating
curves do not show perfect discrimination, the confidence inter-
vals do not include 0.5, confirming the diagnostic ability of the
STOP-Bang questionnaire. The STOP-Bang questionnaire had
the best discrimination with severe OSA.
For a STOP-Bang score of 5, the OR for moderate/severe was
4.8 (95% CI: 2.8–8.0) and for severe OSA was 10.4 (95% CI: 4.5–
24.3). For a STOP-Bang score of 6, the OR for moderate/severe
was 6.3 (95% CI: 3.4–11.7) and for severe OSA was 11.6 (95%
CI: 4.6–28.7). For a STOP-Bang score of 7 and 8, the OR for mod-
erate/severe was 6.9 (95% CI: 3.3–14.3) and for severe OSA was
14.9 (95% CI: 5.6–39.6) (Table 2).
The sensitivity, specificity, PPVs, and NPVs for all OSA, moder-
ate/severe OSA, and severe OSA are summarized in Table 3.As
the STOP-Bang score increased from 3 to 8, the sensitivity
Preoperative patients screened with STOP-Bang questionnaire
during first 2 yr (n = 2870)
Consented for laboratory PSG
n = 414 (14.4%)
PSG not done
n = 195 (47.1%)
STOP-Bang incomplete
n = 7 (3.2%)
STOP-Bang complete
n = 212 (96.8%)
STOP-Bang complete
n = 746 (80.2%)
STOP-Bang complete
n = 534 (75.1%)
STOP-Bang incomplete
n = 177 (24.9%)
PSG done
n = 219 (52.9%)
PSG done
n = 711 (79.2%)
PSG not done
n = 187 (20.8%)
Consented for portable PSG
n = 898 (25.7%)
Preoperative patients screened with STOP-Bang questionnaire
during next 2 yr (n = 3499)
Fig 1 Screening flow chart of patients. PSG, polysomnography.
Table 1 Patient characteristics. Data shown as median (inter-quartile range) or number with percentage in parenthesis. *n¼1634.

STOP-Bang
incomplete due to missing data
PSG not done PSG done
Total STOP-Bang complete STOP-Bang incomplete
n 5439 930 746 184
Gender [male/female] 2504/2935 (46/54) 445/485 (48/52) 365/381 (49/51) 80/104 (44/56)
Age (yr) 58 (47–69) 60 (52–69) 60 (51–68) 61 (54–69)
Neck circumference (cm) 38 (35–40)* 39 (36–42) 39 (36–42)
—†
BMI (kg m
22
) 27 (24–31) 30 (26–34) 30 (26–35) 30 (26–34)
BJA Chung et al.
770

decreased from 68.4% to 0.4% for moderate/severe OSA
patients, and 94.8% to 0% for severe OSA patients. When the
STOP-Bang score was 5, the specificity for moderate/severe
OSA was 56.1% and for severe OSA was 74.2%.
The predicted probabilities of having OSA, moderate/
severe OSA, or severe OSA are shown in Table 4. The probabil-
ities of having OSA were greater as the STOP-Bang score
increased. This trend was the same across the groups of all
OSA, moderate/severe OSA, and severe OSA (Fig. 3). As the
STOP-Bang score increased from 0–2 to 7 and 8, the prob-
ability of having OSA, moderate/severe OSA, and severe
OSA increased from 46% (95% CI: 39–53%) to 86% (95%
CI: 72–93%), 18% (95% CI: 13–24%) to 60% (95% CI: 44–
73%), and 4% (95% CI: 2–8%) to 38% (95% CI: 29–53%),
respectively (Table 4).
Discussion
The results of the study showed that with an increase in the
STOP-Bang score, there was a corresponding increase in the
predicted probability, OR, and specificity for having OSA,
moderate/severe, and severe OSA. This was accompanied
by a progressive decrease in sensitivity. For a STOP-Bang
score of 5, the OR for moderate/severe and severe OSA was
4.8 and 10.4, respectively. For STOP-Bang 7 and 8, the OR
for moderate/severe and severe OSA was 6.9 and 14.9, re-
spectively. The STOP-Bang questionnaire was initially intro-
duced as a scoring model for the preoperative patients.
11
The results from this study further validated the value of
STOP-Bang questionnaire as a screening tool in surgical
patients. The association between the STOP-Bang score and
the probability of OSA would provide the perioperative care
team a useful tool to stratify patients for unrecognized OSA
and triage patients for diagnosis and treatment.
