Low Physical Activity Is a Determinant for Elevated Blood Pressure

in High Cardiovascular Risk Obstructive Sleep Apnea

Monique Mendelson PhD, Renaud Tamisier MD PhD, David Laplaud PhD,
Sonia Dias-Domingos MSc, Jean-Philippe Baguet MD PhD, Laurent Moreau MD,
Christian Koltes MD, Léonidas Chavez MD, Gilles de Lamberterie MD, Frédéric Herengt MD,
Patrick Levy MD PhD, Patrice Flore PhD, and Jean-Louis Pépin MD PhD

INTRODUCTION: Obstructive sleep apnea (OSA) is associated with cardiovascular morbidity,

including hypertension. Beyond the severity of nocturnal hypoxia, other factors such as metabolic
abnormalities but also sedentary behaviors and insufficient physical activity may contribute to
elevated blood pressure (BP). To clarify the respective role of these factors as determinants of BP
in OSA patients, we examined the relationship between BP and anthropometrics, severity of sleep
apnea, and objectively measured physical activity and sedentary behaviors. METHODS: Ninety-
five adults presenting with OSA (apnea-hypopnea index > 10 events/h) and high cardiovascular
risk (63.3 ⴞ 8.8 y; body mass index: 29.9 ⴞ 4.9 kg/m2; apnea-hypopnea index: 41.3 ⴞ 17.5/h; car-
diovascular risk score: 13.5 ⴞ 3.7%) were included. Physical activity and sedentary behaviors were
objectively assessed by actigraphy, and self-measured home BP monitoring was measured. Logistic
regression models adjusted for sex, age, and body mass index were built to identify the predictors
of self-measured morning and evening BP. RESULTS: Physical activity was significantly related to
obesity but not to the severity of sleep apnea or sleepiness. Sedentary behaviors were associated with
self-measured morning and evening systolic BP (r ⴝ 0.32, P ⴝ .002; r ⴝ 0.29, P ⴝ .004). Steps per
day were inversely associated with evening BP (r ⴝ ⴚ0.27, P ⴝ .01). Univariate analysis identified
steps/d and time spent in vigorous physical activity as determinants for evening self-measured BP.
In multivariate analysis, only steps/d were identified as a significant determinant of evening BP.
CONCLUSIONS: Physical activity is the major determinant for evening BP in adults with OSA
presenting high cardiovascular risk. Our results emphasize the need for lifestyle counseling pro-
grams in combination with CPAP to encourage regular physical activity in OSA subjects to obtain
better BP control. (ClinicalTrials.gov registration NCT01226641.) Key words: obstructive sleep ap-
nea; cardiovascular risk; blood pressure; physical activity; sedentary behaviors; sleep. [Respir Care
2014;59(8):1218 –1227. © 2014 Daedalus Enterprises]

Introduction hypertension.1,2 The repetitive occurrence of apneas and

hypopneas with the associated intermittent hypoxia se-
Obstructive sleep apnea syndrome (OSA) has been as- quence causes acute surges in blood pressure (BP), sleep
sociated with cardiovascular morbidity and, in particular, fragmentation, and chronic sympathetic activation that ul-

Dr Mendelson, Dr Tamisier, Ms Dias-Domingos, Dr Levy, Dr Flore, and dipôle Gentilly St Jacques, Nancy; Dr Koltes is affiliated with the Service
Dr Pépin are affiliated with Université Grenoble Alps, HP2 Laboratory, de Pneumologie, Saint Avold; Drs Chavez and de Lamberterie are affil-
with Institut National de la Santé et de la Recherche U1042, Grenoble, iated with Cabinet Marcel Benoit, Grenoble; and Dr Herengt is affiliated
and with Grenoble University Hospital, EFCR and Sleep Laboratory, with Dieulefit Santé, Domaine Chamonix, Dieulefit, France.
Grenoble, France. Dr Laplaud is affiliated with Initiatives pour la Santé
Domicile, Lyon; Dr Baguet is affiliated with the Cardiology Clinic, Cen- Supplementary material related to this paper is available at http://
tre Hospitalier Universitaire, Grenoble; Dr Moreau is affiliated with Mé- www.rcjournal.com.

