Sleep Breath (2005) 9: 73–81
DOI 10.1007/s11325-005-0017-y ORIGINAL ARTICLE
José Haba-Rubio
Georges Darbellay
François R. Herrmann
Jean G. Frey
Alda Fernandes
Jean M. Vesin
Jean P. Thiran
Jean M. Tschopp
Published online: 4 May 2005
# Springer-Verlag 2005
Obstructive sleep apnea syndrome: effect
of respiratory events and arousal on pulse
wave amplitude measured by
photoplethysmography in NREM sleep
Abstract The objective of the study
is to evaluate changes in finger pulse
wave amplitude (PWA), as measured
by photoplethysmography, and heart
rate (HR), related to obstructive
respiratory events and associated
arousals during sleep. We analyzed
1,431 respiratory events in NREM
sleep from 12 patients according to (1)
the type of event (apnea, hypopnea,
upper airway resistance episode) and
(2) the duration of the associated EEG
arousal (>10, 3–10, <3 s). Obstructive
respiratory events provoked a relative
bradycardia and vasodilation followed
by HR increase and vasoconstriction.
Relative PWA changes were signifi-
cantly greater than HR changes. These
responses differed significantly
according to EEG-arousal grades
(time×arousal interaction, p<0.0001),
with longer arousals producing greater
responses, but not to the type of
respiratory event (time×event interac-
tion, p = ns). Obstructive respiratory
events provoke HR and PWA changes,
the magnitude seemingly related to the
intensity of central nervous activation,
with PWA changes greater than HR.
PWA obtained from a simple pulse
oxymeter might be a valuable method
to evaluate sleep fragmentation in
sleep breathing disorders.
Keywords Obstructive sleep apnea .
Arousal . Autonomic markers . O2
pulse wave amplitude
Introduction
Sleep-disordered breathing (SDB) covers a spectrum from
complete upper airway obstruction [obstructive sleep apnea
(OA)] [1] to more subtle breathing abnormalities [ob-
structive hypopnea (OH), upper airway resistance episodes
(UARE)] [2]. The termination of sleep respiratory events is
thought to depend on arousal from sleep, which restores
upper airway patency [3–5]. The arousal response is im-
portant for survival of the affected patient; however, the
sleep fragmentation induced by repetitive arousals and the
physiological events that surround the arousal response
also have adverse consequences which contribute to the
pathophysiology and clinical manifestations of SDB [5].
Arousals are characterized by abrupt changes in central
nervous system activity that produce electroencephalogram
(EEG) activation and marked changes in a wide range of
autonomic nervous system parameters such as heart rate
(HR), blood pressure (BP), ventilation and peripheral vas-
cular resistance [6–10]. Arousals are typically evident and
conventionally detected as changes in cortical EEG ac-
tivity, a labor-intensive approach associated with high inter-
scorer variability [11]. In addition, according to the American
Sleep Disorders Association (ASDA) definition [12], only
about 75% of respiratory events are terminated by an
EEG arousal [13], but it has been shown that stimuli that
produce measurable cardiovascular disturbances without
arousal appear sufficient to produce daytime sleepiness
[14]. The term “autonomic arousal” is used to denote tran-
sient changes during sleep in autonomic parameters which
are not necessarily associated with EEG arousals, and there
is, consequently, considerable interest in using convenient,
non-invasive autonomic markers of sleep fragmentation,
particularly in SDB [15].
Parameters derived from peripheral vasoconstrictor re-
sponses appear to show dramatic changes following arousal
J. Haba-Rubio . J. G. Frey . A. Fernandes .
J. M. Tschopp (*)
Sleep Laboratory,
Centre Valaisan de Pneumologie,
3963 Crans-Montana, Switzerland
e-mail: jean-marie.tschopp@admin.vs.ch
Fax: +41-27-6038181
G. Darbellay . J. M. Vesin . J. P. Thiran
Institute for Signal Processing Laboratory,
Swiss Federal Institute of Technology,
Lausanne, Switzerland
F. R. Herrmann
Departments of Rehabilitation and
Geriatrics, University Hospitals,
Geneva, Switzerland

[6, 10]. Pulse wave amplitude (PWA) measured by photo-
plethysmography may have several advantages over other
cardiovascular markers of arousal; thus, it can easily be
recorded using most conventional pulse oxymeters already
in sleep-laboratory use or in ambulatory screening devices.
