Central sleep apnea indicates autonomic dysfunction in asymptomatic carotid stenosis: a potential marker of cerebrovascular and cardiovascular risk.

Abstract

Arteriosclerosis related stenosis in the carotid bulb causes autonomic imbalance, likely due to carotid chemoreceptor and baroreceptor dysfunction. The latter are associated with increased cerebrovascular and cardiovascular mortality. Chemoreceptor and baroreceptor dysfunction is also involved in the origin of central sleep apnea syndrome (CSA) in different clinical entities. We hypothesized that CSA is associated with stenosis of the internal carotid artery (ICA). The mechanism of this association is an autonomic imbalance induced by stenosis-mediated chemoreceptor and baroreceptor dysfunction.
Cross-sectional prospective study.
University-based tertiary referral sleep clinic and research center.
Fifty-nine patients with various degrees of asymptomatic extracranial ICA (elCA) (n = 49) and intracranial ICA (ilCA) stenosis (n = 10) were investigated.
Polysomnography to detect CSA and analysis of spontaneous heart rate variability (HRV) to detect autonomic imbalance.
CSA occurred in 39% of the patients with elCA stenosis but was absent in patients with ilCA stenosis. CSA was present in patients with severe elCA stenosis of > or = 70% on one side. Independent predictors for CSA were severity of stenosis, asymmetric distribution of stenosis between both elCA and autonomic imbalance, namely a decrease of parasympathetic tone. The specific constellation of HRV-parameters indicated increased chemoreceptor sensitivity and impaired baroreflex sensitivity.
CSA indicates autonomic dysfunction in patients with asymptomatic elCA stenosis. Detection of CSA may help to identify asymptomatic patients with an increased risk of cerebrovascular or cardiovascular events who particularly benefit from carotid revascularization.

Full text

Page 1

SLEEP, Vol. 33, No. 3, 2010
327
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
Carotid artery disease is a major Cause of
isChemiC stroke. in partiCular, stenosis of the
internal Carotid artery (iCa) is responsible
for 30% of ischemic strokes.1 the most common cause of carotid
stenosis is arteriosclerosis.2 the predilection site for extracranial
carotid arteriosclerotic plaque formation is the proximal iCa seg-
ment (carotid bulb).3 Carotid arteries at this level contain periph-
eral chemoreceptors (carotid bodies) and arterial baroreceptors
which are important determinants of ventilatory and short-time
blood pressure regulation.4-6 Traditional views of risk stratifi-
cation of carotid stenosis focus on cerebral hypoperfusion or
thromboembolism.7,8 however, there is growing evidence that au-
tonomic imbalance due to carotid sensor maladaption contributes
to acute cerebrovascular and cardiovascular events.9 particularly
stenosis of the proximal iCa segment may interfere with the ca-
rotid chemoreceptor and baroreceptor function. Concordantly,
carotid stenosis is associated with impaired baroreflex sensitivity
and decreased heart rate variability (hrV).10 depressed barore-
ceptor sensitivity and decreased hrV are associated with an in-
creased cardiovascular and cerebrovascular mortality after stroke
or myocardial infarction.11,12 indices of hrV are closely related to
baroreceptor and chemoreceptor sensitivity.13,14
disturbances in carotid chemoreceptor and baroreceptor func-
tion have also been demonstrated in patients with central sleep ap-
nea syndrome (Csa).15,16 CSA is defined by recurrent episodes of
apnea during sleep resulting from a temporary loss of ventilatory
effort. Csa is associated with an increased cardiovascular risk.17
autonomic dysfunction in iCa stenosis seems to be secondary
to carotid sensor maladaption.9,10 the latter is also involved in the
development of Csa (figure 1).15,16 the insertion of long plastic
tubes into the carotid arteries of dogs induces periodic breathing,
probably caused by the impairment of carotid chemoreceptor and/
or baroreceptor function.18 although these authors assumed a
prolonged heart-to-brain circulation time as the cause of periodic
breathing, a close relationship between a prolonged circulation
time and the occurrence of Csa has not been demonstrated in
human beings.19
the aim of the present study was to detect the prevalence and
characteristics of Csa in patients with asymptomatic iCa steno-
sis. We hypothesized that the occurrence of Csa depends on the
severity and distribution of iCa stenosis. We assumed that the
development of Csa is attributable to disturbances in autonomic
balance. therefore, we performed polysomnographic studies and
analyzed spontaneous hrV in patients with various degrees of
extracranial and intracranial iCa stenosis.
METHODS
Patients
the study was approved by the ethics committee of the
friedrich schiller university. Consecutive patients with variable
degrees of asymptomatic iCa stenosis examined in our neuro-
CENTRAL SLEEP APNEA AND CAROTID STENOSIS
Central Sleep Apnea Indicates Autonomic Dysfunction in Asymptomatic Carotid
Stenosis: A Potential Marker of Cerebrovascular and Cardiovascular Risk
Sven Rupprecht, MD; Dirk Hoyer, PhD; Georg Hagemann, PhD; Otto W. Witte, PhD; Matthias Schwab, PhD
Department of Neurology, Friedrich Schiller University, Jena, Germany
Study Objectives: Arteriosclerosis related stenosis in the carotid bulb causes autonomic imbalance, likely due to carotid chemoreceptor and
baroreceptor dysfunction. The latter are associated with increased cerebrovascular and cardiovascular mortality. Chemoreceptor and baroreceptor
dysfunction is also involved in the origin of central sleep apnea syndrome (CSA) in different clinical entities. We hypothesized that CSA is associ-
ated with stenosis of the internal carotid artery (ICA). The mechanism of this association is an autonomic imbalance induced by stenosis-mediated
chemoreceptor and baroreceptor dysfunction.
Design: Cross-sectional prospective study.
Setting: University-based tertiary referral sleep clinic and research center.
Patients: Fifty-nine patients with various degrees of asymptomatic extracranial ICA (eICA) (n = 49) and intracranial ICA (iICA) stenosis (n = 10)
were investigated.
Interventions: Polysomnography to detect CSA and analysis of spontaneous heart rate variability (HRV) to detect autonomic imbalance.
Measurements and Results: CSA occurred in 39% of the patients with eICA stenosis but was absent in patients with iICA stenosis. CSA was pres-
ent in patients with severe eICA stenosis of ≥ 70% on one side. Independent predictors for CSA were severity of stenosis, asymmetric distribution of
stenosis between both eICA and autonomic imbalance, namely a decrease of parasympathetic tone. The specific constellation of HRV-parameters
indicated increased chemoreceptor sensitivity and impaired baroreflex sensitivity.
Conclusions: CSA indicates autonomic dysfunction in patients with asymptomatic eICA stenosis. Detection of CSA may help to identify asymptom-
atic patients with an increased risk of cerebrovascular or cardiovascular events who particularly benefit from carotid revascularization.
Keywords: Central sleep apnea, carotid stenosis, autonomic dysfunction, heart rate variability
Citation: Rupprecht S; Hoyer D; Hagemann G; Witte OW; Schwab M. Central sleep apnea indicates autonomic dysfunction in asymptomatic
carotid stenosis: a potential marker of cerebrovascular and cardiovascular risk. SLEEP 2010;33(3):327-333.
Submitted for publication April, 2009
Submitted in final revised form December, 2009
Accepted for publication December, 2009
Address correspondence to: Sven Rupprecht, MD, Dept. of Neurology,
Friedrich Schiller University, Erlanger Alle 101, 07740 Jena, Germany; Tel:
+49-3641-9-323480; Fax: +49-3641-9-323402; E-mail: Sven.Rupprecht@
med.uni-jena.de

