Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
https://doi.org/10.1186/s41606-018-0023-1
Sleep Science and Practice
REVIEW
Open Access
Hypopnea definitions, determinants and
dilemmas: a focused review
Q. Afifa Shamim-Uzzaman1*, Sukhmani Singh2 and Susmita Chowdhuri3
Abstract
Obstructive sleep apnea (OSA) is defined by the presence of repetitive obstructive apneas and hypopneas during sleep.
While apneas are clearly defined as cessation of flow, controversy has plagued the many definitions of hypopneas,
which have used variable criteria for reductions in flow, with or without the presence of electroencephalographic (EEG)
arousal, and with varying degrees of oxygen desaturation. While the prevalence of OSA is estimated to vary using the
different definitions of hypopneas, the impact of these variable definitions on clinical outcomes is not clear. This
focused review examines the controversies and limitations surrounding the different definitions of hypopnea, evaluates
the impact of hypopneas and different hypopnea definitions on clinical outcomes, identifies gaps in research
surrounding hypopneas, and makes suggestions for future research.
Keywords: Obstructive sleep apnea, Hypopnea, Obstructive hypopnea, Central hypopnea
Introduction
Obstructive sleep apnea (OSA) is a common disorder,
composed of apneas and hypopneas occurring at least
five times per hour during sleep. Since polysomnographic identification in 1965, the notion of apneas
(absence of airflow for > 10 s, Fig. 1) remains undisputed; however, the definition of hypopneas continues to
evolve and their clinical impact debated over the years.
Bloch et al. first described ‘hypopneas’ as reductions in
oxygen saturation that occurred in association with
reductions in airflow instead of with absence of airflow,
i.e., events suggestive of decreased ventilation that did
not meet criteria for apneas. (Bloch et al., 1979) In this
study “normal” asymptomatic volunteers had 40% more
hypopneas than apneas (105 vs. 60, respectively) with
frequent oxygen desaturation of ≥4%. (Bloch et al.,
1979) Subsequently, in a small study comparing individuals with apneas alone vs. hypopneas alone (n = 50),
Gould et al. noted no differences in age, weight, clinical
symptoms, number of arousals (median 31/h vs. 20/h)
or patterns of oxygen desaturation (median 45 vs. 40, 4%
desaturation per hour) (Gould et al., 1988) between the
two groups, and recommended changing the terminology
* Correspondence: afifa@med.umich.edu
1
VA Ann Arbor Heathcare Center and University of Michigan, 2215 Fuller Rd,
Ann Arbor, MI 48105, USA
Full list of author information is available at the end of the article
from “sleep apnea syndrome” to “sleep hypopnea
syndrome,” defined as 15 or more hypopneas per hour of
sleep in conjunction with 2 or more major clinical
features. Although the term “sleep hypopnea syndrome” did
not gain much popularity, the terminology “sleep apneahypopnea syndrome” (SAHS) was used frequently, until the
current term “obstructive sleep apnea” gained favor.
Objectives
In this focused review, our objective was to describe the
variability in the definitions of hypopneas, limitations of
technology that are used to detect hypopneas, and thereafter, make suggestions for future research to standardize
hypopnea definition and detection. Our literature review
also attempted to identify the potential clinical relevance
of patients with hypopnea-predominant sleep apnea.
These are outlined below.
Background
Defining moments for ‘hypopnea’
Gould’s definition of hypopnea was derived by comparing
75, 50% or 25% reductions in Respitrace thoracoabdominal sum compared to thermocouple flow amplitude with arousal frequency and oxygen desaturations.
(Gould et al., 1988) In this study, a 75% reduction in
movement resulted in much fewer hypopneas than the
number of desaturations or arousals and was excluded
© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
Page 2 of 12
Fig. 1 This figure shows an obstructive apnea. An apnea is a respiratory event lasting ≥10 s, characterized by a decrement in airflow of ≥90% from
the baseline in the oronasal thermocouple signal. Clear crescendo effort in the abdominal belt suggests obstruction. Elevated and progressively
increasing values in the Δ Pes during the event confirm the obstructive etiology
from consideration. While reductions in thoraco-abdominal
movement of 25–50% were of similar accuracy and more
accurate than the frequency of oxygen desaturation alone,
the 50% reduction in effort was significantly closer to the
arousal frequency than was the 25% reduction in thoracoabdominal movement (p < 0.05). Hence, these authors defined ‘hypopnea’ as a “50% reduction in thoracoabdominal
(Respitrace® sum) amplitude for 10 seconds or more when
compared to the peak amplitude lasting for 10s or more that
occurred within the previous 2 minutes in the presence of
continued flow”. (Gould et al., 1988)
In 1997, the AASM created a task force to delineate the
criteria to identify and treat OSA. Their results, presented
as a consensus statement commonly referred to as the
“Chicago Criteria,” defined hypopnea as a ≥ 50% decrement in airflow, or a < 50% reduction in airflow associated
with either an oxygen desaturation or arousal. (Loube et
al., 1999) Despite this, no uniform definition of ‘hypopnea’
was used amongst sleep laboratories within the United
States for the next decade. (Moser et al., 1994; Redline &
Sanders, 1997) A survey of 44 accredited sleep laboratories (labs) showed as many methods and definitions of
hypopneas as number of labs. (Moser et al., 1994)
Methods of detection included use of thermocouple,
pneumotachograph, respiratory inductance plethysmography, intercostal electromyography, microphone or esophageal balloon. Additionally, the requirements for the degree
of airflow reduction and oxygen desaturation also varied
widely. Moreover, 33 of the 44 labs used EEG arousal to
fulfill the definition of hypopnea, even though there was
no consistent definition of arousal at that time. This lack
of precision precluded objective comparison of data from
individual laboratories and raised doubts to the validity
and reproducibility of hypopneas even within the same
individual. In fact, Redline et al. (Redline et al., 2000)
examined the effect of using 11 different criteria for scoring hypopneas on the prevalence of disease in a large
community-based sample and reported that different approaches for measuring apnea-hypopnea index (AHI:
number of apneas and hypopneas per hour of sleep)
resulted in substantial variability in identifying and classifying sleep-disordered breathing.
Findings
A. Sources of variability in hypopnea detection
i) Variability in flow measurements: Hypopnea detection
implies determination of small changes in ventilation
that accompany sleep disordered breathing; the
amplitude of airflow is a measure of these changes.