It is estimated that nearly 80% of men and 93% of
women with moderate-to-severe sleep apnoea are undiag-
nosed,
17
which poses a variety of problems for anaesthesiol-
ogists. OSA patients are known to have a higher incidence of
difficult intubation,
18
postoperative complications,
19 20
increased intensive care unit admissions,
7
and greater dur-
ation of hospital stay.
21
Memtsoudis and colleagues
9
found
that OSA was associated with a significantly higher incidence
of pulmonary complications. However, no association
between postoperative complication and OSA severity was
found in obese patients undergoing bariatric surgery.
22
This
may be due to the fact that most patients with OSA (93%)
received perioperative positive airway pressure therapy, and
all patients were closely monitored after operation with
pulse oximetry on either regular nursing floors or in intensive
or intermediate care units.
22
Recently, a Canadian publica-
tion
23
and the American Society of Anesthesiologists guide-
lines
24
both stressed the importance of preoperative
diagnosis and perioperative management of OSA patients
to avoid postoperative complications.
To identify patients at high risk of OSA is the first step for
the perioperative care of OSA patients and prevention of
adverse events. Although no test or parameter has been
widely accepted as a tool to identify the OSA patients who
are particularly at risk for severe postoperative pulmonary
adverse events, a recent study does show that patients clas-
sified as STOP-Bang high risk had an increased incidence of
postoperative complications.
25
The STOP-Bang questionnaire is concise and easy to use.
It consisted of eight questions with a yes or no answer
25
20
15
10
P
e
r
c
e
n
t
a
g
e

o
f

f
r
e
q
u
e
n
c
y
5
0
01234
STOP-Bang score
5678
Fig 2 Distribution of patients according to their STOP-Bang score.
Table 2 ORs (95% CIs) of different STOP-Bang scores for OSA at different AHI cut-offs. AHI, apnoea–hypopnoea index; OSA, obstructive sleep
apnoea; Mod/Sev OSA, moderate/severe OSA
STOP-Bang score ORs for OSA at different AHI cut-offs
All OSA (AHI>5) Mod/Sev OSA (AHI>15) Severe OSA (AHI>30)
Score 3 vs Score 0–2 3.01 (1.92–4.70) 2.59 (1.58–4.27) 3.56 (1.48–8.58)
Score 4 vs Score 0–2 3.15 (2.01–4.96) 3.33 (2.03–5.46) 5.33 (2.27–12.50)
Score 5 vs Score 0–2 3.98 (2.38–6.66) 4.75 (2.81–8.03) 10.39 (4.45–24.26)
Score 6 vs Score 0–2 4.52 (2.34–8.74) 6.29 (3.39–11.66) 11.55 (4.64–28.71)
Score 7 and 8 vs Score 0–2 7.04 (2.82–17.55) 6.88 (3.32–14.25) 14.86 (5.58–39.56)
STOP-Bang score for predicting OSA BJA
771

and has been used as a preoperative screening tool for
OSA.
12 26 – 28
Recently, the STOP-Bang questionnaire has been validated in
two studies of patients referred to the sleep clinic.
29–30
Farney’s
study showed that the STOP-Bang questionnaire can be used to
estimate the probabilities of no, mild, moderate, and severe
OSA. The greater the cumulative score of risk factors as reflected
by the STOP-Bang model, the greater the probability of severe
OSA.
29
With any score .4, the probability of having severe
OSA increases continuously. With a score of 8, the probability
of severe OSA was 81.9%.
29
Although our results also showed
a similar association between the probabilities of having
severe OSA and the score on STOP-Bang, we did not see such
a high probability of severe OSA with a higher STOP-Bang
score. This may be due to the difference in the study population.
Our patients were preoperative patients. The patients in
Farney’s study were the patients referred to sleep clinic popula-
tion which have a high prevalence of severe OSA.