Page 2

SLEEP, Vol. 33, No. 3, 2010
328
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
vascular laboratory were recruited for the study. We assigned
the patients into two groups: extracranial iCa (eiCa) stenosis
and isolated (without additional eiCa stenosis) intracranial iCa
(iiCa) stenosis. age, gender, body mass index (bmi), presence
of coronary artery disease, diabetes mellitus, and antihyperten-
sive medication of each patient were documented. Written in-
formed consent was obtained.
Extracranial ICA stenosis group
to grade the severity of eiCa stenosis we developed an ul-
trasound based (philips enVisor hd, eindhoven, netherlands)
eiCa stenosis score (eiCa score, table 1).20 the ratio between
the higher and lower scores of both sides (eiCa ratio) was cal-
culated to assess the asymmetric distribution of arteriosclerotic
changes. all stenoses were localized in the proximal iCa seg-
ment including the carotid bulb and the first centimeter distal to
the origin of the eiCa.
Intracranial ICA stenosis group
Carotid siphon stenoses were also assessed ultrasonographi-
cally. severity was graded using an iiCa score according to
table 1.
Exclusion criteria
patients with additional vertebral artery stenosis or occlusion
were excluded from the study. patients with Csa underwent
a diagnostic work-up, including mri scan of the brain, cervi-
cal and intracranial mr angiography, body plethysmography,
daytime capillary blood gas analysis, eCG, and transthoracic
echocardiography. We excluded patients with diseases that pre-
dispose to Csa such as heart failure, previous stroke, pulmo-
nary hypertension, interstitial or obstructive lung disease, and
cardiac rhythm disorders. patients were also excluded if they
exhibited a left ventricular ejection fraction (lVef) < 50%,
forced expiratory volume in one second (feV1) and vital ca-
pacity < 80%, po2 < 60 mm hg and pCo2 ≥ 45 mm Hg.
Apnea Detection
standard polysomnography was performed on all patients
(somnoscreen, somnomedics, randersacker, Germany).
sleep was staged by personnel blinded to the protocol using
standard techniques for scoring sleep stages and arousals.21,22
Central apnea was defined by the absence of tidal volume
excursion for ≥ 10 sec. Central hypopneas were defined as ≥
30% reduction in tidal volume or thoracoabdominal movement
≥ 10 sec with proportional thoracoabdominal in-phase move-
ments followed by ≥ 4% oxyhemoglobin desaturation and/or an
arousal.23 Obstructive apnea/ hypopnea was defined similarly,
except that thoracoabdominal out-of-phase motions had to be
present. mixed apneas including central and obstructive com-
ponents were classified as obstructive events because of their
uncertain etiology.
total apnea-hypopnea index (ahi), central apnea-hypopnea
index (cahi) and obstructive apnea-hypopnea index (oahi)
were defined as number of respective events per hour of sleep.
Obstructive sleep apnea (OSA) was diagnosed at an oAHI ≥ 5
per hour, CSA at a cAHI ≥ 5 per hour with ≥ 50% of total AHI
consisting of central events.23
Heart Rate Variability
We used eCG tracks sampled at 256 hz from polysomnogra-
phy for hrV analysis, because nocturnal hrV is more closely
associated with baroreflex sensitivity than HRV at wakeful-
ness.13 ECG samples were collected from first stable NREM
sleep stage 2 after sleep onset to avoid influences of different
sleep stages and/or sleep related respiratory events on hrV pa-
rameters. r-r intervals were calculated from artifact and apnea/
hypopnea-free periods of 10 min by the detection of Qrs com-
plexes using a steepest ascent criterion. equidistant time series
were calculated by linear interpolation and resampling (200 hz,
after anti-aliasing filtering). High frequency (HF: 0.15-0.4 Hz),
low frequency (lf: 0.04-0.15 hz), very low frequency (Vlf:
0.003-0.04 hz), and total band power (tp: 0.003-0.4 hz) were
calculated by fast fourier transformation.24
Table 1—Score system to determine the degree of ICA stenosis and
group assessment.
Extracranial ICA Stenosis
Ultrasonographic Assessment
No Arteriosclerosis
Arteriosclerosis without Stenosis
< 50%
Mild–Moderate Stenosis
50%
60%
Severe Stenosis or Occlusion
70%
80%
90%
Occlusion
Score
0
1
2
3
4
5
6
7
Intracranial ICA Stenosis
Ultrasonographic Assessment
No Arteriosclerosis
Mild Stenosis
Moderate Stenosis
Severe Stenosis
0
1
2
3
The scores were determined in both ICA and added to form a total score
ranging from 0 to 14 for eICA and from 0 to 6 for iICA arteriosclerosis/
stenosis.
Figure 1—Hypothetic mechanisms for the development of CSA and
autonomic dysfunction in eICA stenosis.

