Management of Obstructive Sleep Apnea in Adults: A Clinical Practice
Guideline From the American College of Physicians
Amir Qaseem, MD, PhD, MHA; Jon-Erik C. Holty, MD, MS; Douglas K. Owens, MD, MS; Paul Dallas, MD; Melissa Starkey, PhD; and
Paul Shekelle, MD, PhD, for the Clinical Guidelines Committee of the American College of Physicians*
Description: The American College of Physicians (ACP) developed
this guideline to present the evidence and provide clinical recom-
mendations on the management of obstructive sleep apnea (OSA)
in adults.
Methods: This guideline is based on published literature from 1966
to September 2010 that was identified by using MEDLINE, the
Cochrane Central Register of Controlled Trials, and the Cochrane
Database of Systematic Reviews. A supplemental MEDLINE search
identified additional articles through October 2012. Searches were
limited to English-language publications. The clinical outcomes eval-
uated for this guideline included cardiovascular disease (such as
heart failure, hypertension, stroke, and myocardial infarction), type
2 diabetes, death, sleep study measures (such as the Apnea–
Hypopnea Index), measures of cardiovascular status (such as blood
pressure), measures of diabetes status (such as hemoglobin A1c
levels), and quality of life. This guideline grades the evidence and
recommendations using ACP’s clinical practice guidelines grading
system.
Recommendation 1: ACP recommends that all overweight and
obese patients diagnosed with OSA should be encouraged to lose
weight. (Grade: strong recommendation; low-quality evidence)
Recommendation 2: ACP recommends continuous positive airway
pressure treatment as initial therapy for patients diagnosed with
OSA. (Grade: strong recommendation; moderate-quality evidence)
Recommendation 3: ACP recommends mandibular advancement
devices as an alternative therapy to continuous positive airway
pressure treatment for patients diagnosed with OSA who prefer
mandibular advancement devices or for those with adverse effects
associated with continuous positive airway pressure treatment.
(Grade: weak recommendation; low-quality evidence)
For author affiliations, see end of text.
This article was published at www.annals.org on 24 September 2013.
Obstructive sleep apnea (OSA) is a common disorderthat affects persons in all age groups, especially
middle-aged and elderly persons (Table 1). Evidence shows
that OSA rates are increasing, and this is probably because
of escalating obesity rates (1). The goal of OSA treatment
is to alleviate airway obstruction during sleep. The stan-
dard first-line OSA treatment involves continuous positive
airway pressure (CPAP) devices, which deliver compressed
air into the airway to keep it open. Many patients do not
tolerate CPAP and often do not adhere to the instructions
for many reasons, including discomfort, skin irritation,
noise, and claustrophobia (2–4). To improve adherence,
many technological modifications have been made to
CPAP devices (mostly in alterations of when air pressure is
delivered), although the utility of the modified devices is
unknown. Because adherence is often an issue in OSA
treatment, additional patient education or interventions
may be warranted. Dental or mandibular advancement de-
vices (MADs) that are worn while the patient is sleeping
have been used to treat OSA. Alternative therapeutic strat-
egies include surgical interventions to remove obstructive
tissue, positional therapy, pharmacologic treatment, and
weight-loss interventions for obese patients.
All interventions have the potential for adverse effects,
and the purpose of this American College of Physicians
(ACP) guideline is to present information on both the
benefits and harms of interventions to assess the net bene-
fits of available treatments. The target audience for this
guideline includes all clinicians, and the target patient pop-
ulation comprises all adults with OSA.