Sources of variability that contribute to poor reliability
of these measurements of airflow include:
1) positioning of thermo-elements, as slight
displacements could produce major changes in
signal amplitude,
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
2) alterations in proportion between nasal and oral
breathing,
3) nasal cycle causing alterations in nasal airflow
(which could change with changes in body
position), (Cole & Haight, 1986)
4) variation in sensitivity and frequency response
between different thermo-elements, (Berg et al., 1997)
5) displacement of the Respitrace® girdles that could
alter signal amplitude.
ii) Type of device: Variability can also arise from the
type of devices used during the recording. One
study demonstrated that despite relatively high
correlation coefficients between the methods of
detecting hypopneas, agreement between the
devices detecting changes in ventilation (using
thermistor, nasal pressure and/or Respitrace®) were
low, with poor agreement with minute ventilation
measured by head-out body plethysmography in
awake subjects. (Berg et al., 1997) The best agreement
was noted with plethysmographic minute ventilations
and the amplitudes of the summed Respitrace signals,
and from the nasal-pressure signals. In fact,
nasal-pressure measurements provided the greatest sensitivity and negative predictive values.
Combination of nasal pressure and Respitrace®
provided more consistent results – 86% sensitivity
and 83% specificity – and better agreement between
both methods (Cohen’s K = 0.65).
iii) Observer reproducibility: Finally, Whyte et al.
showed reproducibility in scoring of hypopneas by
different observers. (Whyte et al., 1992) When two
polysomnographers were asked to independently
score both apneas and hypopneas on all-night
polysomnograms of patients with OSA using the
same methodology, there was close agreement
between the polysomnographers for the number of
hypopneas (r = 0.98; mean difference 11%) and for
the number of apneas (r = 0.99; mean difference
8%). The agreement was similar for the durations of
both hypopneas (r = 0.99; mean difference 13%) and
apneas (r = 0.99; mean difference 11%). There was
also close agreement between the total number of
respiratory events scored with and without reference
to the flow signal (r = 0.99; mean difference 1.4%)
with a maximum under-recognition of 18 events per
night in a subject with 237 apneas per night. (Whyte
et al., 1992) Hence, it was possible for different
observers to score hypopneas reliably.
iv) Variability in baseline: The lack of clear
determination of “baseline” or normative values for
each patient lends itself to inherent variability. If the
baseline (SpO2, flow, EEG, muscle tone, etc) is not
clear, variations from the baseline are subject to
Page 3 of 12
interpretation. For example, subjective variations in
detection of arousals can lead to variations in
scoring hypopneas related to arousals. Since
arousals can vary in their intensity and subsequent
autonomic responses, (Azarbarzin et al. SLEEP
2014;37(4):645–653) they are not always detected
by current scoring methods. The threshold visual
intensity that causes different scorers to score
arousals varies considerably, with some scoring
arousals with minimal, equivocal changes in EEG
whereas others score arousals only when the
changes are unequivocal. When arousals are
generally intense this is not a problem but when
arousal changes are mild, large differences in
AHI can arise. While the AASM scoring rules
require that only arousal lasting 3 s be scored,
the rules do not specify the minimum time
difference between an arousal following a
hypopnea. This can also can lead to variations in
scoring arousals and ultimately to scoring
hypopneas associated with arousals.
Attempts at reducing variability
Identification of factors affecting scoring:
A decade after the Chicago criteria, in an attempt to
standardize definitions used by sleep laboratories and
researchers, the American Academy of Sleep Medicine
(AASM) published the AASM Manual for the Scoring
of Sleep and Associated Events in 2007. This manual
defined a hypopnea as a 30% reduction in airflow, as
measured by the nasal pressure transducer flow signal, with a concomitant 4% drop in oxygen saturation;
alternatively, a hypopnea was also defined as a 50%
or greater decline in the flow signal associated with a
3% drop in oxygen saturation and/or an EEG arousal
lasting at least 3 s in duration. (Iber et al., 2007)
Controversy regarding the best definition led to the
adoption of both definitions in the scoring manual;
the first being referred to as rule “4A” (or “recommended”) (Fig. 2) and the latter as rule “4B” (or
“alternative”) (Fig. 3).
However, the use of the recommended vs. alternative definitions of hypopnea led to highly variable
apnea/hypopnea indices. Ruehland et al. scored the
same 323 consecutive sleep studies using different
hypopnea definitions and found considerable variability in the median apnea-hypopnea index (AHI, 8.3 vs.
14.9) as well as the hypopnea index (HI, 2.2 vs. 7.2)
using the recommended and alternative definitions,
respectively. (Ruehland et al., 2010) Greater than half
of the inconsistencies in AHI was due to the inclusion of arousals in the alternative definition, and a
quarter due to the reduction of the desaturation
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
Page 4 of 12
Fig. 2 This figure shows a hypopnea scored using the recommended criteria IVA of the 2012 AASM Scoring guidelines – requiring a ≥ 30%
decrement in flow associated with a ≥ 4% decrease in oxygen saturation. Note the progressively increasing and elevated Δ Pes that confirm the
obstructive etiology
requirement from 4 to 3%. (Ruehland et al., 2010)
This translated to differences in the identification and
classification of sleep apnea in the same patient.
Hence, further clarification, with consideration of the
clinical implications, was sought and is outlined
below.
i) Effect of the arousal criterion on hypopnea scoring
and classification of severity of sleep apnea The association between the arousal index and cardiovascular
morbidities is not as robust as that of oxygen desaturation indices, below. However, correlations have been
shown between the arousal index and hypertension
(Sulit et al., 2006) as well as white matter disease in the
elderly. (Ding et al., 2004) In fact, the Cleveland Family
Study showed a greater correlation of hypertension risk
with the arousal index than with oxygen desaturation.
This may, in part, be due to the activation of the sympathetic nervous system when arousals occur during sleep,
(Loredo et al., 1999; Somers et al., 1993) and the resultant sleep fragmentation leads to clinically significant
symptoms. (Bonnet, 1986; Thomas, 2006; Guilleminault
et al., 2009) With respect to scoring, Guilleminault et al.
showed that using criteria 4A to score hypopneas (i.e., a
30% flow reduction with 4% oxygen desaturation, without consideration of arousals) would have missed 40% of
patients identified using the criteria incorporating
arousals and who were responsive to positive airway
Fig. 3 This figure shows a hypopnea scored using the alternative criteria IVB, i.e., ≥50% decrement in flow associated with a ≥ 3% decrease in
oxygen saturation or an arousal. This event would have been missed if using the recommended criteria IVA of the 2012 AASM Scoring guidelines
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
pressure (PAP) therapy (with both reductions in AHI
and sleepiness). (Guilleminault et al., 2009)
ii) Effect of oxygen criterion There are clear, strong
associations between obstructive respiratory events and
cardiovascular events, stroke, and hyperglycemia, regardless of the SpO2 reduction criteria (3% vs. 4%) used.