Page 3

SLEEP, Vol. 33, No. 3, 2010
329
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
the band powers were logarithmized to obtain the normally
distributed parameters lnVlf, lnlf, and lnhf. hf and lf band
power normalized as percentage to tp minus Vlf band power
were assigned as hfnu and lfnu (normalized units, nu). Vlf
band power normalized to tp was assigned as Vlfnu.
HF band power reflects the parasympathetic component of
autonomic cardiovascular regulation and partially the vagal
modulation of baroreflex activity. LF band power is modu-
lated by the parasympathetic and sympathetic system. the lf/
hf ratio is considered as a marker of sympathovagal balance,
i.e., an increase in LF/HF ratio reflects an increased sympa-
thetic and/or decreased parasympathetic tone.25 the underly-
ing mechanisms influencing VLF band power are still unclear
but there is strong evidence for a relationship between Vlf
oscillations and peripheral chemoreceptor acitivity.26,27 altera-
tions in hrV are associated with chemoreceptor and barore-
ceptor dysfunction in Csa/ Cheyne-stokes respiration 26,27 as
well as baroreceptor dysfunction in eiCa arteriosclerosis.10,28
Statistical Analysis
We examined the hypothesis that the stenosis dependent
extent of Csa is autonomically mediated by the following ar-
gumentation line:
the extent of Csa (cahi) in dependence on markers of au-
tonomic function (lnVlf, lnlf, lnhf) was evaluated by mul-
tivariate linear regression models, whereas the contribution of
the hrV parameters was evaluated using a stepwise build pro-
cedure (inclusion criterion P < 0.05) by regression coefficient
with 95% confidence interval, standardized regression coef-
ficient (β) and P-value. The dependence of CSA on stenosis
characteristics (eiCa score, eiCa ratio) was similarly evalu-
ated. finally, a multivariate model based on autonomic and
stenosis parameters was built up. due to the limited number of
patients, we restricted the number of covariates in the respec-
tive multivariate models to a maximum of 5 parameters.
bmi, age, oahi, diabetes, coronary artery disease, and anti-
hypertensive medication were not considered as confounders in
the models, as none of them significantly contributed to cAHI.
inter-individual differences in age, bmi, cahi, oahi, mean
spo2, eiCa score, eiCa ratio, and hrV parameters were as-
sessed by the mann-Whitney u-test. differences in the eiCa
score for the left and right eiCa were assessed by the Wil-
coxon signed-rank test; differences in distribution of stenoses
(unilateral vs. bilateral) by the binomial test; differences in
gender and the presence of confounders as diabetes mellitus,
coronary artery disease, and antihypertensive medication by
the 2-tailed fisher exact test. the spearman rank correlation
coefficient was used to study the relationships between cAHI
or oahi and eiCa score, eiCa ratio, lnVlf, lnlf, lnhf, age,
and bmi in patients with eiCa stenosis. the patient charac-
teristics were provided as mean ± standard deviation. p < 0.05
was considered significant.
RESULTS
Patients
fifty-nine patients were recruited for the study. age ranged
between 43 and 86 years (mean age: 65.0 ± 11.0 y). the eiCa
stenosis group included 49 patients (36 men, 13 women). ar-
teriosclerosis without relevant stenosis was present in 17 pa-
tients. thirty-two patients had relevant eiCa stenosis: 7 had
mild-to-moderate eiCa stenosis, and 25 had severe stenoses
(table 2). stenoses were distributed unilaterally in 12 patients
and bilaterally in 20 patients. patients with severe eiCa steno-
sis were older than patients without eiCa stenosis (p < 0.05)
but did not differ in bmi, oahi, or mean spo2 over the night
(table 2).
the iiCa stenosis group included 10 patients (5 men, 5
women) with isolated moderate-to-severe carotid siphon
stenosis. stenoses were unilateral in 4 patients and bilateral in
6 patients (table 3).
Prevalence of CSA
none of the 10 patients with iiCa stenosis showed Csa,
while 7 patients (70%) had osa. Csa was present in 19 (39%)
and osa in 16 (33%) of 49 patients with eiCa stenosis (table
2). ten of 19 patients with Csa (53%) also had obstructive
events, but central apnea was predominant. patients with Csa
were predominately men (p < 0.008) and older than patients
without Csa (p < 0.02) but did not show differences in bmi,
oahi, or mean spo2 (table 3). the degree of stenosis was
higher in patients with than without Csa (p < 0.001, table
3) and an independent predictor for Csa in the multivariate
regression model (p < 0.002, table 4).
Csa was associated with nrem sleep (p < 0.04). Capillary
blood gas analysis showed normoxic and normocapnic venti-
lation conditions at daytime (ph: 7.43 ± 0.0, po2: 69.2 ± 12.9
mm hg, pCo2: 39.2 ± 2.2 mm hg, base excess: 2.0 ± 1.6).
Table 2—Prevalence of sleep related breathing disorders and demographic and polysomnographic parameters in relation to the degree of extracranial and
intracranial ICA stenosis.
Patients
CSA
Patients
n = 0 (0%)
n = 1 (14%)
n = 18 (72%) n = 4 (16%)
OSA
Patients
n = 8 (47%)
n = 3 (43%)
Age
Years
61.7 ± 9.4*
61.7 ± 8.6
68.2 ± 10.1
BMI
kg/m2
27.8 ± 5.2
26.7 ± 3.1
29.6 ± 0.1
cAHI
Events/h
1.7 ± 1.5**
3.4 ± 1.7**
15.3 ± 15.7**
oAHI
Events/h
9.1 ± 8.4
5.5 ± 3.5
6.7 ± 5.0
Mean SpO2
%
92.7 ± 2.1
92.3 ± 0.9
92.0 ± 1.4
Extracranial ICA (eICA)
Arteriosclerosis without Stenosis
Mild—Moderate Stenosis
Severe Stenosis
Intracranial ICA (iICA)
Moderate—Severe Stenosis
n = 17
n = 7
n = 25
n = 10 n = 0 (0%) n = 7 (70%) 59.9 ± 12.828.4 ± 2.7 2.4 ± 2.315.8 ± 20.2 92.2 ± 4.3
*P < 0.05 compared to the severe eICA stenosis group
**P < 0.05 compared to the group with the next lower degree of eICA stenosis