Table 3 Predictive parameters of different STOP-Bang score cut-offs. *Percentage out of the 746 patients (n, number of patients in the AHI group
who scored the STOP-Bang score indicated or higher). AHI, apnoea–hypopnoea index; PPV, positive predictive value; NPV, negative predictive
value
STOP-Bang score cut-off n (%)* Sensitivity (%) Specificity (%) PPV (%) NPV (%)
All OSA (AHI.5)
1 504 (67.6) 98.8 2.5 68.7 50.0
2 488 (65.4) 95.7 17.8 71.6 65.6
3 429 (57.5) 84.1 40.3 75.3 54.0
4 306 (41.1) 60.0 60.6 76.7 41.2
5 185 (24.8) 36.3 79.7 79.4 36.7
6 90 (12.1) 17.7 91.5 81.8 34.0
7 36 (4.8) 7.1 97.5 85.7 32.7
8 4 (0.5) 0.8 98.7 57.1 31.5
Moderate/severe OSA (AHI.15)
1 285 (38.2) 97.8 0.7 16.7 61.2
2 283 (37.9) 86.9 1.4 6.3 58.5
3 256 (34.3) 68.4 10.8 17.6 55.1
4 195 (26.1) 44.4 32.1 26.5 51.1
5 126 (16.9) 23.3 56.1 31.4 45.9
6 64 (8.6) 10.0 77.7 35.1 41.8
7 25 (3.4) 3.7 91.3 37.2 40.5
8 1 (0.1) 0.4 98.7 14.3 61.3
Severe OSA (AHI.30)
1 134 (18.0) 100.0 2.0 18.3 100.0
2 134 (18.0) 100.0 10.5 19.7 100.0
3 127 (17.0) 94.8 27.6 22.3 96.0
4 105 (14.1) 78.4 52.0 26.3 91.6
5 75 (10.1) 56.0 74.2 32.2 88.5
6 38 (5.1) 28.4 88.2 34.6 84.9
7 16 (2.1) 11.9 95.8 38.1 83.2
8 0 (0) 0 98.9 0 81.9
Table 4 Predicted probabilities per score for all OSA, moderate/severe OSA, and severe OSA. CI, confidence interval; AHI, apnoea–hypopnoea
index; n, number; Mod/Sev OSA, moderate/severe OSA
Score All OSA (AHI>5) Mod/Sev OSA (AHI>15) Severe OSA (AHI>30)
n Probability (95% CI) n Probability (95% CI) n Probability (95% CI)
0–2 81 0.46 (0.39–0.53) 31 0.18 (0.13–0.24) 7 0.04 (0.02–0.08)
3 123 0.72 (0.65–0.78) 61 0.36 (0.29–0.43) 22 0.13 (0.09–0.19)
4 121 0.73 (0.66–0.79) 69 0.42 (0.34–0.49) 30 0.18 (0.13–0.25)
5 95 0.77 (0.69–0.84) 62 0.50 (0.42–0.59) 37 0.30 (0.23–0.39)
6 54 0.79 (0.68–0.87) 39 0.57 (0.45–0.69) 22 0.32 (0.22–0.44)
7 and 8 36 0.86 (0.72–0.93) 25 0.60 (0.44–0.73) 16 0.38 (0.29–0.53)
BJA Chung et al.
772

Since a STOP-Bang score of ≥3 demonstrated a very high
sensitivity and NPV for moderate/severe OSA, this cut-off
may be good for a surgical population with high OSA preva-
lence such as bariatric surgical patients. We would be confi-
dent in excluding the possibility of moderate/severe or severe
OSA in patients with a STOP-Bang score of 0–2. On the other
hand, the patients with a STOP-Bang score of 5–8 have a
high specificity to detect moderate and severe OSA. These
scores may be useful in the general patient population
which has a low OSA prevalence to reduce false-positive
rate. It enables identification of those patients most in
need of urgent evaluation and to exclude patients from pos-
sible harm due to unrecognized sleep apnoea.
29
However,
further research is needed so that the STOP-Bang can be vali-
dated in the different clinical populations.
It is a challenge to establish a practical perioperative care
pathway for OSA patients. It is not known whether patients
with a STOP-Bang score of 5–8 with co-morbidities having
major surgery would benefit from sleep medicine referral,
expedited polysomnography (PSG), and continuous positive
airway pressure (CPAP) treatment. There have been no
studies in the literature to prove that preoperative PSG is of
benefit to the surgical patients with suspected OSA.
Overnight-attended PSG is an old standard in the diagnosis
of OSA, but it is expensive and cumbersome. Often, there is
a timeline for patients undergoing surgery. Portable home-
based monitoring devices or single channel recording such
as nocturnal oximetry might be used as an alternative for
the diagnosis of OSA in patients with high probability of
moderate-to-severe OSA.