(Berry et al., 2012a) In addition, the correlation between
AHIs scored with 3% oxygen desaturation and 4%
oxygen desaturation was > 0.95 (Redline et al., 2000),
showing excellent concordance. Hence, a 3% reduction
criteria was recommended in the update to the scoring
manual.
Of note, however, in 2015, Myllymaa et al. examined
the effects of different oxygen desaturation threshold
(ODT) levels on the AHI of 54 patients (Myllymaa et al.,
2016). Hypopneas were defined as a decrement in airflow of ≥30% for over 10s along with one of the following: an ODT ≥ 2% (ODT2%), ODT ≥ 3% (ODT3%),
ODT ≥ 4% (ODT4%), ODT ≥ 5% (ODT5%) or ODT ≥ 6%
(ODT6%). Not only was there a significant increase in
the median AHI with ODT3% vs. ODT4% (6.5 events/
hr.; p = 0.003), different ODT’s resulted in patients being
classified under different categories of AHI severity.
Using ODT3% instead of ODT4% resulted in a 44% increase (from 29.4 to 73.5%) in the number of patients with
moderate or severe OSA (AHI ≥ 15). Thus, any changes in
ODT, although slight, could result in significant differences in AHI, which could in turn, result in highly variable
classifications of disease severity. (Myllymaa et al., 2016)
iii) Effect of flow reduction criterion Hypopneas
defined with either 30% decrements in flow or 50%
decrement in flow, if resulting in a desaturation or an
arousal, carried clinical consequences, be it disrupted
sleep, daytime sleepiness, or cardiovascular morbidity.
However, a hypopnea based only on desaturation criteria
alone (without arousals), would miss much clinically
significant disease, as noted above.
iv) Calibration model for apnea-hypopnea indices:
Impact of alternative criteria for defining hypopneas
Analysis of 6441 polysomnograms showed that AHI
values were sensitive and changed substantially depending on the hypopnea criteria used. (Ho et al., 2015) Also,
there was greater concordance (or “stability”) in AHI between the two hypopnea definitions as AHI increased
above 30, but greater variability (or “divergence”) at
lower AHIs. (Ho et al., 2015) Additionally, in 2 Spanish
cohorts of 1116 women and 939 elderly individuals, the
prevalence of an AHI ≥30 events/h increased by 14%
when using AHI with 3% desaturation plus arousal
criterion (AHI3%a), compared to the AHI using 4%
(AHI4%) desaturation criterion. (Campos-Rodriguez et
Page 5 of 12
al., 2016) The percentage of women with an AHI < 5
events/h decreased from 13.9% with AHI4 to 1.1% with
the AHI3%a definition; almost one-third (31%) of the
investigated subjects moved from normal to OSA labels
or vice versa. Moreover, the proportion of moderate
(15 ≤ AHI < 30 per hour) and severe (AHI ≥ 30 per hour)
OSA changed 13.5 and 10%, respectively, depending on
the hypopnea definition used. (Farre et al., 2015) Thus,
although using different hypopnea criteria may not make
a significant difference in OSA diagnosis for patients
with more severe disease (AHI > 30), it could result in
the misclassification of disease at lower AHI levels.
Standardization of scoring
These findings expounded the need for further
standardization. The 2012 update to the scoring manual
attempted to do just that, refining the definition of
hypopnea to a 30% decrease in airflow lasting at least
10 s and associated a ≥3% SpO2 desaturation or an
arousal. (Berry et al., 2012b) In addition, it included consensus definitions for obstructive and central hypopneas
for the first time. From previous operational definitions
used in heart failure with an obstructive event, obstructive hypopneas required any of the following indicators
relative to baseline: paradoxical thoraco-abdominal
movement, snoring, and inspiratory flattening of the
flow signal whereas central hypopneas required the
absence of all of these indicators (Fig. 4). Simply put, an
obstructive hypopnea was a reduction in flow secondary
to increased resistance of the upper airways (i.e.,
obstruction), whereas a central hypopnea was a result of
decreased effort, not increased resistance (Fig. 5). However, the differences between central and obstructive
hypopneas were not validated using esophageal catheter
pressure changes, a gold standard measure of respiratory
effort. Iber cautioned that given the substantial evidence
supporting interaction between central and obstructive
events, more emphasis should be placed on identifying
causes such as heart failure, sleep disruption, and hypoxemia, rather than just distinguishing between obstructive
and central events. (Iber, n.d.)
Randerath compared polysomnography (PSG) and
esophageal manometry in 41 patients suspected of having
sleep apnea; hypopneas were independently discriminated
by blinded investigators based on either esophageal
pressure or the visual PSG-based algorithm (presence or
absence of flattening of the flow curve, paradoxical breathing effort, termination of the hypopnea, position of the
arousal, and correlation with sleep stages). (Randerath et
al., 2013) Of the 1837 scorable hypopneas, 1175 (64%)
could be further defined by esophageal pressure and 1812
(98.6%) by the PSG-based algorithm; notably, evaluation
of hypopneas using esophageal pressure was limited by
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
Page 6 of 12
Fig. 4 An obstructive hypopnea. A hypopnea is classified as an obstructive hypopnea if the event meets all criteria for hypopnea and signs of
obstruction (snoring, flow limitation, crescendo effort, or paradoxical breathing) are seen during the event
poor signal quality and artifact. Of those hypopneas that
could be differentiated with both methods, using esophageal pressure as a reference, the PSG-based algorithm correctly defined 76.9% of central and 60.5% of obstructive
hypopneas. However, because the esophageal manometry
was not interpretable in 36% of their cases, the accuracy
of a combined logic for hypopnea definition was only 68%.
Thus, although 77% of central hypopneas were correctly
identified, nearly 40% of obstructive events were misclassified. (Randerath et al., 2013) Thus, variability in the definitions of hypopneas has led to re-classification of the type
and severity of OSA.
