[Relationship between age and cardiovascular complications in obstructive sleep apnoea].

Abstract

Obstructive sleep apnoea (OSA) is a risk factor for cardiovascular morbidity and mortality. The aim of this study was to assess relations between cardiovascular diseases (CVD) and age in OSA subjects.
Consecutive OSA subjects (AHI/RDI > 10, Epworth score > 9 points) were evaluated. The chest X-ray, spirometry, arterial blood gases, ECG, blood morphology and biochemistry were performed during trial treatment with autoCPAP.
We studied 533 consecutive OSA patients, mean age 55.6 +/- 10.3 years (range 24-81), with obesity (BMI 34.4 +/- 6.6 kg/m(2)) and severe OSA (AHI/RDI 37.8 +/- 21.8). To evaluate relations between CVD and age, patients were divided into three groups. Group 1 < 50 years (123 subjects, 23.1%), Group 2 aged 50-60 years (250 subjects, 46.9%) and Group 3 > 60 years (160 subjects, 30%). Subjects < 50 years were more obese and had higher AHI/RDI when compared to older groups. Incidence of arterial hypertension, coronary artery disease, atrial fibrillation, heart failure and stroke increased with age (higher in subjects > 60 years).
Cardiovascular diseases were prevalent in OSA patients > 60 years. However the youngest group presented with more severe obesity and higher AHI/RDI.

Full text

Page 1

www.pneumonologia.viamedica.pl
ORIGINAL ARTICLE
235
Address for correspondence:Address for correspondence:
Address for correspondence:Address for correspondence:
Address for correspondence: Robert Pływaczewski, Department of Respiratory Failure, National Research Institute of Tuberculosis and Lung Diseases,
Płocka 26, 01–138 Warsaw, Poland
Received: 11.08.2008
Copyright © 2009 Via Medica
ISSN 0867–7077
Robert Pływaczewski1, Monika Czystowska2, Agnieszka Skoczylas2, Przemysław Bieleń1,
Luiza Jonczak1, Dorota Górecka2, Paweł Śliwiński1
1Department of Respiratory Failure, National Research Institute of Tuberculosis and Lung Diseases, Warsaw, Poland
Head: Prof. P. Śliwiński
2II Department of Respiratory Medicine, National Research Institute of Tuberculosis and Lung Diseases, Warsaw, Poland
Head: Prof. D. Górecka
Relationship between age and cardiovascular complications
in obstructive sleep apnoea
Abstract
Introduction: Obstructive sleep apnoea (OSA) is a risk factor for cardiovascular morbidity and mortality. The aim of this
study was to assess relations between cardiovascular diseases (CVD) and age in OSA subjects.
Material and methods: Consecutive OSA subjects (AHI/RDI > 10, Epworth score > 9 points) were evaluated. The chest X-ray,
spirometry, arterial blood gases, ECG, blood morphology and biochemistry were performed during trial treatment with autoCPAP.
Results: We studied 533 consecutive OSA patients, mean age 55.6 ± 10.3 years (range 24–81), with obesity (BMI 34.4 ±
± 6.6 kg/m2) and severe OSA (AHI/RDI 37.8 ± 21.8). To evaluate relations between CVD and age, patients were divided into
three groups. Group 1 < 50 years (123 subjects, 23.1%), Group 2 aged 50–60 years (250 subjects, 46.9%) and Group 3
> 60 years (160 subjects, 30%). Subjects < 50 years were more obese and had higher AHI/RDI when compared to older
groups. Incidence of arterial hypertension, coronary artery disease, atrial fibrillation, heart failure and stroke increased with
age (higher in subjects > 60 years).
Conclusions: Cardiovascular diseases were prevalent in OSA patients > 60 years. However the youngest group presented
with more severe obesity and higher AHI/RDI.
Key words: obstructive sleep apnoea, cardiovascular complications, age
Pneumonol. Alergol. Pol. 2009; 77: 235–241
Introduction
Previous epidemiological studies confirmed
that respiratory disturbances during sleep (RDDS)
are more frequent among elderly people (from 30%
to 80%) than among middle age subjects (4% of
men and 2% of women). In middle age RDDS oc-
curs about twice as often among men as among
women. In older age groups (men > 60–65 years
and women after menopause without hormonal
substitutive therapy — HTZ) the frequency of
RDDS occurrence is similar in both sexes [2–4].
Ancoli-Israel et al. [5] assessed RDI occurren-
ce among persons over 65 years in a group of
427 people. Respiratory disturbance index (RDI)
> 5 was confirmed among 81% of the study’s par-
ticipants; 62% of the group (70% of men and 56%
of women) had RDI > 10. RDDS risk factors: high
index of body mass (BMI) and sex (RDDS was more
often diagnosed among men).
Elsewhere, the same authors studied relations be-
tween the RDDS occurrence and age among 233 occu-
pants of a foster home (151 men and 82 women). The
average age was 83.5 ± 8.5 years for women and
79.7 ± 7.5 years for men. RDI > 5 was diagnosed among
70% of subjects, (68% of women and 76% of men).
Bixler et al. [4, 7] assessed the influence of age
on the frequency of RDDS occurrence in the po-