Page 4

SLEEP, Vol. 33, No. 3, 2010
330
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
Degree of Extracranial ICA Stenosis and CSA
the cahi but not the oahi increased with the degree of
eiCa stenosis, particularly, in patients with severe eiCa steno-
sis (p < 0.05, table 2, figure 2). Consequently, the degree of
eiCa stenosis (eiCa score) was positively correlated to cahi
(r = 0.58, p < 0.01, figure 2).
Localization of Extracranial ICA Stenosis and CSA
in a second analysis, we asked whether the localization (right
vs. left) and the distribution of arteriosclerotic lesions in both
eiCa (unilateral vs. bilateral, symmetry of stenoses) contribute
to Csa development. Csa was not associated with left or right
eiCa stenosis or unilateral or bilateral eiCa stenosis (table 3).
however, Csa was associated with asymmetric distribution of
bilateral stenoses. thus, the eiCa ratio was higher in patients
with than without Csa (p < 0.001, table 3) and positively cor-
related to cahi (r = 0.61, p < 0.01). the eiCa ratio was an
independent predictor of Csa in the multivariate linear regres-
sion analysis (p < 0.001, table 4).
Extracranial ICA Stenosis, CSA, and Autonomic Function
We also asked whether Csa is associated with autonomic
dysregulation. eight of 49 patients with eiCa stenosis were ex-
cluded because of insufficient ECG recording or the absence of
apnea-free periods at sleep onset for hrV analysis.
Cardiac Vlf band power (p < 0.01) and lf/hf ratio (p < 0.02)
were increased in patients with Csa compared to patients with-
out Csa. lf band power tended to be higher in patients with
CSA. HF band power was significantly reduced in patients with
Csa (p < 0.03, table 5). Vlf band power (r = 0.45, p < 0.01)
and lf/hf ratio (r = 0.34, p < 0.05) were positively correlated
to cAHI. No significant correlations were found between LF or
hf band power and cahi.
Vlf (p < 0.005) was associated with cahi in the univariate
regression model (table 4). asymmetric distribution (eiCa ra-
tio, p < 0.001), degree of eiCa stenosis (eiCa score, p < 0.002),
and hf band power (p < 0.013), but not Vlf band power, were
independent predictors of cahi in the stepwise multivariate re-
Table 3—Demographic, polysomnographic and vascular parameters in
patients with ICA stenosis with and without CSA
Extracranial ICA
Patients
Age (y)
Men: Women
BMI (kg/m2)
Diabetes Mellitus
Coronary Artery Disease
Antihypertensive Medication
β Blockers
Calcium Blockers
Diuretics
ACE + AT II Inhibitors
Polysomnographic Data
AHI (Events/h)
Total
Central
Obstructive
SpO2 Mean (%)
Vascular Data
Degree of Stenoses (eICA
Score)
Total
Left
Right
Ratio Between of Higher and
Lower Score (eICA Ratio)
Localization of Stenoses
(Number of Stenoses)
Unilateral Stenoses
Bilateral Stenoses
Intracranial ICA
Patients
Age (y)
Men: Women
BMI (kg/m2)
Polysomnographic Data
AHI (Events/h)
Total
Central
Obstructive
SpO2 Mean (%)
Vascular Data
Degree Of Stenoses (iICA Score)
Total
Left
Right
Localization Of Stenoses
(Number of Stenoses)
Unilateral Stenoses
Bilateral Stenoses
CSA
n = 19
68.9 ± 9.7
18:1
28.9 ± 2.6
n = 6
n = 6
Non-CSA P-Value
n = 30
62.6 ± 9.6
18:12
28.4 ± 9.9
n = 5
n = 6
0.02
0.008
n.s
n.s
n.s
n = 10
n = 4
n = 8
n = 15
n = 12
n = 5
n = 9
n = 15
n.s
n.s
n.s
n.s
27.1 ± 16.8
20.1 ± 15.2
7.1 ± 5.5
91.9 ± 1.5
9.6 ± 7.7 < 0.001
2.0 ± 1.6 < 0.001
7.6 ± 6.8
92.5 ± 1.6
n.s
(0.07)
8.1 ± 2.6
4.5 ± 2.1
3.5 ± 2.1
3.9 ± 2.6 < 0.001
1.9 ± 1.6 < 0.001
1.9 ± 1.4 0.02
2.9 ± 2.0 1.6 ± 1.1 < 0.001
n = 5
n = 13
n = 7
n = 7
n.s
n.s
n = 10
59.9 ± 12.8
5:5
28.4 ± 2.7
18.2 ± 20.3
2.4 ± 2.3
15.8 ± 20.2
92.2 ± 4.3
4.2 ± 1.6
1.9 ± 1.4
2.3 ± 0.9
n = 4
n = 6
Figure 2—Relationship between the severity of eICA arteriosclerosis/
stenosis and CSA. Circles reflect individual patients.