In a retrospective study, PSGs of 112 consecutive
patients for suspected OSA were re-scored for respiratory
events
using
either
2007
AASM
recommended (AASM2007Rec), 2007 AASM alternate
(AASM2007Alt), Chicago criteria (AASM1999), or
2012 AASMrecommended (AASM2012) respiratory
event criteria (Duce et al., 2015). The median AHI
using AASM2012 definitions, was approximately 90%
greater than the AHI obtained using the AASM2007
recommended criteria, approximately 25% greater
than the AASM2007Alt AHI, and approximately 15%
lower than the AASM1999 AHI. These changes increased OSA diagnoses by approximately 20 and 5%
for AASM2007Rec and AASM2007Alt, respectively.
Minimal changes in OSA diagnoses were observed
between AASM1999 and AASM2012 criteria. Differences between the AASM2007 using recommended
criteria and AASM2012 hypopnea indices were
Fig. 5 A central hypopnea lacks the obstructive features seen in Fig. 4. The lack of elevated Pes values also confirms the central etiology of
the hypopnea
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
predominantly due to the change in desaturation
levels required.
Results from such studies point to the growing importance of finding consistent methods for scoring hypopneas. Approaches designed to “calibrate AHI thresholds
to the event definitions employed” or create equations to
measure AHI specific to the technology in different
laboratories have been considered. (Ho et al., 2015)
Clinical factors determining the type of hypopnea
Although the diagnostic value of apnea-hypopnea indices (AHIs), as determined by different hypopnea definitions, has been evaluated by investigators, it is as yet
unclear what determines the type of obstructive respiratory event an individual will have. Are there physiologic
characteristics that predetermine whether an individual
will have primarily apneas or primarily hypopneas? What
underlying differences lead to some individuals having
hypopneas associated with oxygen desaturations while
others have hypopneas terminating in arousals? The
literature detailing this, outlined below, is sparce.
Determinants of arousal-based vs. desaturation-based
hypopneas
Tsai et al. reported that regardless of the hypopnea criteria used to define sleep apnea, there were no significant differences in patient characteristics [age, sex, body
mass index (BMI), and neck circumference], or in consequent Epworth Sleepiness Scale, time spent at an SaO2
below 90%, arousal index, or apnea index between
patients with predominantly arousal-based hypopneas
versus those with desaturation-based hypopneas. (Tsai et
al., 1999) No patient characteristics predicted the type of
hypopnea, regardless of which hypopnea scoring method
was used; however, while the addition of arousal-based
scoring criteria for hypopnea caused only small changes
in the AHI, OSA defined solely by an AHI value increased the prevalence of OSA. (Tsai et al., 1999)
Determinants of hypopneas vs. apneas
i) Effect of BMI
In a retrospective study of 90 adults with OSA, comparing two groups with body mass indices (BMI) ≥45 vs.
BMI < 35, matched for age and gender, the hypopnea-toapnea ratio (HAR) was significantly higher in the BMI
≥45 group (38.8 ± 50.7) compared to the BMI < 35 group
(10.6 ± 16.5), p = 0.0006. (Mathew & Castriotta, 2014)
The hypopnea index, but not the apnea index, was also
higher in the BMI ≥45 vs. BMI < 35 group (28.7 ± 28.6
vs 12.6 ± 8.4, p = 0.0005), as was the AHI (35.5 ± 33.8 vs
22 ± 23, p = 0.03). In addition, the end-tidal CO2 was
higher in the higher BMI group. However, the
Page 7 of 12
hypopnea-to-apnea ratio did not appear to be influenced
by the presence or absence of hypoventilation and was
similar for those with or without obesity hypoventilation
syndrome. (Mathew & Castriotta, 2014) In fact, BMI
was the only significant predictor of HAR (adjusted
r2 = 0.138; p = 0.002) when adjusting for age, gender,
race, and ETCO2. Of note, a small sample size may
have confounded the study findings. The authors suggested that different pathophysiologic mechanisms
may have been involved in the generation of apneas
and hypopneas.
ii) Effect of Sex Hormones
A study of 118 patients with ‘occlusive’ sleep apnea
syndrome, defined as daytime hypersomnolence and an
AHI > 10/h, reported that, in women, only about 30% of
respiratory events during sleep were occlusive apneas
while 70% were hypopneas; conversely, in men, only 50%
of events were hypopneas. The authors highlighted that
both premenopausal and postmenopausal women had
more hypopneas than apneas and “some of the most
severely affected women were never observed to have
complete cessation of airflow during sleep”. (Leech et al.,
1988) Notably, there were fewer sleep disordered breathing events associated with oxygen desaturation in
women than men (p < 0.003); 19 women did not experience oxygen desaturation at all, and only three had a
total of nine episodes of apnea, whereas 20 men
accounted for 264 episodes of nocturnal oxygen desaturation or abnormal breathing. (Bloch et al., 1979)
Thus, gender differences exist in the prevalence of
hypopneas, and these may be conferred by differences in
upper airway anatomy or control of ventilation. The
latter may be attributed to hormonal differences that in
turn modify ventilatory responsiveness during sleep.
Rowley et al. showed that the determinants of the
change in end-tidal CO2 at the apnea threshold included
sex and menopausal status, with changes in end-tidal
CO2 at the apnea threshold highest in premenopausal
women (4.6+/− 0.6 mmHg), with no difference between
the postmenopausal women (3.1+/− 0.5 mmHg) and men
(3.4+/− 0.7 mmHg) (Rowley et al., 2006). Hormone replacement therapy increased the change in end-tidal CO2 (CO2
reserve) at the apnea threshold from 2.9+/− 0.4 mmHg to
4.8+/− 0.4 mmHg (P < .001) indicating that estrogens and
progestins stabilize breathing in women during non-rapid
eye movement sleep. (Rowley et al., 2006) Moreover, studies
suggest that testosterone increases the risk for central
events during sleep in men. (Zhou et al., 2003; Chowdhuri
et al., 2013)
Thus, although no patient characteristics can determine the predominant type of hypopnea (arousal- vs.
desaturation-based) an individual may have, obesity
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
and female sex may be associated with hypopneapredominant OSA, rather than apnea-dominant.