Page 2

Pneumonologia i Alergologia Polska 2009, vol 77, no 3, pages 235–241
236
www.pneumonologia.viamedica.pl
pulation of men and women (1000 women and
741 men over 20 years were chosen to PSG study).
The RDDS occurred more than twice as often in
the oldest group of men and women (65–100 years)
compared to middle aged patients (46–64 years).
Among men, AHI ≥ 10 was diagnosed among
23.9% of subjects aged between 65–100 and 11.8%
of subjects aged 45–64. Among women AHI ≥ 15 was
confirmed in 7% of subjects aged 65–100 and in
2% of subjects aged 45–64. On the basis of poli-
somnography and daily symptoms, OSA was dia-
gnosed among 3.9% of men and 1.2% of women.
The OSA frequency was low among women befo-
re menopause (0.6%) and among these who were
taking HTZ (0.5%). Among women who were un-
dergoing their menopause without HTZ, the RDDS
frequency was significantly higher and amounted
to 2.7% (p = 0.02).
Bigger differentiation in the RDDS frequency
in middle and old age patients were diagnosed by
Spanish authors [8, 9]. In these works, two popu-
lations were studied: the first aged between 30–70
and the second aged 71–100. AHI ≥ 5 was diagno-
sed/found among 81% of men and 80% of women
aged 30–70. AHI ≥ 15 was diagnosed among 57%
of men and 49% of women in the older group and
among 14.2% of men and 7% of middle age women.
In Cleveland Family Study [10] the RDDS oc-
currence was compared among men and women in
the following age groups: 25–60 years and > 60 years.
AHI ≥ 15 was diagnosed among 4% of women in the
first group and 32% of those in the second group and
among 22% and 42% of men respectively.
In the Sleep Heart Health Study (SHHS) the
percentage of persons with AHI ≥ 15 in the age gro-
up between 60–99 years was 1.7 times bigger than
among subjects aged 40–60 [11].
The aim of this study is to assess the relations
between age and the degree of OSA progression and
between obesity and cardiovascular diseases.
Material and methods
During the first consultation, subjects filled in
a questionnaire about respiratory disturbances during
sleep (the questionnaire referred to such symptoms
as snoring, sleep apnoeas, awakening, daily sleepi-
ness [the Epworth Sleepiness Scale or ESS], morning
fatigue, nycturia and co-existing diseases, applied
drugs, laryngological operations experienced, and
smoking). Polisomnography (PSG) was made using
Somnostar a (Sensormedics, USA) and Alice 5 (Re-
spironics, USA) equipment. Some people were sub-
jected to polygraphical examination (without the
estimation of sleep phases) with the use of appara-