31
Thus, a combination of STOP-
BANG questionnaire to identify patients at risk of OSA and
nocturnal oximetry may allow for a more rapid diagnosis of
OSA. Alternatively, in the patients classified as high risk of
OSA by the STOP-Bang questionnaire, especially those with
a STOP-Bang score of ≥5, practicing perioperative precau-
tions (preparation for possible difficult intubation, using
short-acting anaesthesia agents, adequate neuromuscular
blocking agent reversal, and use of CPAP after operation)
and postoperative monitoring is helpful to prevent adverse
outcomes.
23 24 32
If patients get earlier treatment for their
OSA because of screening in preoperative clinics, there may
be long-term health benefits for the patients, besides redu-
cing risk for OSA-related perioperative adverse event. More
collaboration between anaesthesiologists, surgeons, and
sleep physicians is needed.
There are a few limitations with our study. The study could
be criticized because PSG was performed with both the stand-
ard PSG in the laboratory and the portable PSG at home.
Embletta X-100 is a level 2 diagnostic device for SDB. When in-
stalled by a well-trained technician and scored by a certified
PSG technologist, parameters measuring sleep-disordered
breathing and sleep architecture from Embletta X-100 were
comparable with in-laboratory standard PSG.
15
Although
home monitoring is validated
15 33
and all PSG recordings
were scored by certified PSG technologists, some inconsistency
in the two approaches may exist. Secondly, the study popula-
tion is surgical patients referred to preoperative clinics. These
results may not be applicable to other patient populations.
Further validation in the different population, especially the
general population, needs to be done. Also, there may be a se-
lection bias involved in the patient recruiting process, the sub-
jects having some OSA-related symptoms might be more
motivated to give consent to this study. Finally, like all other
screening studies for sleep apnoea, central apnoeas were
also not evaluated separately in the report.
In conclusion, the predicted probabilities were greater as
the STOP-Bang score increased, showing that patients had
a greater probability of having OSA when they scored
higher on the STOP-Bang questionnaire. A STOP-Bang score
of ,3 will allow the healthcare team to rule out patients
who do not have OSA. A STOP-Bang score of 5–8 will allow
the team to identify patients with increased probability of
moderate/severe OSA. The STOP-Bang score can help the
healthcare team to stratify patients for unrecognized OSA,
practice perioperative precautions, or triage patients for
diagnosis and treatment.
Authors’ roles
F.C. is the principal investigator. F.C. helped design the study,
conduct the study, and write the manuscript and had
the overall responsibility for the study. R.S. helped design
the study, and write the manuscript. P.L. helped design the
study, conduct the study, and write the manuscript. E.S. ana-
lysed the data and helped write the manuscript. C.S. helped
1.0
0.0
234
STOP-Bang score
567
0.1
0.2
0.3
0.4
0.5
O
S
A

p
r
o
b
a
b
i
l
i
t
y
0.6
0.7
0.8
0.9
AHI cut-off: AHI>5 AHI>15 AHI>30
Fig 3 Plot of predicted probabilities for AHI cut-offs of .5, .15,
and .30 with the corresponding STOP-Bang score. The vertical
bars indicate the 95% confidence intervals. STOP-Bang scores
of 0, 1, and 2 are grouped together and are shown as score
2. Scores 7 and 8 are grouped together and is shown as score
7. As the STOP-Bang scores increased, the predicted probabilities
were greater. AHI, apnoea–hypopnoea index.
STOP-Bang score for predicting OSA BJA
773

design the study and supervised sleep studies. Y.S. was re-
sponsible for the scoring of PSG.
Registration Site and Number: Mount Sinai Hospital,
Toronto, Canada: 06-0143-E and 07-0183-E; Toronto
Western Hospital: 06-0135-AE and 07-0515-AE.
Acknowledgements
We acknowledged the help of Santhira Vairavanathan, MBBS,
Sazzadul Islam, MSc, Hisham Elsaid, MD, Babak Amirshahi, MD,
and Hoda Fazel, MD, for the help in the conduct of the study.
Declaration of interest
None declared.
Funding
This work was supported by Physicians Services Incorporated
Foundation, University Health Network Foundation, ResMed
Foundation, Respironic Foundation and Department of Anes-
thesia, University Health Network-Mount Sinai Hospital, Uni-
versity of Toronto.
Appendix
STOP-Bang questionnaire
11
1. Snoring: Do you snore loudly (loud enough to be heard through closed
doors)?