METHODS
This guideline is based on a systematic evidence review
sponsored by the Agency for Healthcare Research and
* This paper, written by Amir Qaseem, MD, PhD, MHA; Jon-Erik C. Holty, MD, MS; Douglas K. Owens, MD, MS; Paul Dallas, MD; Melissa Starkey, PhD; and Paul Shekelle, MD,
PhD, was developed for the Clinical Guidelines Committee of the American College of Physicians. Individuals who served on the Clinical Guidelines Committee from initiation of the
project until its approval were Paul Shekelle, MD, PhD (Chair); Roger Chou, MD; Molly Cooke, MD; Paul Dallas, MD; Thomas D. Denberg, MD, PhD; Nick Fitterman, MD; Mary
Ann Forciea, MD; Robert H. Hopkins Jr., MD; Linda L. Humphrey, MD, MPH; Tanveer P. Mir, MD; Douglas K. Owens, MD, MS; Holger J. Schu¨nemann, MD, PhD; Donna E.
Sweet, MD; David S. Weinberg, MD, MSc; and Timothy Wilt, MD, MPH. Approved by the ACP Board of Regents on 17 November 2012.
See also:
Web-Only
CME quiz
Clinical Guideline
© 2013 American College of Physicians
This online-first version will be replaced with a final version when it is included in an issue. The final version may differ in small ways.
Ann Intern Med. 2013;159. www.annals.org
Downloaded From: http://annals.org/ on 09/24/2013 www.medlive.cn
Quality (AHRQ) (5) that addressed the following key
questions related to OSA management:
1. What is the comparative effect of different treat-
ments for OSA in adults?
a. Does the comparative effectiveness of treatments
vary based on presenting patient characteristics, OSA se-
verity, or other pretreatment factors? Are any of these char-
acteristics or factors predictive of treatment success?
i. Characteristics: Age, sex, race, weight, bed partner,
airway, other physical characteristics, and specific comor-
bid conditions.
ii. Obstructive sleep apnea severity or characteristics:
Baseline questionnaire (and similar tools) results, formal
testing results (including hypoxemia levels), baseline qual-
ity of life, positional dependency.
iii. Other: Specific symptoms.
2. In patients with OSA who are prescribed nonsurgi-
cal treatments, what are the associations of pretreatment,
patient-level characteristics with treatment adherence?
3. What is the effect of interventions to improve ad-
herence to device use (positive airway pressure, oral appli-
ances, and positional therapy) on clinical and intermediate
outcomes?
The Tufts Evidence-based Practice Center conducted
the systematic evidence review. The literature search in-
cluded studies identified using MEDLINE (1966 to Sep-
tember 2010), the Cochrane Central Register of Con-
trolled Trials, and the Cochrane Database of Systematic
Reviews and included peer-reviewed studies on adult hu-
man patients published in English (6). Further details
about the methods and inclusion and exclusion criteria
applied in the evidence review are available in the full
AHRQ report (5). No randomized, controlled trial (RCT)
on OSA treatment with regard to mortality outcomes was
identified. The ACP supplemented the AHRQ review
(MEDLINE search, 1946 to October 2012) to identify
English-language observational studies in humans report-
ing death or cardiovascular or cerebrovascular illness asso-
ciated with OSA treatment strategies (that is, CPAP, sur-
gery, or MADs), as well as more recent relevant RCTs. To
guide our recommendations, we prioritized outcomes on
the basis of clinical importance, starting with death and
including cardiovascular outcomes. In the absence of sta-
tistically significant effects on clinical outcomes, we con-
sidered symptoms (such as Epworth Sleepiness Scale [ESS]
scores) and other physiologic measures (such as the Apnea–
Hypopnea Index [AHI]). This guideline rates the evidence
and recommendations using ACP’s grading system, which
is based on the GRADE (Grading of Recommendations
Assessment, Development, and Evaluation) system (Table
2). Details of ACP’s guideline development process can be
found in the methods paper (7).
COMPARATIVE EFFECTIVENESS OF OSA TREATMENTS
Evidence on the comparative effectiveness of OSA
treatments is summarized in the Appendix Table (available
at www.annals.org).
CPAP for OSA Treatment
Although moderate-quality evidence showed that
CPAP improves sleep measures compared with control or
sham devices in patients with at least moderate OSA (AHI
score �15 events/h), there was little or no evidence on the
effects of CPAP on other important clinical outcomes.