Clinical consequences of hypopneas
Impact of differing definitions on clinical outcomes
The immediate consequences of hypopneas do not appear
to differ from those of apneas. In 39 sleep apnea patients
who underwent polysomnography, 80 events/subject were
evaluated for clinical consequences — i.e., oxygen desaturation of ≥4% from the baseline, EEG arousal, and an increase in heart rate by 6 bpm. (Ayappa et al., 2005) Both
apneas and hypopneas were not significantly different in
frequency for oxygen desaturation (78% vs. 54%,
respectively) arousals (63% vs.47%, respectively) and associated increase in heart rate (73% vs. 55%, respectively). In
contrast, of the events with minimal (25–50%) amplitude
reduction, only 25% caused desaturation, 42% arousal, and
42% heart rate increase. No specific consequence occurred
after every event. Thus, the immediate consequences of
individual respiratory events (oxygen desaturation, EEG
arousal and heart rate) overlapped and were not specific
to any particular event. The same may not hold true for
excessive daytime sleepiness or for long term cardiovascular sequelae.
i) Excessive daytime sleepiness Hosselet et al. observed
that the respiratory disturbance index (RDItotal), calculated from the sum of apneas, hypopneas and flow limitation events regardless of the level of desaturation or
arousal (Hosselet et al., 2001), predicted daytime sleepiness. In this study, the highest sensitivity and specificity
in separating patients with excessive daytime sleepiness
(EDS) from patients without EDS (non-EDS) was provided by the RDItotal. For RDItotal, the optimal combination of sensitivity and specificity was obtained at a
cutoff value of 18 events/h. However, the cutoff value of
5/h for the AHI per AASM results in sensitivity of 100%
but specificity for EDS of only 15%.
Similarly, Ciftici et al. studied 90 patients who had an
AHI > 5/h, scored according to the hypopnea definition
of the AASM (Ciftci et al., 2004). The records of these
patients were scored according to different hypopnea
definitions (hypopnea-arousal, hypopnea-desaturation,
hypopnea-effort). AHI (AASM), AHI (arousal), AHI (desaturation), and AHI (effort) were determined. Patients’
daytime sleepiness was evaluated by the Epworth Sleepiness Scale (> 10). When all of three major symptoms
(snoring, observed apnea, and daytime sleepiness) were
found in a patient’s history, the term “clinical OSAS”
was applied. ESS was strongly correlated with each
index. In addition, an AHI-AASM cutoff value > 5 had
the highest sensitivity and specificity from the viewpoint of
separation between EDS and non-EDS, and also between
clinical OSAS and nonclinical OSAS. (Ciftci et al., 2004)
Page 8 of 12
Chervin & Aldrich noted that the rate of apneas as
opposed to the rate of hypopneas had a greater impact
on the degree of excessive daytime sleepiness in patients
with OSA (Chervin & Aldrich, 1998). In 1146 subjects
(30% females), the mean number of apneas per hour of
sleep (AI) was 14.3 ± 27.0 and the mean number of
hypopneas per hour of sleep (HI) was 16.5 ± 16.1. A
regression model showed that the AI explained 9.6% of
the variance in mean sleep latency (MSL) (p ≤ 0.0001) on
Mean Sleep Latency Tests, after controlling for total
sleep time, but the HI explained only 5.4% (p ≤ 0.0001)
of the variance. When AI, HI, and TST (total sleep time)
were included in a single multiple-regression model, AI
explained 8.3% of the variance in MSL and HI explained
4.0% (p < 0.0001 for each). The AHI during supine sleep
(recorded in a subgroup of n = 169 subjects), the rate of
apneas (n = 1146), and the rate of obstructive apneas
were useful in explaining variation in measured levels of
sleepiness; however, rates of hypopneas and central apneas were not as useful. The minimum recorded oxygen
saturation (n = 1097) was as important as the AHI to the
level of sleepiness. (Chervin & Aldrich, 1998)
ii) Metabolism In 2656 subjects from the Sleep Heart
Health Study, hypopneas, even with mild degrees of oxygen
desaturation of 2–3%, were associated with fasting hyperglycemia, independent of multiple covariates. Hypopneas were
further stratified on the degree of associated oxyhemoglobin
desaturation into: 0.0–1.9%, 2.0–2.9%, 3.0–3.9%, and ≥ 4.0%
reductions in SaO2. Hypopneas based solely on the arousal
criteria were not identified. The adjusted cumulative odds
ratios for the hypopnea index (HI) and impaired fasting glucose were 1.15 (95% CI: 0.90–1.47), 1.44 (95% CI: 1.09–1.90),
2.25 (95% CI: 1.59–3.19) and 1.47 (95% CI: 1.13–1.92)
respectively. (Stamatakis et al., 2008)
iii) Stroke Association between incident stroke and OSA
using a hypopnea definition of ≥3% oxygen desaturation
has been reported (Redline et al., 2010; Shahar et al., 2001)
and may be somewhat stronger than the association with
coronary heart disease or heart failure. This association of
stroke and OSA may be mediated through ischemic pathways. Potential mechanisms: Andreas et al. simulated
obstructed breaths using the Muller maneuver (generating
high negative intrathoracic pressures against an obstruction) and showed a significant reduction in blood flow to
the middle cerebral artery (MCA) during the period of
obstruction, in conjunction with a drop in flow across the
mitral and aortic valves. (Andreas et al., 1991) Using
Doppler sonography, Netzer et al. showed that blood flow
through the MCA was significantly reduced (i.e., > 50%
reduction in velocity) more frequently with obstructive
hypopneas (76%) and obstructive apneas (80%) than with
central apneas (14%) (p ≤ 0.0001); the level of reduced
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
blood flow during obstructive apneas vs. obstructive
hypopneas was not significantly different. However, there
was a significant association between MCA blood flow reduction and the duration of obstructive hypopnea (p < 0.05),
which was not seen with obstructive apneas or central
apneas, although mean event durations were similar
(18.1 ± 6.5 s for hypopnea, 17.2 ± 5.9 s for central
apneas, and 14.8 ± 5.0 s for obstructive apneas; p = 0.3).
Similarly, a statistically significant correlation (p < 0.05)
was seen between the fall in oxygen saturation with
obstructive hypopnea and reduction in MCA blood
flow, not seen with central or obstructive apneas.