In previous studies, changes in PWA were measured by
photoplethysmography in relation to arousals induced by
acoustical stimulus in healthy subjects [6, 10]. In sleep-
related breathing disorders, blood gas disturbances and
other factors potentially further modulate activation-related
autonomic activity [16, 17] and could modify this cardio-
vascular response. If reliable changes in PWA could be
detected in relation to obstructive sleep respiratory events,
the photoplethysmography could become an interesting
screening method for detecting respiratory events and for
assessing sleep fragmentation in obstructive sleep apnea
patients.
The first aim of this preliminary study was, then, to
determine if changes in PWA related to sleep respiratory
events could be detected by photoplethysmography and to
compare the relative PWA magnitude changes with changes
in HR. The second aim was to determine if these changes
were related to the nature of the stimuli (i.e., type of
respiratory event) or if they were proportional to the in-
tensity of the associated arousal.
Materials and methods
Patients
We analyzed data from 12 patients referred to our labo-
ratory for evaluation because of presumed SDB. To qualify
for the study, the patients had to fulfill the following criteria:
(1) Nocturnal polysomnography performed using esoph-
ageal pressure recording without technical problems
during the monitoring and without artifacts on EEG
tracings
(2) A confirmed diagnosis of SDB with a respiratory dis-
turbance index >30 respiratory events per hour of
sleep, at the end of the diagnostic procedure. We delib-
erately chose six patients who essentially presented OA
and OH and six patients presenting a combination of
OH and UARE
(3) Absence of unstable medical condition.
Patients were not considered for the study if they were
taking psychoactive, cardiac, antihypertensive or drugs in-
terfering with the autonomic system. They were informed
that some of the collected data would be used for research
purposes, and they gave written informed consent.
Measurements
Overnight polysomnography was performed according to
standard laboratory protocol, with MAP polysomnography
system (Monet 27 channels, MAP, Germany), between
2230 and 0600 h the following day. Signals recorded were
as follows: EEG (C3–A2, O1–A2, C4–A1, O2–A1, sam-
pling rate 125 Hz), left and right electro-oculogram (EOG,
sampling rate 100 Hz), submental electromyogram (EMG,
sampling rate 100 Hz), arterial blood oxygen saturation
(SaO2) and finger photoplethysmogram (using Nonin
XPOD3012 oxymeter, Nonin Medical Inc., Plymouth, MN,
USA, sampling rates 20 and 80 Hz, respectively), electro-
cardiogram (ECG, sampling rate 800 Hz), chest and ab-
dominal movements (sampling rate 20 Hz), esophageal
pressure (using a Jaeger esophagus catheter connected to a
pressure transducer, sampling rate 50 Hz) and surface EMG
of the anterior tibialis muscles (sampling rate 100 Hz).
Respiratory airflow was monitored with a nasal cannula
connected to a pressure transducer (Protech2, Minneapolis,
MN, USA, sampling rate 20 Hz). Snoring was measured by
a microphone placed at the suprasternal notch. A body
position sensor attached to a thoracic belt was used to
monitor body position. Data were digitized, amplified and
evaluated using sleep-stage software (Analysis Manager
7.18, MAP).
Data analysis
Sleep staging, arousal and respiratory events scoring
Sleep staging was performed using the criteria of
Rechtschaffen and Kales [18] for epochs of 20 s.
In a first phase, respiratory events were scored by a
single trained observer (J.H.) blinded to the cardiovascular
channels and using standard criteria [1], irrespective of the
presence or absence of EEG arousal. An OA was defined
as a complete cessation of airflow for more than 10 s. An
OH was defined as a 50% or greater reduction in airflow
from the baseline value lasting at least 10 s or a clear
amplitude airflow reduction lasting >10 s and associated
with either an oxygen desaturation of >3% or an arousal.
An UARE was defined as a progressive change in the
shape of the inspiratory flow contour (characterized by
increasing limitation) and of an increase in respiratory
effort (in a crescendo pattern), in the absence of decrease
in airflow, with termination of the event as abrupt nor-
malization of both inspiratory flow contour and the esoph-
ageal pressure swing. All obstructive respiratory events
detected from stage 2 NREM sleep were included for
cardiovascular analysis.
In a second phase, EEG changes related to the respi-
ratory events were scored by the same investigator. Again,
EEG changes related to all respiratory events from stage 2
74