tus PolyMesam (MAP, Germany). A description of the
above studies is presented in another work [12].
AHI/RDI which was associated with excessi-
ve daily sleepiness (SSE > 9) was assumed as
a criterion for diagnosing OSA [7].
Some 533 people suffering from OSA (412 men,
77.3% and 121 women, 22.7%) aged between
24 and 81 (the average age was 55.6 plus or minus
10.3) were examined. The subjects suffered from
moderate and severe forms of the disease (the ave-
rage AHI/RDI — 37.8 ± 21.8).
Chest spirometry, ECG, arterial blood gasome-
try and basic blood examinations were made
among patients during the hospitalization connec-
ted with OSA using autoCPAP apparatus.
Co-existing diseases were diagnosed according
to criteria described below.
Arterial hypertension was diagnosed in cases
of positive history (previously prescribed hypoten-
sive drugs) and/or increased blood pressure during
repetitive measurements. Ischaemic heart disease
was diagnosed on the basis of the history (corona-
ry ailments, applied treatment) and/or diagnosed
changes in ECG (post-infarction scar. Left bundle
branch block — after excluding other reasons
— trails of ischemia). Decisive factors in the diagno-
sis of cardiac insufficiency were: historic data (re-
sults of previous examinations from cardiologic/
/internistic units); effort dyspnoea — II–IV class
according to NYHA; objective examination — pe-
ripheral oedemas; traits of pulmonary haemosta-
sis completed with results of actual echocardiogra-
phy (ejection fraction < 50% or end-diastolic func-
tion) and roentgenoscopy of chest (extended heart
silhouette, Karley’s lines). Endured stroke was dia-
gnosed on the basis of previous examination results
and medical documentation from internistic and
neurological units.
The diagnosis of chronic obstructive pulmo-
nary disease (COPD) was confirmed after making
spirometry (FEV1%FVC < low range of norm and
negative end-diastolic test) which was most often
completed with typical history (chronic cough,
effort dyspnoea, exposure to cigarette smoke).
Diabetes was diagnosed when the following
factors occurred: glycemia on empty stomach twi-
ce > 125 mg% or ‘incidental’ glycemia > 200 mg%
or glycemia > 200 mg% in the 75 g glucose load
test (diabetes was diagnosed and was healing be-
fore confirmation of OSA among most of the pe-
ople suffering from diabetes). Hypertriglyceride-
mia, hypercholesterolemia and mixed hyperlipide-
mia were diagnosed if: triglycerides concentration
in fasting state exceeded 160 mg%, total choleste-
rol concentration in fasting state was > 200 mg%