Page 5

SLEEP, Vol. 33, No. 3, 2010
331
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
Csa was associated with a loss of parasympathetic tone
as indicated by the reduction in the hf band power and the
increase in the lf/hf ratio. the increase in the lf/hf ratio
in parallel with the upward trend in lf band power may also
suggest augmentation of sympathetic activity. furthermore, an
increase in Vlf band power was present in our patients with
Csa. such a shift in sympathovagal balance and an increase in
Vlf band power has also been demonstrated in patients with
CSA/ Cheyne-Stokes respiration reflecting enhanced chemo-
sensitivity and down-regulation of baroreceptor function.26,27,40
studies examining the chemosensitivity in patients with eiCa
stenoses are absent. at least our study allows some conclusions
regarding alterations of chemosensitivity in these patients. the
VLF band power, which reflects chemosensitivity26,27, was the
strongest autonomic predictor of Csa in our patients. it indi-
cates a stenosis-related enhancement of peripheral chemosen-
sitivity. there is some evidence that Vlf oscillations can also
be influenced by CSA related ventilatory oscillations.41,42 in our
study, eCG for hrV analysis were taken only from Csa-free
intervals. therefore, we can assure that slow oscillations in
ventilation did not contribute to hrV changes particularly to
increased Vlf band power.
it has been shown that decreased parasympathetic activity
and impaired baroreflex sensitivity are present in patients with
eiCa stenosis.10 in agreement with this, hf band power was
reduced in our patients. loss of baroreceptors sensitivity may
additionally facilitate chemoreceptor sensitivity and sympa-
thetic overactivity by central baroreflex-chemoreflex interac-
tions (figure 3).43-45
taken together, increased chemosensitivity caused by isch-
emia of the peripheral chemoreceptors and/or baroreflex dys-
function is highly likely to be involved in the development of
Csa in eiCa stenosis (figure 3). although some studies sug-
gest that Csa contributes to autonomic dysfunction by frequent
nocturnal apnea-related arousals and hypoxemia,46 our findings
indicate that Csa is rather the consequence than the cause of
the autonomic dysfunction in agreement with studies in heart
failure.47-49
Given the involvement of carotid sensors in the development
of Csa, association of Csa with an asymmetric distribution of
gression model (Table 4). The insufficiency of the VLF band
power to be an independent predictor is due to its close correla-
tion with the hf band power (r = 0.52, p < 0.01), eiCa ratio
(r = 0.33, p < 0.05) and eiCa score (r = 0.41, p < 0.01).
DISCUSSION
our data demonstrate a relationship between extracranial but
not intracranial iCa stenosis and Csa. severity and asymmetric
distribution of the eiCa stenosis as well as a decrease in para-
sympathetic activity (revealed by a decrease of cardiac hf band
power) were independent predictors for Csa development. We
found a threshold of approximately 70% eiCa stenosis at least
at one side for an increased occurrence of Csa.
Csa was present in 39% of our patients with eiCa stenosis
and was comparable to the prevalence of Csa/ Cheyne-stokes
respiration in heart failure.29 osa was present in 33% of our pa-
tients, which is in line with epidemiological studies adapted for
age and gender similar to our population.30 osa and Csa can
coexist in the same individual, particularly in diseases that pre-
dispose to Csa.31 osa potentially aggravates Csa by elevat-
ing pulmonary capillary wedge pressure and reducing cardiac
output.32,33 however, osa (oahi) did not contribute to Csa
development in our study. We also did not find an association
between osa and the severity of eiCa stenosis, although osa
is a predisposing risk factor for carotid arteriosclerosis.34 this
may be due to the fact that stenosis is a sign of late stage arte-
riosclerosis, while osa is associated with the early stages of
carotid arteriosclerosis.34
the carotid bulb contains peripheral chemoreceptors (ca-
rotid bodies) and arterial baroreceptors. arteries of the carotid
bodies arise directly from carotid arteries.35 therefore, eiCa
arteriosclerosis may lead to ischemia of the carotid bodies,
which sensitizes chemoreceptors.36 The resulting chemoreflex
hypersensitivity enhances the probability for Csa by elevat-
ing the ventilatory response (figure 3).15,37 indeed, a progres-
sive decline in carotid blood flow by bilateral chronic carotid
occlusion triggers peripheral chemoreflex hypersensitivity in
rabbits.38 also, stimulation of carotid chemoreceptors by ade-
nosine, which enhances peripheral chemoresponsiveness, leads
to Csa in human beings.39
Table 4—Resulting multivariate regression models of autonomic and eICA stenosis
parameters
Factors
Considered
Factors in the
Resulting Model B (95% CI)β P-Value
Autonomic
Parameters
eICA Stenosis
Parameters
eICA Stenosis
+
Autonomic
Parameters
VLF, LF, HF VLF 5.79 (1.86; 9.72) 0.43 0.005
eICA score,
eICA ratio
eICA score
eICA ratio
1.58 (0.68; 2.48)
3.04 (1.26; 4.81)
0.40
0.41
0.001
0.001
eICA score,
eICA ratio
VLF, LF, HF
eICA score
eICA ratio
HF
1.48 (0.59; 2.37)
3.54 (1.81; 5.23)
3.29 (0.74; 5.80)
0.39
0.48
0.29
0.002
< 0.001
0.013
Multivariate validation of the dependence of cAHI on autonomic and stenosis
parameters by linear regression models. Results were presented by regression
coefficient (B), 95% confidence intervals (CI), standardized regression coefficient
(β), and P-value.
Table 5—HRV parameters in patients with eICA stenosis with
and without CSA
CSA
n = 17
68.0 ± 9.8
Non-CSA
n = 24
62.3 ± 10.5
P-Value
Patients
Age
Heart Rate
Variability
lnTP (ms2)
lnVLF (ms2)
VLF (nu)
lnLF (ms2)
LF (nu)
lnHF (ms2)
HF (nu)
LF/HF Ratio
n.s
6.7 ± 0.9
6.0 ± 0.9
52.3 ± 14.1
5.5 ± 1.2
68.6 ± 18.9
4.4 ± 1.3
28.7 ± 18.7
4.8 ± 5.0
6.0 ± 0.8
5.0 ± 0.7
40.8 ± 11.6
4.8 ± 1.1
57.2 ± 19.8
4.4 ± 1.0
42.4 ± 21.9
2.1 ± 1.8
0.04
0.002
0.01
n.s
(0.07)
n.s
0.03
0.02