CPAP Versus Control Treatment
Twenty-two studies (reported in 23 articles) compared
CPAP with control treatment, including patients with
mean baseline AHI scores between 10 and 65 events per
hour. These studies apply to many patients with OSA, and
follow-up ranged from 1 to 24 months. Moderate-quality
evidence showed that CPAP reduced AHI scores, im-
proved ESS scores, reduced arousal index scores, and raised
the minimum oxygen saturation compared with control
Table 1. Terms and Definitions Related to OSA
Term Definition
Apnea Complete airflow cessation during sleep
Hypopnea Reduced airflow during sleep
AHI A measure of the number of apnea and hypopnea episodes
per hour of monitored sleep. According to the American
Academy of Sleep Medicine, an OSA diagnosis is defined
by �15 events/h (with or without OSA symptoms) or
�5 events/h with OSA symptoms. Severity of OSA
classified according to AHI is defined as mild if 5–14
events/h, moderate if 15–30 events/h, and severe if
�30 events/h.
ESS A self-administered questionnaire in which patients rate
their chances of dozing in various situations.
Arousal index The frequency of arousals per hour of sleep, measured by
electroencephalography.
AHI � Apnea–Hypopnea Index; ESS � Epworth Sleepiness Scale; OSA � ob-
structive sleep apnea.
Table 2. The American College of Physicians’ Guideline
Grading System*
Quality of
Evidence
Strength of Recommendation
Benefits Clearly Outweigh
Risks and Burden or Risks
and Burden Clearly
Outweigh Benefits
Benefits Finely Balanced
With Risks and Burden
High Strong Weak
Moderate Strong Weak
Low Strong Weak
Insufficient evidence to determine net benefits or risks
* Adopted from the classification developed by the GRADE (Grading of Recom-
mendations Assessment, Development, and Evaluation) workgroup.
Clinical Guideline Clinical Practice Guideline on Management of OSA in Adults
This online-first version will be replaced with a final version when it is included in an issue. The final version may differ in small ways.
Annals of Internal Medicine www.annals.org
Downloaded From: http://annals.org/ on 09/24/2013 www.medlive.cn
treatment (Appendix Table). Ten studies assessed the ef-
fect of CPAP on quality of life; however, they could not be
compared because of the use of various quality-of-life sub-
scales (8–17). Most studies found no statistically signifi-
cant improvement with CPAP; however, some showed im-
provement in physical and vitality scales of the Short
Form-36 Health Survey. Eight studies (9–11, 13, 15–18)
evaluated neurocognitive and psychological test results, and
most showed no improvement. None of the 7 studies (9,
10, 19–23) that addressed blood pressure differences or the
1 study (20) that assessed hemoglobin A1c level differences
found any statistically significant differences between
CPAP and control treatment. A recent RCT in patients
with OSA who did not have daytime sleepiness (24)
showed no statistically significant difference in the inci-
dence of hypertension or cardiovascular events for patients
treated with CPAP compared with control treatment (in-
cidence density ratio, 0.83 [95% CI, 0.63 to 1.1]; P �
0.20). Another RCT treated patients with acute stroke and
OSA (AHI scores�10 events/h) with either CPAP or stan-
dard of care (25). Authors reported a statistically significant
improvement in National Institutes of Health Stroke Scale
scores at day 8 between patients randomly assigned to
CPAP who were excellent users (CPAP device use�4 h for
the first night and mean use �4 h/night for the first 3
nights; National Institutes of Health Stroke Scale score of
2.3 in 10 excellent CPAP users out of 25 total CPAP users)
compared with those randomly assigned to standard of care
(National Institutes of Health Stroke Scale score of 1.4 in
25 patients) (P � 0.022). Only 1 study examined a clinical
outcome and found no statistically significant improve-
ment of heart failure symptoms with CPAP compared with
no specific treatment (8).