Page 3

Robert Pływaczewski et al., Relationship between age and cardiovascular complications in obstructive sleep apnoea
237
www.pneumonologia.viamedica.pl
and when both criteria occurred at the same time.
Hyperuricemia was diagnosed when uric acid con-
centration in blood serum in fasting state exceeded
5.7 mg% in women and 7 mg% in men.
Statistical analysis
The results obtained were analyzed with the
help of statistical program Statistica 6.0. The results
of research were presented as mean and standard
deviations. The qualitative differences between stu-
died variables were estimated with Pearson’s c2 test
or in Yates’s and Fisher’s modification for groups
of fewer people. The quantitative differences betwe-
en studied variables in selected sub-groups of pa-
tients were estimated with the ANOVA test. The
multiple regression test was used to select variables,
which significantly influenced AHI/RDI value.
Results
The average index of apneas and shallow-bre-
athing AHI/RDI amounted in the whole group to
37.8 ± 21.8 episodes for an hour of sleeping du-
ring the PSG examination and registration time du-
ring the PolyMesam examination. The average arte-
rial blood saturation with oxygen (SaO2 mid.) was
90.3 ± 5.1%. The lowest arterial blood saturation
with oxygen (SaO2 min.) was 74 ± 12%. The sub-
jects were spending almost one third of the night in
oxygen deficiency. Most subjects were overweight or
obese and had an average BMI of 34.4 ± 6.6 kg/m2.
The score on the Epworth scale was on average
11.9 ± 5.7 (the norm is 9; the range of scale is 0–24).
Diseases of the cardiovascular system were
often diagnosed in this subject group. Arterial hy-
pertension was diagnosed in 381 patients and 126
suffered from ischemic heart disease. Heart failu-
re was diagnosed in 63 persons (11.8%). Auricu-
lar fibrillation (periodic or fixed) was diagnosed in
45 patients (8.4%). Nineteen subjects (3.6%) had
suffered a stroke.
COPD was diagnosed in 91 subjects (17.1%).
Among metabolic disorders hypercholesterolemia
was most often diagnosed: 251 subjects (47.2%);
230 subjects (43.2%) had hypertriglyceridemic;
mixed hyperlipidemia was diagnosed in 149 (28%)
and hyperuricemia was diagnosed in 202 (38.5%).
Diabetes was confirmed in 87 patients (16.3%).
To assess the relationship between age and the
degree of OSA progression, obesity and cardiova-
scular disease, the subjects were divided into three
sub-groups according to age. The first group con-
sisted of 123 patients (23.1%) aged < 50 years, the
second of people aged 50–60 (250 people, 46.9%),
and the third aged > 60 years (160 people, 30%).
The progression of OSA in the age groups
The highest AHI/RDI values and the highest
score on the Epworth scale were found in the youn-
gest age group. In all groups, the average BMI exce-
eded 30 kg/m2 (the lowest was in the third group).
Minimal and average SaO2 were lowest, and time of
night oxygen deficiency/anoxia (T90) the shortest,
in the oldest group. The results of PSG/Poly-Mesam
examination, BMI and Epworth scale of sleepiness
in the age groups are presented in Table 1.
Co-existing diseases
Arterial hypertension, ischemic heart disease
and auricular fibrillation were most often diagno-
sed in the third group and most rarely in the first
group. Significant differences were found betwe-
en all groups. Similar results were observed for
heart failure (significant differences only between
groups 1 and 3) and stroke (in the first group no-
body had a stroke and significantly important dif-
ferences were between groups 1 and 2 and groups
1 and 3). Detailed results are presented in Table 2.
Table 1. Comparison of PSG/PolyMesam, BMI, Epworth score in age groups 1, 2 and 3
Variable Group 1Group 2Group 3p
BMI [kg/m2]35.5±7*35.1±6.6#
32.3±5.7*#
*p = 0.0003
#p = 0.0004
AHI/RDI (n/h)44±24.8*38.8±21.2#
31.4±18.4*#
*p < 0.0001
#p = 0.005
Mean SaO2 (%)
Lowest SaO2 (%)
T90 (%)
89.9±5.689.9±5.2*91.3±4.2**p = 0.045
74.4±11.573±1375.4±10.6NS
35.5±31.5*33±31.225.8±27.3*p = 0.03
Epworth Sleepiness Score (points)13.4±5.6*12.6±5.6#
9.7±5.2*#
*p < 0.0001
#p < 0.0001
Explanations of abbreviations in the text