(Netzer et al., 1998) Hence, the occurrence of MCA
blood flow reduction increases as the duration of the
obstructive hypopnea increases and its associated drop
in oxygen saturation increases.
iv) Cardiovascular disease In a cohort of 6106 adults
from the Sleep Heart Health Study hypopneas with ≥4%
oxygen desaturations were independently associated with
cardiovascular disease, whereas hypopneas with less than
a 4% desaturation or arousal only were not associated
with prevalent cardiovascular disease, after controlling
for apnea index, age, sex, race, body mass index, waist
circumference, neck circumference, total cholesterol,
smoking status, and hypertension. (Punjabi et al., 2008)
Mehra et al. found significant associations between
SDB and the risk of atrial fibrillation and complex ventricular ectopy (CVE) amongst 2911 elderly men without
heart failure where hypopneas were defined by a desaturation criterion of ≥3%. However, whether hypopneas
predicted arrthymias was not investigated. The authors
compared central vs. obstructive forms of sleep disordered breathing, and found that central sleep apnea was
more strongly associated with atrial fibrillation (Odds
Ratio 2.69, 95% CI: 1.61–4.47) than CVE (OR 1.27, 95%
CI: 0.97–1.66) while OSA was associated with CVE,
especially when associated with hypoxia; those in the
highest hypoxia category had an increased odds of CVE
(OR 1.62, 95% CI: 1.23–2.14) compared with those with
the lowest associated hypoxia. (Mehra et al., 2009)
Proposed mechanisms for the arrhythmic potential of
apneas and hypopneas include intermittent hypoxia
leading to increased oxidative stress, systemic inflammation, and sympathetic activity; repetitive blood pressure
elevations secondary to sympathetic activation; and
excessive intrathoracic pressure changes leading to
mechanical stress on the heart and blood vessel walls
(including large caliber vessels such as the aorta).
(Camen et al., 2013; Kohler & Stradling, 2010)
In patients with congestive heart failure (CHF), the criteria used to define hypopnea significantly influenced
the AHI and the prevalence of sleep-disordered breathing (SDB). (Ward et al., 2013) The number of patients
Page 9 of 12
with CHF in whom SDB was diagnosed, using an AHI
cutoff of ≥15/h, increased by 16% using the AASM
‘alternative’ hypopnea rule (≥50% reduction in airflow
with ≥3% oxygen desaturation or arousal) compared
with the ‘recommended’ hypopnea scoring rule (≥ 50%
decrease in nasal airflow with a ≥ 4% oxygen desaturation). Median AHI increased from 9.3/h to 13.8/h
(median difference 4.6/h) and SDB prevalence increased
from 29 to 46% with the AASM alternative scoring rule
(p < 0.001). However, classification of SDB as OSA or
central sleep apnea was not significantly altered by the
hypopnea scoring rules.
Recent large scale studies in the non-sleep literature
(McEvoy et al., 2016; Yu et al., 2017) boldly called into
question the benefit of treating sleep apnea on cardiovascular outcomes and death. Although riddled with
confounders such as non-adherence to PAP therapy,
(McEvoy et al., 2016; Yu et al., 2017) different types of
sleep apnea being treated (central vs. obstructive, (Yu et
al., 2017) different modes of PAP therapy used, (Yu et al.,
2017) and different diagnostic criteria for sleep apnea,
(McEvoy et al., 2016) these studies raise important questions on the validity of comparing data using different recording and scoring methodologies.
Of the ten studies reviewed in Yu’s meta-analysis
(which included the McEvoy study), only 2 used any
AASM criteria for scoring hypopneas, and though published in 2012 (Kushida et al., 2012) & 2015 (Huang et
al., 2015), both of these used the 1999 Chicago Criteria.
One study from Spain (Barbe et al., 2012) used a modification of the 2012 AASM criteria (scoring hypopneas
with 50% decrement in flow associated with a 4% oxygen desaturation) while another (Bradley et al., 2005)
scored hypopnea as a 50% decrement in flow only
(without a consequence). The remaining six studies
used cardiopulmonary or respiratory polygraphy, which
could not measure arousals, so any arousal-based
hypopneas would have been missed. Of these limited
channel studies, three used a 4% oxygen desaturation
index (ODI) of > 7.5 (Craig et al., 2012; McMillan et al.,
2014) or > 12 (4%-drops from baseline/hour) (McEvoy
et al., 2016) to diagnose sleep apnea; one (Parra et al.,
2015) used a “discernible reduction in airflow or thoracic motion lasting >10 seconds and associated with a
cyclical dip in SaO2 of > 3%” and calculated the AHI
based on time in bed. In the remaining 2 studies
(Cowie et al., 2015; Peker et al., 2016), scoring criteria
were not clearly defined.
This raises many unanswered questions and reflects
the current dilemmas. How did differences in diagnostic
criteria affect the overall interpretation of the metaanalysis? Would the conclusions have been the same if
there was a standardized definition of the disorder? Is it
conceivable that treatment of apnea-predominant versus
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
hypopnea predominant sleep apnea responded differently to PAP therapy? We currently do not have answers
to these important questions.
v) Mortality In the clinical Spanish cohorts, AHI ≥30
events/h was associated with increased cardiovascular
mortality risk in women after adjusting for multiple
covariates, regardless of the AHI4%, AHI3% or AHI3%arousal hypopnea definition, whereas in elderly individuals
the mortality risk was higher in those diagnosed using the
AHI4% and AHI3% definitions but not using the AHI3%a
definition. (Campos-Rodriguez et al., 2016)
Summary & recommendations
A. Technical specifications
While a number of studies have investigated the physiology and clinical significance of hypopneas, the data are
sparse and inconclusive, mainly because the definitions
and diagnostic methods have varied across studies. Thus,
there remains a crucial gap in knowledge regarding the
clinical presentation and prognosis of hypopneas. A
clear, standard, and consistent definition of hypopnea is
vital to this understanding. How can we claim that sleep
apnea has consequences if the disorder itself is not
clearly defined?
To this end, we recommend that the following specific, concrete recommendations be incorporated into
the scoring guidelines:
i) Clear definition of, or guidance on, determination
of baseline values for flow or SpO2. With today’s
technological advancements, digital methods to
determine these, especially when the pre-event
signals are unstable, could be helpful to avoid
subjectivity.
ii) Criteria for identification of poor or unreliable
signals (e.g., EEG, flow or SpO2 signals) and
guidance on when to exclude these from the
calculation of respiratory events or sleep time.
iii) Clear guidelines on arousal criteria that minimize
subjectivity and bias.
iv) Specifications on the use of sensors that meet
specific performance calibration criteria.
Clinical impact
Few studies have reported on the impact of the different definitions of hypopneas on chronic medical conditions. Also, studies evaluating the clinical impact of
these variable definitions of respiratory events on cardiovascular or neurocognitive sequelae are lacking. Specifically, whether combinations of respiratory events,
hypoxia and EEG arousals have variable physiological
effects on daytime sleepiness, cardiovascular morbidity
and mortality cannot be ascertained from these studies.