Page 6

SLEEP, Vol. 33, No. 3, 2010
332
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
with carotid artery stenosis are necessary and
would help to identify patients at high risk for
cardiovascular and/or cerebrovascular events.
ABBREVIATIONS
Csa, Central sleep apnea
iCa, internal carotid artery
eiCa, extracranial iCa
iiCa, intracranial iCa
hrV, heart rate variability
bmi, body mass index
eiCa score, eiCa stenosis score
eiCa ratio, ratio between the higher and lower
eiCa stenosis score of both sides
iiCa score, iiCa stenosis score
lVef, left ventricular ejection fraction
feV1, forced expiratory volume in one second
ahi, apnea-hypopnea index
cahi, central apnea-hypopnea index
oahi, obstructive apnea-hypopnea index
osa, obstructive sleep apnea
hf, high frequency band power
lf, low frequency band power
Vlf, very low frequency band power
tp, total band power
nu, normalized units
ACKNOWLEDgMENTS
We thank i. Gaertner and t. schulze for per-
forming the polysomnograms and j. Geiling
for the anatomical drawings.
DISCLOSURE STATEMENT
this was not an industry supported study. the authors have
indicated no financial conflicts of interest.
REFERENCES
1. timsit sG, sacco rl, mohr jp, et al. early clinical differentiation of cere-
bral infarction from severe atherosclerotic stenosis and cardioembolism.
stroke 1992;23:486-91.
2. fisher Cm, Gore i, okabe n, White pd. atherosclerosis of carotid and
vertebral arteries-extracranial and intracranial. j neuropathol exp neurol
1965;24:455-76.
3. javid h. development of carotid plaque. am j surg 1979;138:224-7.
4. dahan a, nieuwenhuijs d, teppema l. plasticity of central chemorecep-
tors: effect of bilateral carotid body resection on central Co2 sensitivity.
plos med 2007;4:1195-204.
5. Kara T, Narkiewicz K, Somers VK. Chemoreflexes-physiology and clini-
cal implications. acta physiol scand 2003;177:377-84.
6. Lanfranchi PA, Somers VK. Arterial baroreflex function and cardiovascu-
lar variability: interactions and implications. am j physiol regul integr
Comp physiol 2002;283:r815-r26.
7. derdeyn Cp, Grubb rl, powers Wj. Cerebral hemodynamic impairment
- methods of measurement and association with stroke risk. neurology
1999;53:251-9.
8. markus hs, mackinnon a. asymptomatic embolization detected by dop-
pler ultrasound predicts stroke risk in symptomatic carotid artery stenosis.
stroke 2005;36:971-5.
9. doux jd, yun aj. the link between carotid artery disease and ischemic
stroke may be partially attributable to autonomic dysfunction and failure
of cerebrovascular autoregulation triggered by darwinian maladapta-
tion of the carotid baroreceptors and chemoreceptors. med hypotheses
2005;66:176-81.
arteriosclerotic changes between the left and right iCa implies
that a mismatch between the two afferent sensory inputs arising
from the peripheral chemoreceptors has an additional potentiat-
ing effect on chemoreflex sensitization, possibly mediated by
central interactions.
the high prevalence of Csa in patients with eiCa stenosis
compared with the absence of Csa in patients with iiCa steno-
sis allows us to conclude that there is an association between
eiCa stenoses and Csa, in spite of the limited sample size.
however, the cross-sectional study design does not allow us to
draw conclusions about the causal relationship between eiCa
stenosis and Csa.
the annual risk of stroke is approximately 2% in patients
with asymptomatic carotid artery stenosis. risk of non-stroke
cardiovascular events is 2- or 3-fold higher in these patients.50
our data suggest that cardiovascular risk arises from a shift in
sympathovagal balance rather than generalized arteriosclerosis.
such a shift in sympathovagal balance with an increase of sym-
pathetic activity and loss of parasympathetic modulation, as well
as an augmented chemoresponsiveness, are well-established
predictors for mortality in heart failure and myocardial infarc-
tion.24,51 today, the stroke-preventing effect of interventional
therapy is only marginal, possibly because current diagnostic
procedures merely focus on cerebrovascular risk induced by
plaque morphology, stenosis-related embolization, and cerebral
hypoperfusion.50 studies examining the prognostic value of Csa
as a clinical expression of autonomic dysfunction in patients
Figure 3—Scheme of the proposed chemoreflex and baroreflex dysfunction in eICA stenosis.
The eICA stenosis induces carotid body ischemia that leads to increased peripheral chemoreflex
sensitivity which induces sympathetic overactivity (chemoreflex–broken lines). In parallel, stenosis
mediated reduced stretch sensitivity of the carotid arterial baroreceptors decreases baroreflex
sensitivity (baroreflex–continuous line) resulting in a loss of parasympathetic tone (baroreflex–
continuous lines). The decreased baroreflex sensitivity also reduces the tonic baroreflex mediated
inhibition of the sympathetic tone and the chemoreflex in the solitary tract nucleus. Consequently,
the decreased baroreflex potentiates chemoreflex sensitivity at the brainstem level in addition to
the increased stenosis mediated peripheral chemoreflex sensitivity. The increased peripheral and
central chemoreflex sensitivity induces respiratory instability (CSA).