Yes No
2. Tired: Do you often feel tired, fatigued, or sleepy during daytime?
Yes No
3. Observed: Has anyone observed you stop breathing during your sleep?
Yes No
4. Blood pressure: Do you have or are you being treated for high blood
pressure?
Yes No
5. BMI: BMI more than 35 kg m
22
?
Yes No
6. Age: Age over 50 yr old?
Yes No
7. Neck circumference: Neck circumference .40 cm?
Yes No
8. Gender: Male?
Yes No
High risk of OSA: Yes to ≥3 questions.
Low risk of OSA: Yes to ,3 questions.
Questionnaire reproduced from Chung et al.
11
with permis-
sion from Wolters Kluwer Health.
References
1 Young T, Hutton R, Finn L, Badr S, Palta M. The gender bias in sleep
apnea diagnosis: are women missed because they have different
symptoms? Arch Intern Med 1996; 156: 2445–51
2 Lettieri CJ, Eliasson AH, Andrada T, Khramtsov A, Raphaelson M,
Kristo DA. Obstructive sleep apnea syndrome: are we
missing an at-risk population? J Clin Sleep Med 2005; 1: 381–5
3 Young T, Finn L, Peppard PE, et al. Sleep disordered breathing and
mortality: eighteen year follow up of the Wisconsin Sleep Cohort.
Sleep 2008; 31: 1071–8
4 Marshall NS, Wong KK, Liu PY, Cullen SR, Knuiman MW, Grunstein RR.
Sleep apnea as an independent risk factor for all cause mortality: the
Busselton Health Study. Sleep 2008; 31:1079–85
5 Kaw R, Michota F, Jaffer A, Ghamande S, Auckley D, Golish J. Un-
recognized sleep apnea in the surgical patient: implications for
the perioperative setting. Chest 2006; 129: 198–205
6 Liao P, Yegneswaran B, Vairavanathan S, Zilberman P, Chung F.
Postoperative complications in patients with obstructive sleep
apnea: a retrospective matched cohort study. Can J Anaesth
2009; 56: 819–28
7 Gupta RM, Parvizi J, Hanssen AD, Gay PC. Postoperative complica-
tions in patients with obstructive sleep apnea syndrome undergo-
ing hip or knee replacement: a case–control study. Mayo Clin Proc
2001; 76: 897–905
8 Gallagher SF, Haines KL, Osterlund LG, Mullen M, Downs JB. Post-
operative hypoxemia: common, undetected, and unsuspected
after bariatric surgery. J Surg Res 2010; 159: 622–6
9 Memtsoudis S, Liu SS, Ma Y, et al. Perioperative pulmonary out-
comes in patients with sleep apnea after noncardiac surgery.
Anesth Analg 2011; 112: 113–21
10 Ancoli-Israel S, Kripke DF, Klauber MR, et al. Sleep-disordered
breathing in community-dwelling elderly. Sleep 1991; 14: 486–95
11 Chung F, Yegneswaran B, Liao P, et al. STOP questionnaire: a tool
to screen patients for obstructive sleep apnea. Anesthesiology
2008; 108: 812–21
12 Chung F, Yegneswaran B, Liao P, et al. Validation of the Berlin
questionnaire and American Society of Anesthesiologists check-
list as screening tools for obstructive sleep apnea in surgical
patients. Anesthesiology 2008; 108: 822–30
13 Finkel KJ, Searleman AC, Tymkew H, et al. Prevalence of undiag-
nosed obstructive sleep apnea among adult surgical patients in
an academic medical center. Sleep Med 2009; 10: 753–8
14 Chung F, Liao P, Sun Y, et al. Perioperative practical experiences in
using a level 2 portable polysomnography. Sleep Breath 2010,
doi:10.1007/s11325-010-0340-9
15 Iber C, Ancoli-Israel S, Cheeson A, Quan SF. The AASM Manual for
the Scoring of Sleep and Associated Events, Rules, Terminology
and Technical Specifications. Westchester, IL: American
Academy of Sleep Medicine, 2007. Available from http://www
.aasmnet.org/store/product.aspx?pid=176 (accessed 23 January
2011)