Page 10 of 12
There are no data available regarding effects of sleep
hypopneas in patients with asthma, COPD or other
lung and/or neuromuscular diseases. Whether treatment of ‘hypopnea-predominant’ OSA leads to reduced
cardiovascular morbidity or mortality or metabolic and
neurocognitive dysfunction is also not known. And, although studies suggest that sleep apnea may be related
to adverse clinical consequences such as cardiovascular
disease, stroke, abnormal glucose metabolism, excessive
daytime sleepiness, and increased mortality; further research is still needed to determine the effect that treating sleep apnea has on these condition.
Conclusion
Notwithstanding the numerous attempts at standardizing the scoring rules, the qualitative nature of scoring flow via visual inspection causes inter-observer
variability, and the semi-quantitative sensors (thermistors, nasal prongs, or thoraco-abdominal bands) used
to obtain uncalibrated signals for flow or effort, all
lead to a level of uncertainty when scoring hypopneas. And several unanswered questions still remain
regarding the final impact of using these variable
hypopnea definitions for the diagnosis of OSA. Therefore, we emphasize the importance of standardizing
the scoring of hypopneas across all sleep labs, regardless of their status of accreditation by the AASM.
Future research needs to focus on carefully delineating
the pathophysiological significance and long-term clinical
implications of the various hypopnea definitions and
hypopneas per se on neurocognitive, cardiovascular and
metabolic outcomes.
Abbreviations
AASM: American Academy of Sleep Medicine; AHI: Apnea-Hypopnea Index;
AI: Apnea Index; CI: Confidence Interval; COPD: Chronic Obstructive
Pulmonary Disease; CVE: Complex Ventricular Ectopy; EDS: Excessive daytime
sleepiness; EEG: Electroencephalogram; EMG: Electromyogram; ETCO2: Endtidal Carbon Dioxide; HAR: Hypopnea-to-apnea Ratio; HI: Hypopnea Index; IL6: Interleukin-6; MCA: Middle Cerebral Artery; MSL: Mean Sleep Latency;
ODT: Oxygen Desaturation Index; OSA: Obstructive Sleep Apnea;
OSAS: Obstructive Sleep Apnea Syndrome; PSG: Polysomnography;
RDI: Respiratory Disturbance Index; SAHS: Sleep Apnea-Hypopnea Syndrome;
SaO2/SpO2: Oxygen saturation; SDB: Sleep disordered breathing;
SE: Standard Error; TST: Total Sleep Time
Funding
No funding was provided for the development of this manuscript.
Availability of data and materials
Not applicable. This manuscript is a review of the literature. All articles that
were reviewed and referenced are available on PubMed.
Author’s contributions
All authors participated in the review of the literature and in the writing of
this manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Not applicable.
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
Competing interests
The authors declare that they have no competing interests.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Author details
1
VA Ann Arbor Heathcare Center and University of Michigan, 2215 Fuller Rd,
Ann Arbor, MI 48105, USA. 2Oakland University, Rochester, MI, USA. 3John D.
Dingell VA Medical Center and Wayne State University, Detroit, MI, USA.
Received: 23 October 2017 Accepted: 7 May 2018
References
Andreas S, et al. Doppler echocardiographic analysis of cardiac flow during the
Mueller manoeuver. Eur J Clin Investig. 1991;21(1):72–6.
Ayappa I, et al. Immediate consequences of respiratory events in sleep
disordered breathing. Sleep medicine. 2005;6(2):123–30.
Barbe F, et al. Effect of continuous positive airway pressure on the incidence of
hypertension and cardiovascular events in nonsleepy patients with obstructive
sleep apnea: a randomized controlled trial. JAMA. 2012;307(20):2161–8.
Berg S, et al. Comparison of direct and indirect measurements of respiratory
airflow: implications for hypopneas. Sleep. 1997;20(1):60.
Berry RB, et al. Rules for scoring respiratory events in sleep: update of the 2007
AASM manual for the scoring of sleep and associated events. J Clin Sleep
Med. 2012a;8(5):597–619.
Berry RB, et al. The AASM manual for the scoring of sleep and associated events.
Rules, Terminology and Technical Specifications. Darien, Illinois: American
Academy of Sleep Medicine; 2012b.
Bloch A, et al. Sleep apnea, hypopnea and oxygen desaturation in normal
subjects. N Engl J Med. 1979;300:513–7.
Bonnet MH. Performance and sleepiness as a function of frequency and
placement of sleep disruption. Psychophysiology. 1986;23(3):263–71.
Bradley TD, et al. Continuous positive airway pressure for central sleep apnea and
heart failure. N Engl J Med. 2005;353(19):2025–33.
Iber C, Ancoli-Israel S, Chesson AL Jr. Quan SF for the American Academy of
Sleep Medicine. The AASM manual for the scoring of sleep and associated
events: rules, terminology and technical specifications. 1st ed. Westchester, IL:
American Academy of Sleep Medicine; 2007.
Campos-Rodriguez F, et al. Impact of different hypopnea definitions on
obstructive sleep apnea severity and cardiovascular mortality risk in women
and elderly individuals. Sleep Med. 2016;27-28:54–8.
Chervin RD, Aldrich MS. Characteristics of apneas and hypopneas during sleep
and relation to excessive daytime sleepiness. Sleep. 1998;21(8):799–806.
Chowdhuri S, et al. Testosterone conversion blockade increases breathing
stability in healthy men during NREM sleep. Sleep. 2013;36(12):1793–8.
Ciftci TU, Kokturk O, Ozkan S. Apnea-hypopnea indexes calculated using different
hypopnea definitions and their relation to major symptoms. Sleep and
Breathing. 2004;8(03):141–6.
Cole P, Haight JS. Posture and the nasal cycle. Annals of Otology, Rhinology &
Laryngology. 1986;95(3):233–7.
Cowie MR, et al. Adaptive servo-ventilation for central sleep apnea in systolic
heart failure. N Engl J Med. 2015;373(12):1095–105.
Craig SE, et al. Continuous positive airway pressure improves sleepiness but not
calculated vascular risk in patients with minimally symptomatic obstructive sleep
apnoea: the MOSAIC randomised controlled trial. Thorax. 2012;67(12):1090–6.
Ding J, Nieto F, Beauchamp N Jr. Sleep-disordered breathing and white matter
disease in the brainstem in older adults. Sleep. 2004;27(3):474–9.
Duce B, Milosavljevic J, Hukins C. The 2012 AASM respiratory event criteria
increase the incidence of hypopneas in an adult sleep center population. J
Clin Sleep Med. 2015;11(12):1425–31.
Farre R, et al. A step forward for better interpreting the apnea-hypopnea index.
Sleep. 2015;38(12):1839–40.
Gould G, et al. The sleep hypopnea syndrome. Am Rev Respir Dis. 1988;137(4):895–8.