Page 7

SLEEP, Vol. 33, No. 3, 2010
333
Carotid Artery Stenosis and Central Sleep Apnea—Rupprecht et al
31. javaheri s, parker tj, liming jd, et al. sleep apnea in 81 ambulatory
male patients with stable heart failure - types and their prevalences, con-
sequences, and presentations. Circulation 1998;97:2154-9.
32. buda aj, schroeder js, Guilleminault C. abnormalities of pulmonary-ar-
tery wedge pressures in sleep-induced apnea. int j Cardiol 1981;1:67-74.
33. Guilleminault C, motta j, mihm f, melvin k. obstructive sleep apnea and
cardiac index. Chest 1986;89:331-4.
34. schulz r, seeger W, fegbeutel C, et al. Changes in extracranial arteries in
obstructive sleep apnoea. eur respir j 2005;25:69-74.
35. jago r, heath d, smith p. structure of the glomic arteries. j pathol
1982;138:205-18.
36. peng yj, overholt jl, kline d, kumar Gk, prabhakar nr. induction
of sensory long-term facilitation in the carotid body by intermittent
hypoxia: implications for recurrent apneas. proc natl acad sci u s a
2003;100:10073-8.
37. Xie al, rutherford r, rankin f, Wong b, bradley td. hypocapnia and
increased ventilatory responsiveness in patients with idiopathic central
sleep apnea. am j respir Crit Care med 1995;152:1950-5.
38. Ding YF, Li YL, Cornish KG, Schultz HD. Reduced blood flow in carotid
arteries is a trigger contributing to peripheral chemoreflex hypersensitiv-
ity in chronic heart failure rabbits. faseb j 2007;21:a1268-a9.
39. Gleeson k, Zwillich CW. adenosine infusion and periodic breathing dur-
ing sleep. j appl physiol 1992;72:1004-9.
40. ponikowski p, Chua tp, piepoli m, et al. augmented peripheral chemo-
sensitivity as a potential input to baroreflex impairment and autonomic
imbalance in chronic heart failure. Circulation 1997;96:2586-94.
41. mortara a, sleight p, pinna Gd, et al. abnormal awake respiratory pat-
terns are common in chronic heart failure and may prevent evaluation
of autonomic tone by measures of heart rate variability. Circulation
1997;96:246-52.
42. szollosi i, krum h, kaye d, naughton mt. sleep apnea in heart fail-
ure increases heart rate variability and sympathetic dominance. sleep
2007;30:1509-14.
43. heistad dd, abboud fm, mark al, schmid pG. effect of barorecep-
tor activity on ventilatory response to chemoreceptor stimulation. j appl
physiol 1975;39:411-16.
44. somers Vk, mark al, abboud fm. interaction of baroreceptor and
chemoreceptor reflex control of sympathetic-nerve activity in normal hu-
mans. j Clin invest 1991;87:1953-7.
45. miura m, reis dj. role of solitary and paramedian reticular nuclei in
mediating cardiovascular reflex responses from carotid baroreceptors and
chemoreceptors. j physiol-london 1972;223:525-48.
46. naughton mt, benard dC, liu pp, rutherford r, rankin f, bradley td.
effects of nasal Cpap on sympathetic activity in patients with heart-fail-
ure and central sleep apnea. am j respir Crit Care med1995;152:473-9.
47. Mansfield D, Kaye DM, La Rocca HB, Solin P, Esler MD, Naughton MT.
raised sympathetic nerve activity in heart failure and central sleep apnea
is due to heart failure severity. Circulation 2003;107:1396-400.
48. mortara a, sleight p, pinna Gd, et al. association between hemodynamic
impairment and Cheyne-stokes respiration and periodic breathing in
chronic stable congestive heart failure secondary to ischemic or idiopathic
dilated cardiomyopathy. am j Cardiol 1999;84:900-4.
49. solin p, kaye dm, little pj, bergin p, richardson m, naughton mt.
impact of sleep apnea on sympathetic nervous system activity in heart
failure. Chest 2003;123:1119-26.
50. redgrave jn, rothwell pm. asymptomatic carotid stenosis: what to do.
Curr opin neurol 2007;20:58-64.
51. ponikowski p, Chua tp, anker sd, et al. peripheral chemoreceptor hyper-
sensitivity-an ominous sign in patients with chronic heart failure. Circula-
tion 2001;104:544-9.
10. Nasr N, Pavy-Le Traon A, Larrue V. Baroreflex sensitivity is impaired in