The literature review identified no RCTs evaluating
the effect of CPAP on mortality rates. The ACP’s supple-
mental search also identified no long-term RCT of CPAP
in the general OSA population but did identify 1 recent
trial that randomly assigned patients with OSA who previ-
ously had a stroke to CPAP treatment versus conservative
therapy; however, no statistically significant improvement
in cardiovascular mortality rates was reported at 24 months
(0% vs. 4.3%; P � 0.161) (26). Five out of 8 identified
observational studies of CPAP in general OSA populations
reported statistically significant reductions in overall mor-
tality rates associated with CPAP (or CPAP adherence)
compared with no CPAP (or CPAP nonadherence) (27–
35). Four out of 6 observational studies reported statisti-
cally significant reductions in cardiovascular mortality rates
associated with CPAP (or CPAP adherence) (27, 29, 31,
32, 36, 37). Eight observational studies reported multivar-
iate mortality analyses (overall or cardiovascular-specific
death), and 7 of them reported statistically significant re-
ductions in mortality rates associated with CPAP therapy
(27, 30–32, 36–39).
CPAP Versus Sham CPAP
Twenty-four trials (reported in 30 articles) compared
CPAP devices with sham CPAP treatment and included
patients with mean baseline AHI scores between 22 and 68
events per hour. Follow-up ranged from 1 week to 3
months. These studies apply to many patients with OSA
and included both fixed CPAP devices and autotitrating
CPAP (auto-CPAP) devices, which involve a modification
in which the machine increases positive pressure in re-
sponse to airflow resistance. Moderate-quality evidence
showed that CPAP was more effective than sham CPAP
and reduced AHI scores, improved ESS scores, and re-
duced arousal index scores compared with sham CPAP
(Appendix Table). One study (40) found better quality-of-
life scores for auto-CPAP than for sham CPAP, although 5
other studies showed no benefit (41–45). No significant
differences were found between CPAP and sham CPAP for
oxygen saturation, sleep efficiency, Multiple Sleep Latency
Test results, or sleep quality assessed by the Functional
Outcomes of Sleep Questionnaire (5). Data on blood pres-
sure were inconsistent across the 12 trials that assessed it
(46–57), and no study compared CPAP with sham CPAP
in regard to death or cardiovascular illness.
Oral CPAP Versus Nasal CPAP
Evidence from 3 small trials (46–48) to determine the
comparative efficacy of oral CPAP versus nasal CPAP was
insufficient. The trials reported inconsistent results, and
effect estimates were generally imprecise, mostly because of
the small sample sizes.
Auto-CPAP Versus Fixed CPAP
Twenty-one RCTs (58–78) compared auto-CPAP
with fixed CPAP in patients with mean AHI scores be-
tween 15 and 55 events per hour and included many obese
and overweight patients. These studies are mostly applica-
ble to patients with AHI scores greater than 15 events per
hour and body mass indices (BMIs) greater than 30 kg/m2.
Follow-up ranged from 3 weeks to 9 months. Despite
some miniscule differences in ESS (0.5 point), minimum
oxygen saturation (1%), and adherence (11 minutes), these
differences are probably not clinically significant. Overall,
moderate-quality evidence showed that auto-CPAP and
fixed CPAP have similar adherence and treatment effects
for patients with OSA (Appendix Table). The authors of
the original evidence review published an updated review
that included 3 additional studies and reported similar
findings, concluding that the 2 treatments are similarly
effective (79).
Bilevel CPAP Versus Fixed CPAP
Bilevel CPAP devices are designed to alleviate the dif-
ficulty and discomfort of exhaling against the fixed pressure
of CPAP by delivering lower pressure during exhalation
rather than during inhalation. Evidence from 5 trials (80,
Clinical GuidelineClinical Practice Guideline on Management of OSA in Adults
www.annals.org Annals of Internal Medicine
This online-first version will be replaced with a final version when it is included in an issue. The final version may differ in small ways.
Downloaded From: http://annals.org/ on 09/24/2013 www.medlive.cn
81) to determine the comparative efficacy of bilevel CPAP
versus fixed CPAP was insufficient. The studies were small
and highly clinically heterogeneous, and most had impre-
cise treatment effect estimates and showed null findings.