16 Sleep-related breathing disorders in adults: recommendations for
syndrome definition and measurement techniques in clinical re-
search. The Report of an American Academy of Sleep Medicine
Task Force. Sleep 1999; 22: 667–89
17 Young T, Evans L, Finn L, Palta M. Estimation of the clinically diag-
nosed proportion of sleep apnea syndrome in middle-aged men
and women. Sleep 1997; 20: 705–6
18 Hiremath AS, Hillman DR, James AL, Noffsinger WJ, Platt PR,
Singer SL. Relationship between difficult tracheal intubation and
obstructive sleep apnoea. Br J Anaesth 1998; 80: 606–11
19 Reeder MK, Goldman MD, Loh L, Muir AD, Casey KR, Lehane JR.
Late postoperative nocturnal dips in oxygen saturation in
patients undergoing major abdominal vascular surgery. Predict-
ive value of pre-operative overnight pulse oximetry. Anaesthesia
1992; 47: 110–5
20 Rosenberg J, Wildschiødtz G, Pedersen MH, von Jessen F, Kehlet H.
Late postoperative nocturnal episodic hypoxaemia and asso-
ciated sleep pattern. Br J Anaesth 1994; 72: 145–50
BJA Chung et al.
774

21 Ballantyne GH, Svahn J, Capella RF, et al. Predictors of prolonged hos-
pital stay following open and laparoscopic gastric bypass for morbid
obesity: body mass index, length of surgery, sleep apnea, asthma,
and the metabolic syndrome. Obes Surg 2004; 14:1042–50
22 Weingarten TN, Flores AS, McKenzie JA, et al. Obstructive sleep
apnoea and perioperative complications in bariatric patients. Br
J Anaesth 2011; 106: 131–9
23 Seet E, Chung F. Management of sleep apnea in adults—func-
tional algorithms for the perioperative period. Can J Anaesth
2010; 57: 849–64
24 Gross JB, Bachenberg KL, Benumof JL, et al. American Society of
Anesthesiologists Task Force on Perioperative Management. Practice
guidelines for the perioperative management of patients with ob-
structive sleep apnea: a report by the American Society of Anesthe-
siologists Task Force on Perioperative Management of patients
with obstructive sleep apnea. Anesthesiology 2006; 104:1081–93
25 Vasu TS, Doghramji K, Cavallazzi R, et al. Obstructive sleep apnea
syndrome and postoperative complications: clinical use of the
STOP-BANG questionnaire. Arch Otolaryngol Head Neck Surg
2010; 136: 1020–4
26 Ramachandran SK, Josephs LA. A meta-analysis of clinical
screening tests for obstructive sleep apnea. Anesthesiology
2009; 110: 928–39
27 Ong TH, Raudha S, Fook-Chong S, Lew N, Hsu AA. Simplifying
STOP-BANG: use of a simple questionnaire to screen for OSA in
an Asian population. Sleep Breath 2010; 14: 371–6
28 Gafsou B, Marsac L, Fournier JL, Be´loucif S, Baillard C. Validation of
the STOP-Bang questionnaire as screening tools for obstructive
sleep apnea in patients scheduled for bariatric surgery: 1AP3-5.
Eur J Anesthesiol 2010; 27:13
29 Farney RJ, Walker BS, Farney RM, Snow GL, Walker JM. The
STOP-Bang equivalent model and prediction of severity of
obstructive sleep apnea: relation to polysomnographic mea-
surements of the apnea/hyponea index. J Clin Sleep Med
2011; 7: 459–65
30 Silva GE, Vana KD, Goodwin JL, Sherrill DL, Quan SF.
Identifications of patients with sleep disorderd breathing:
comparing the four-variable screening tool, STOP, STOP-Bang
and Epworth Sleepiness Score. J Clin Sleep Med 2011; 7:
467–72
31 Centers for Medicare and Medicaid Services. Decision memo
for sleep testing for obstructive sleep apnea. Available from
http://www.cms.hhs.gov/mcd/viewdecisionmemo.asp?id=227
(accessed October 2011)
32 Gali B, Whalen FX, Schroeder DR, Gay PC, Plevak DJ. Identification
of patients at risk for postoperative respiratory complications
using a preoperative obstructive sleep apnea screening tool
and postanesthesia care assessment. Anesthesiology 2009; 110:
869–77
33 Santos-Silva R, Sartori DE, Truksinas V, et al. Validation of a port-
able monitoring system for the diagnosis of obstructive sleep
apnea syndrome. Sleep 2009; 32: 629–36
STOP-Bang score for predicting OSA BJA
775