bilateral carotid atherosclerosis. stroke 2005;36:1891-5.
11. robinson tG, dawson sl, eames pj, panerai rb, potter jf. Cardiac
baroreceptor sensitivity predicts long-term outcome after acute ischemic
stroke. stroke 2003;34:705-11.
12. La Rovere MT, Bigger JT, Marcus FI, et al. Baroreflex sensitivity and
heart-rate variability in prediction of total cardiac mortality after myocar-
dial infarction. lancet 1998;351:478-84.
13. Bigger JT, Larovere MT, Steinman RC, et al. Comparison of baroreflex
sensitivity and heart period variability after myocardial infarction. j am
Coll Cardiol 1989;14:1511-8.
14. ponikowski p, Chua tp, piepoli m, et al. Ventilatory response to exer-
cise correlates with impaired heart rate variability in patients with chronic
congestive heart failure. am j Cardiol 1998;82:338-44.
15. solin p, roebuck t, johns dp, Walters eh, naughton mt. peripheral and
central ventilatory responses in central sleep apnea with and without con-
gestive heart failure. am j respir Crit Care med 2000;162:2194-200.
16. strassburg a, majunke b, notges jk, et al. Central sleep apnea is as-
sociated with blunted baroreflex sensitivity in patients with myocardial
infarction. int j Cardiol 2007.
17. lanfranchi pa, somers Vk, braghiroli a, Corra u, eleuteri e, Giannuzzi
p. Central sleep apnea in left ventricular dysfunction - prevalence and
implications for arrhythmic risk. Circulation 2003;107:727-32.
18. Guyton aC, Crowell jW, moore jW. basic oscillating mechanism of
cheyne-stokes breathing. am j physiol 1956;187:395-8.
19. pepin jl, Chouri-pontarollo n, tamisier r, levy p. Cheyne-stokes res-
piration with central sleep apnoea in chronic heart failure: proposals for a
diagnostic and therapeutic strategy. sleep med rev 2006;10:33-47.
20. mrC european Carotid surgery trial: interim results for symptoma-mrC european Carotid surgery trial: interim results for symptoma-
tic patients with severe (70-99%) or with mild (0-29%) carotid steno-
sis. european Carotid surgery trialists’ Collaborative Group. lancet
1991;337:1235-43.
21. rechtschaffen a, kales a. a manual of standardized terminology: tech-
niques and scoring systems for sleep stages of human subjects. los ange-
les, uCla, brain information service/brain research institute, 1968.
22. asda. eeG arousals: scoring rules and examples: a preliminary report
from the sleep disorders atlas task force of the american sleep disor-
ders association. sleep 1992;15:173-84.
23. kushida Ca, littner mr, morgenthaler t, et al. practice parameters for
the indications for polysomnography and related procedures: an update
for 2005. sleep 2005;28:499-521.
24. Camm aj, malik m, bigger jt, et al. heart rate variability - standards of
measurement, physiological interpretation, and clinical use. Circulation
1996;93:1043-65.
25. malliani a, pagani m, lombardi f, Cerutti s. Cardiovascular neural regu-
lation explored in the frequency domain. Circulation 1991;84:482-92.
26. Ponikowski P, Chua TP, Amadi AA, et al. Detection and significance of a
discrete very low frequency rhythm in rr interval variability in chronic
congestive heart failure. am j Cardiol 1996;77:1320-6.
27. ponikowski p, anker sd, Chua tp, et al. oscillatory breathing patterns
during wakefulness in patients with chronic heart failure-Clinical impli-
cations and role of augmented peripheral chemosensitivity. Circulation
1999;100:2418-24.
28. Chao aC, Chern Cm, kuo tb, et al. noninvasive assessment of spon-
taneous baroreflex sensitivity and heart rate variability in patients with
carotid stenosis. Cerebrovasc dis 2003;16:151-7.
29. javaheri s. sleep disorders in systolic heart failure: a prospective study of
100 male patients. The final report. Int J Cardiol 2006;106:21-8.
30. young t, peppard pe, Gottlieb dj. epidemiology of obstructive sleep
apnea- a population health perspective. am j respir Crit Care med
2002;165:1217-39.