No study showed a difference in adherence between the
devices.
Flexible Bilevel CPAP Versus Fixed CPAP
Flexible bilevel CPAP delivers reduced positive airway
pressure at the end of inspiration and beginning of expira-
tion, which is determined partly by the user’s respiration.
This differs from the fixed positive airway pressure deliv-
ered by CPAP and was designed to improve comfort. Ev-
idence from 1 moderate-quality study (52) to determine
the comparative efficacy of flexible bilevel CPAP versus
fixed CPAP was insufficient.
C-Flex Versus Fixed CPAP
C-Flex (Philips Healthcare, Andover, Massachusetts)
CPAP is a proprietary technology that slightly reduces the
pressure at the beginning of exhalation. Four trials (53, 54,
82, 83) compared fixed CPAP with C-Flex CPAP in pa-
tients with mean baseline AHI scores between 35.4 and
53.3 events per hour. Follow-up ranged from 1.5 to 6
months. Low-quality evidence showed that C-Flex CPAP
and fixed CPAP are similarly tolerated and efficacious for
OSA treatment (Appendix Table).
CPAP With or Without Humidification
Evidence from 5 studies (55, 84–87) to determine the
benefit of CPAP with or without humidification was insuf-
ficient. The studies were low- to moderate-quality, small,
and clinically heterogeneous and reported inconsistent re-
sults. Although 2 trials reported improved adherence to
humidification, the other 3 trials reported no significant
difference.
MADs for OSA Treatment
MADs Versus No Treatment
Five trials (11, 12, 56, 88, 89) compared several
MADs with no treatment in patients with mean baseline
AHI scores between 19 and 34 events per hour. These
studies generally apply to patients with AHI scores of 15
events or greater per hour, although patients with comor-
bid conditions or excessive sleepiness were excluded from
some trials. Follow-up ranged from 4 to 23 weeks.
Moderate-quality evidence showed that MAD use im-
proved the signs and symptoms of sleep apnea, including
AHI score, arousal index score, and minimum oxygen sat-
uration (Appendix Table) compared with no treatment.
No clear survival benefits or reductions in cardiovascular
illness were associated with MADs.
MADs Versus Inactive Oral Devices
Five studies (reported in 7 publications) (56, 57, 90–
94) compared MADs with inactive (sham) oral devices in
patients with mean baseline AHI scores between 25 and 36
events per hour and excluded patients with significant co-
morbid conditions or periodontal disease. Follow-up
ranged from 8 days to 6 weeks. Moderate-quality evidence
showed that MAD use improves the signs and symptoms
of sleep apnea (including AHI, ESS, and arousal index
scores and minimum oxygen saturation) compared with
inactive (sham) oral devices (Appendix Table). One study
each showed improvements in quality of life and neurocog-
nitive test results (56, 57, 90).
MADs Versus CPAP
Ten studies (reported in 11 articles) (11, 12, 95–103)
compared the efficacy of CPAP with MADs in patients
with mean baseline AHI scores between 18 and 40 events
per hour. Follow-up ranged from 1 to 10 months.
Moderate-quality evidence showe
本文档为【2013+ACP+阻塞性睡眠呼吸障碍指南】,请使用软件OFFICE或WPS软件打开。作品中的文字与图均可以修改和编辑,
图片更改请在作品中右键图片并更换,文字修改请直接点击文字进行修改,也可以新增和删除文档中的内容。
该文档来自用户分享,如有侵权行为请发邮件ishare@vip.sina.com联系网站客服,我们会及时删除。
[版权声明] 本站所有资料为用户分享产生,若发现您的权利被侵害,请联系客服邮件isharekefu@iask.cn,我们尽快处理。
本作品所展示的图片、画像、字体、音乐的版权可能需版权方额外授权,请谨慎使用。
网站提供的党政主题相关内容(国旗、国徽、党徽..)目的在于配合国家政策宣传,仅限个人学习分享使用,禁止用于任何广告和商用目的。