Page 4

Pneumonologia i Alergologia Polska 2009, vol 77, no 3, pages 235–241
238
www.pneumonologia.viamedica.pl
Diabetes was diagnosed significantly more
often in groups 2 and 3 compared to group 1. COPD
was most rarely diagnosed in the first group (the dif-
ferences between groups were not statistically signi-
ficant). Hypercholesterolemia, hypertriglyceridemia
and mixed hyperlipidemia were significantly more
often diagnosed in groups 1 and 2 than in group 3.
Hyperuricemia was diagnosed most often in group
1, but the differences were not statistically signifi-
cant. Detailed results are presented in Table 3.
Lung function tests
The results of FVC and FEV1 (expressed in
percentage of predicted value) were similar in all
groups. FEV1%FVC was significantly higher in gro-
up 1. PaO2 measured during the day (while bre-
athing with atmospheric air) was also significan-
tly higher in group 1. The results of spirometry and
gasometry are presented in Table 4.
The analysis of multiple regression (the follo-
wing factors were taken into consideration in the
model: BMI, sex, age, SSE, neck circumference,
COPD, ischaemic heart disease, arterial hypertension,
heart failure, auricular fibrillation, diabetes) showed
significant correlations (R = 0.49, R2 = 0.24) betwe-
en AHI/RDI and BMI (b = 0.37, p < 0.0001), Epworth
scale (b = 0.12, p = 0.01) and age (b = 0.12, p = 0.02).
Discussion
Among the 533 suffering from OSA, the most
severe forms of the disease were diagnosed among
patients aged under 50 years and cardiovascular
complications were most often found among sub-
jects aged over 60 years.
Previous results of epidemiologic studies con-
firm more frequent occurrence of RDDS among
older age groups [1, 2, 4, 5, 7–10]. The results of
studies pertaining to RDDS’s influence on the oc-
currence of cardiovascular complications and de-
ath rate in elderly age are unequivocal.
Bliwise et al. [13] observed 198 subjects over
12 years (average age 66 ± 8.2 years). Among sub-
jects with RDDS confirmed (RDI > 10) the risk of
Table 3. Other concomitant diseases in groups 1, 2 and 3
Variable Group 1Group 2 Group 3P
COPD (n/%)
Diabetes (n/%)
14 (11.4%)
11 (8.9%)*#
47 (18.8%)
47 (18.8%)*
30 (18.8%)
29 (18.1%)#
NS
*p = 0.01
#p = 0.03
*p = 0.002
#p = 0.007
*p = 0.04
#p = 0.01
*p = 0.004
#p = 0.003
NS
Hypertriglicerydaemia (n/%) 62 (50.4%)*116 (46.4%)#
52 (32.7%)*#
Hypercholesterolaemia (n/%)62 (50.4%)*128 (51.2%)#
61 (38.4%)*#
Mixed hyperlipidaemia (n/%) 40 (32.5%)*80 (32%)#
29 (18.2%)*#
Hyperuricaemia (n/%)52 (42.6%)93 (37.8%) 57 (36.5%)
Table 2. Cardiovascular diseases in groups 1, 2 and 3
VariableGroup 1Group 2Group 3p
Arterial hypertension (n/%)70 (56.9%)*#
182 (72.9%)*129 (80.6%)#
*p = 0.002
#p = 0.00002
*p = 0.0003
#p < 0.0001
+p < 0.0001
*p = 0.008
*p = 0.01
#p = 0.0001
+-p = 0.04
*p = 0.04
#p = 0.003
Coronary artery disease (n/%)8 (6.5%)*#
53 (21.2%)*+
65 (40.6%)#+
Heart failure (n/%)
Atrial fibrillation (n/%)
8 (6.5%)*
2 (1.6%)*#
28 (11.2%)
20 (8%)*+
27 (16.9%)*
23 (14.4%)#+
Stroke (n/%) 0 (0%)*#
8 (3.2%)*11 (6.9%)#
Explanations of abbreviations in the text