Guilleminault C, Hagen C, Huynh N. Comparison of hypopnea definitions in lean
patients with known obstructive sleep apnea hypopnea syndrome (OSAHS).
Sleep and Breathing. 2009;13(4):341–7.
Ho V, et al. Calibration model for apnea-hypopnea indices: impact of alternative
criteria for hypopneas. Sleep. 2015;38(12):1887–92.
Page 11 of 12
Hosselet J-J, et al. Classification of sleep-disordered breathing. Am J Respir Crit
Care Med. 2001;163(2):398–405.
Iber C. Are We Ready to Define Central Hypopneas? Sleep. 2013;36(3):305-306.
https://doi.org/10.5665/sleep.2434.
Iber C. Are we ready to define central hypopneas? Sleep. 36(3):363–8.
Iber C, et al. The AASM manual for the scoring of sleep and associated events:
rules, terminology and technical specifications: American Academy of Sleep
Medicine; 2007.
Kohler M, Stradling JR. Mechanisms of vascular damage in obstructive sleep
apnea. Nat Rev Cardiol. 2010;7(12):677–85.
Kushida CA, et al. Effects of continuous positive airway pressure on neurocognitive
function in obstructive sleep apnea patients: the apnea positive pressure longterm efficacy study (APPLES). Sleep. 2012;35(12):1593–602.
Leech JA, et al. A comparison of men and women with occlusive sleep apnea
syndrome. Chest. 1988;94(5):983–8.
Loredo JS, et al. Relationship of arousals from sleep to sympathetic nervous system
activity and BP in obstructive sleep apnea. CHEST Journal. 1999;116(3):655–9.
Loube DI, et al. Indications for positive airway pressure treatment of adult obstructive
sleep apnea patients: a consensus statement. Chest. 1999;115(3):863–6.
Mathew R, Castriotta RJ. High hypopnea/apnea ratio (HAR) in extreme obesity. J
Clin Sleep Med. 2014;10(4):391–6.
McEvoy RD, et al. CPAP for prevention of cardiovascular events in obstructive
sleep apnea. N Engl J Med. 2016;375(10):919–31.
McMillan A, et al. Continuous positive airway pressure in older people with
obstructive sleep apnoea syndrome (PREDICT): a 12-month, multicentre,
randomised trial. Lancet Respir Med. 2014;2(10):804–12.
Mehra R, et al. Nocturnal arrhythmias across a spectrum of obstructive and
central sleep-disordered breathing in older men: outcomes of sleep disorders
in older men (MrOS sleep) study. Arch Intern Med. 2009;169(12):1147–55.
Moser NJ, et al. What is hypopnea, anyway? Chest. 1994;105(2):426–8.
Myllymaa K, et al. Effect of oxygen desaturation threshold on determination of
OSA severity during weight loss. Sleep and Breathing. 2016;20(1):33–42.
Netzer N, et al. Blood flow of the middle cerebral artery with sleep-disordered
breathing correlation with obstructive hypopneas. Stroke. 1998;29(1):87–93.
Parra O, et al. Efficacy of continuous positive airway pressure treatment on 5-year
survival in patients with ischaemic stroke and obstructive sleep apnea: a
randomized controlled trial. J Sleep Res. 2015;24(1):47–53.
Peker Y, et al. Effect of positive airway pressure on cardiovascular outcomes
in coronary artery disease patients with nonsleepy obstructive sleep
apnea. The RICCADSA randomized controlled trial. Am J Respir Crit Care
Med. 2016;194(5):613–20.
Punjabi NM, et al. Sleep-disordered breathing and cardiovascular disease: an outcomebased definition of hypopneas. Am J Respir Crit Care Med. 2008;177(10):1150–5.
Randerath WJ, et al. Evaluation of a noninvasive algorithm for differentiation of
obstructive and central hypopneas. Sleep. 2013;36(3):363–8.
Redline S, Sanders M. Hypopnea, a floating metric: implications for prevalence,
morbidity estimates, and case finding. Sleep. 1997;20(12):1209–17.
Redline S, et al. Effects of varying approaches for identifying respiratory disturbances
on sleep apnea assessment. Am J Respir Crit Care Med. 2000;161(2):369–74.
Redline S, et al. Obstructive sleep apnea–hypopnea and incident stroke: the sleep
heart health study. Am J Respir Crit Care Med. 2010;182(2):269–77.
Rowley J, et al. The determinants of the apnea threshold during NREM sleep in
normal subjects. Sleep. 2006;29(1):95–103.
Ruehland W, et al. The New AASM Criteria for Scoring Hypopneas: Impact on the
Apnea Hypopnea Index. Sleep 32: 150–157, 2009. Year Book of Pulmonary
Disease. 2010;2010:244–5.
Shahar E, et al. Sleep-disordered breathing and cardiovascular disease: crosssectional results of the sleep heart health study. Am J Respir Crit Care Med.
2001;163(1):19–25.
Somers VK, et al. Sympathetic-nerve activity during sleep in normal subjects. N
Engl J Med. 1993;328(5):303–7.
Stamatakis K, et al. Fasting glycemia in sleep disordered breathing: lowering the
threshold on oxyhemoglobin desaturation. Sleep. 2008;31(7):1018–24.
Sulit L, Storfer-Isser A, Kirchner H. Differences in poly-somnography
predictors for hypertension and impaired glucose tolerance. Sleep.
2006;29(6):777–83.
Thomas RJ. Sleep fragmentation and arousals from sleep—time scales,
associations, and implications. Clin Neurophysiol. 2006;117(4):707–11.
Tsai WH, et al. A comparison of apnea–hypopnea indices derived from different
definitions of hypopnea. Am J Respir Crit Care Med.
1999;159(1):43–8.
Shamim-Uzzaman et al. Sleep Science and Practice (2018) 2:7
Ward NR, et al. The effect of respiratory scoring on the diagnosis and
classification of sleep disordered breathing in chronic heart failure. Sleep.
2013;36(9):1341–8.
Whyte K, et al. Accuracy and significance of scoring hypopneas. Sleep. 1992;
15(3):257–60.
Yu J, et al. Association of Positive Airway Pressure with Cardiovascular Events and
Death in adults with sleep apnea: a systematic review and meta-analysis.
JAMA. 2017;318(2):156–66.
Zhou XS, et al. Effect of testosterone on the apneic threshold in women during
NREM sleep. J Appl Physiol. 2003;94(1):101–7.
Page 12 of 12