Page 5

Robert Pływaczewski et al., Relationship between age and cardiovascular complications in obstructive sleep apnoea
239
www.pneumonologia.viamedica.pl
death was 2.7 times higher in comparisson with
subjects without respiratory disturbances during
sleep. Cardiovascular diseases were most often the
death cause in this subject group.
Ancoli-Israel et al. [14] analyzed death causes
among 426 elderly persons (average age while be-
ginning the study 72.5 years). Subjects with RDI
> 30 lived significantly shorter than persons with RDI
£ 30 (p = 0.0034). The independent factors of de-
ath risk were age, diseases of the circulatory system/
/cardiovascular system and of the respiratory system.
He et al. [15] evaluated the survival time
among 385 sufferers from OSA over eight years.
Those with advanced forms of the disease (AI
> 20 than AI < 20; p < 0.05), who were not treated
for OSA, died significantly more often. The above
phenomenon was especially found among those
aged under 50 (significant differences in the length
of survival were observed between the second and
eighth year of observation). In the group of patients
aged over 50, a significant influence of AI > 20 on
survival shortening was found only in the eighth
year of observation.
Lavie et al. [16] studied the RDDS influence
on the general death rate in a large group of 14 589
men aged between 20 and 93 years. The average
length of observation was 4.6 years. The subjects
with RDI > 30 had a significantly higher risk of
death in comparison to persons with RDI £ 10. The
biggest death rate was observed among obese pa-
tients (BMI ≥ 31) with RDI > 40 (11.47/1000 per-
sons/year). In the group of people with RDI > 30,
after considering age, a significantly higher risk of
death was found in the group aged 20–29 years
— hazard ratio (HR) 5.8 (95% CI, 1.45–23.42) and
among those ≥ 80 years — HR 1.92 (95% CI, 1.19–
–3.09) in comparison to control groups selected ac-
cording to age. In the group of patients with RDI
> 50 the increased risk of death in comparison to the
general population was stated for the following age
groups: 20–29, 30–39, 40–49 years (HR was appro-
priately 9.8% [95% CI, 2.44–39.25], 3.12 [95% CI,
1.17–8.35] and 1.89 [95% CI, 1.04–3.44]).
According to some authors, OSA is an inde-
pendent (also from age) risk factor of arterial hy-
pertension, ischaemic heart disease and stroke [17].
However in other studies, the RDDS influence on
the development of cardiovascular diseases was
not confirmed.
Philips et al. [18] observed a group of 82 heal-
thy persons aged between 50 and 80 over five years.
After the first polisomnography, AHI > 5 was dia-
gnosed in 16 subjects. After five years, control exa-
minations were done among 42 persons (including
ten people with AHI > 5 from the initial examina-
tion). Neither a significant increase in BMI nor blo-
od pressure among people with AHI < 5 and AHI
> 5 were observed.
Enright et al. [19] assessed the relation betwe-
en snoring, apnoea’s occurrence and the functioning
of the circulatory system in 5201 people > 65 years.
Snoring was declared by 33% of men and 19% of
women; apnoeas were observed among 13% of men
and 4% of women. Snoring, observed apnoeas and
daily excessive sleepiness were not in the subject
group connected with arterial hypertension nor
other diseases of the cardiovascular system.
Hass et al. [20] assessed the influence of age
on arterial hypertension occurrence among 6120
subjects participating in the Sleep Heart Health
Study (SHHS). The authors compared two groups
of subjects: the first consisting of 2477 people aged
between 40–59 years, and the second of people
aged 60 and above. In the first group, significant
connections between AHI and arterial hyperten-
sion were diagnosed (for AHI from 15 to 39.9, OR
was 2.38, 95% CI, 1.30–4.38; for AHI ≥ 30 OR was
2.24, 95% CI, 1.10–4.54). In the group of subjects
≥ 60 years the connections between AHI and arte-
rial hypertension were not observed.
Other studies of RDDS, with a sizeable propor-
tion of persons over 60 years of age, confirm more
Table 4. Comparison of spirometry and arterial blood gases in groups 1, 2 and 3
VariableGroup 1 Group 2Group 3P
FVC (%n)
FEV1 (%n)
FEV1%FVC (%)
94.2±16.5
89.6±17.2
77.2±7.4*#
93.4 ±18.1
87.8 ±19.8
74.2 ±8.1*
95.7± 18.1
90.3± 21.1
72.2± 9.1#
NS
NS
*p = 0.01
#p < 0.0001
*p < 0.0001
#p = 0.02
NS
PaO2 [mm Hg]74.1±9.1*#
70 ±8.2*71.2± 8.4#
PaCO2 [mm Hg]39.2±3.939.5 ±4.239.2± 4.6
Explanations of abbreviations in the text