1218 RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

timately lead to sustained hypertension.3 A large body of

evidence from animal models, intermittent hypoxia expo- QUICK LOOK
sure of healthy humans4 and epidemiological studies sup- Current knowledge
port an independent role of OSA in the pathogenesis of
Obstructive sleep apnea (OSA) is associated with car-
daytime hypertension.2,5,6 However, it has been shown that
diovascular morbidity including hypertension. Noctur-
CPAP, the first line therapy for OSA, has a limited effect
nal hypoxia, metabolic abnormalities, sedentary behav-
on blood pressure control,7,8 and the size of the effect is
iors, and insufficient physical activity contribute to
highly related to the duration of nightly CPAP usage.9,10
elevated blood pressure (BP).
This suggests that, in these frequently obese and comorbid
OSA patients, the severity of abnormalities during sleep is What this paper contributes to our knowledge
only one of the numerous factors explaining elevated day-
time blood pressure. In a group of subjects with OSA and cardiovascular
Lack of physical activity is an established modifiable risk, steps/d and sedentary behaviors were the most
risk factor for cardiovascular disease and premature mor- important predictors of self-measured BP. Lifestyle
tality.11,12 Most studies examining physical activity and counseling to improve physical activity may improve
cardiovascular risk are based on subjective questionnaires, BP control.
and the majority of surveys have reported higher rates of
elevated blood pressure in inactive versus active individ-
uals.13,14 One study recognizes that time spent in self-
ever, there are limited data regarding objective measures
reported sedentary behaviors is a unique and independent
of different intensities of physical activity and sedentary
risk factor for cardiovascular disease.15 Previous studies
behaviors and their relation to cardiovascular risk factors
have also reported an association between sedentary be-
and blood pressure in OSA patients.
haviors and increased risk of hypertension.16,17 OSA se-
We hypothesized that daytime sleepiness, fatigue, and
verity has been associated with self-reported lack of exer-
obesity frequently exhibited by OSA patients would lead
cise and potentially impairs exercise performance.18
to lower levels of physical activity and increased sedentary
Sleep duration is another modifiable risk factor that may
behavior, which, in turn, can negatively impact BP. The
impact BP, physical activity, and sedentary behaviors.
aim of the present study was to assess the predictors of
Short-term sleep deprivation studies suggest that acute sleep
elevated BP in OSA patients with high cardiovascular risk.
deprivation increases BP.19,20 Furthermore, a relationship
Accordingly, we examined the relationship between BP
between habitual sleep duration and BP21 and an inverse
and anthropometrics, severity of sleep apnea, and objec-
relation between sleep duration and body mass index
tively measured physical activity and sedentary behaviors.
(BMI)22,23 have been demonstrated. Hypertension in short
sleepers may then also be favored by the development of
obesity.24,25 One study showed that short-term sleep loss Methods
reduces daytime overall spontaneous physical activity and
shifts the intensity of physical activity toward lower levels Study Design and Participants
under free-living conditions.26 Accordingly, alterations in
sleep quantity and quality during OSA and the associated Ninety-five adults with OSA were prospectively in-
daytime sleepiness and fatigue may be reasons for reduced cluded. Inclusion criteria were: men and women 18 – 85 y
physical activity and increased sedentary behaviors.27 How- of age, allowing a representation of OSA adults who are
active in the workforce or retired and who are capable of
demonstrating at least light physical activity; BMI ⬍ 40
kg/m2; OSA diagnosed with polysomnography or polyg-
This study was supported by a grant from Initiative pour la Santé. The raphy; cardiovascular score ⬎ 5%28; or secondary preven-
authors have disclosed no conflicts of interest.
tion with a past history of cardiovascular disease (transient
Dr Pépin has full access to all of the data in the study, and he takes full ischemic attack, stroke, cerebral hemorrhage, myocardial
responsibility for the integrity of all of the data and the accuracy of the infarction, angina, coronary revascularization, arteriopa-
data analysis, including and especially any adverse effect. thy, aortic aneurysm). Subjects were recruited from 13
Drs Flore and Pépin are senior co-authors. centers across France. Mean recruitment per center was 8
subjects. Stratification was carried out to adjust the models
Correspondence: Monique Mendelson PhD. E-mail: monique with regards to the recruiting center. Noninclusion criteria
mendelson1@gmail.com. Jean-Louis Pépin MD PhD. E-mail:
jpepin@chu-grenoble.fr. were: central sleep apnea syndrome, cardiovascular score
⬍ 5% in primary prevention,28 cardiac failure, history of
DOI: 10.4187/respcare.02948 hypercapnic chronic respiratory failure, incapacitated pa-

RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8 1219

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

tients in accordance with article L 1121-6 of the French normal cuff was used for circumferences below 32 cm;
public health code, or subjects taking part in another clin- otherwise, a large cuff was used.
ical trial. The study protocol was approved by the ethical Target BPs for home readings were based on the then-
committee (Comité de Protection des Personnes Sud Est current United Kingdom NICE guidelines for hyperten-
V, 09-PROR-1). All of the subjects provided written in- sion and diabetes,32 adjusted down by 10/5 mm Hg in ac-
formed consent to participate in the study. The different cordance with the recommendations of the British
assessments in this study corresponded to a baseline eval- Hypertension Society (home readings tend to be lower
uation before being included in a prospective randomized, than office readings). Target values were therefore 135/85
controlled trial targeting the efficiency of a telemedicine mm Hg for subjects without diabetes and 125/80 mm Hg
program in addition to standard CPAP treatment (Clinical for subjects with diabetes.32,33
Trials registration NCT01226641).
Sleep Studies
Dyspnea and Fatigue Questionnaires
OSA diagnosis was obtained by full polysomnography
Levels of fatigue over the past 6 months were measured
or by simplified polygraphy without electroencephalogram
using the 11-item Chalder fatigue scale,29 which consists
recordings. Sleep was scored manually according to stan-
of 4 items for mental fatigue and 7 items for physical
dard criteria. Polysomnography used continuous acquisi-
fatigue. Dyspnea was evaluated with the Sadoul question-
tion of the following recordings: electro-oculogram (3 chan-
naire.30
nels), electroencephalogram (3 channels), electromyogram
(1 channel), and electrocardiogram (1 channel). Air flow
Cardiovascular Score
was measured using nasal pressure associated with the
sum of oral and nasal thermistor signals. Respiratory effort
Ten-year risk of fatal cardiovascular event (Systematic
was monitored with abdominal and thoracic bands. An
Coronary Risk Evaluation Project [SCORE] calculation
apnea was defined as a complete cessation of air flow for
for European countries) was calculated.28 For data analy-
at least 10 s and a hypopnea as a reduction of at least 50%
sis, cardiovascular SCORE was automatically set at 15% if
in the nasal pressure signal or a decrease between 30% and
participants were in secondary prevention (ie, suffered a
50% associated with either oxygen desaturation of at least
cardiovascular event in the past).
4% or electroencephalogram arousal. Apneas were classi-
fied as obstructive, central, or mixed according to the pres-
Blood Pressure and Heart Rate Measurements
ence or the absence of respiratory effort. The criterion for
sleep apnea in this study was an apnea-hypopnea in-
Clinic BP was measured in the supine position by mer-
dex ⱖ 10 events/h of sleep.
cury sphygmomanometer on 2 occasions (2 consecutive
days), with 3 measurements spaced by 1 min on each
occasion, according to European guidelines on hyperten- Accelerometer Data Collection and Analysis
sion.31 Clinic heart rate was measured by pulse palpation
(30 s) after the third measurement in the supine position. Daily physical activity was assessed using an activity
The following clinic parameters were assessed: systolic monitor (Sensewear Pro2 armband, Body Media, Pitts-
BP (SBP), diastolic BP (DBP), and heart rate. Office hy- burgh, Pennsylvania).34 The Sensewear Pro2 armband con-
pertension was defined as a mean (average of the 3 mea- tains accelerometers that sense movement in 2 planes, a
surements at each of the 2 office sessions) office SBP galvanic skin sensor, a temperature sensor, and a near-
ⱖ 140 mm Hg and/or a mean office DBP ⱖ 90 mm Hg.31 patient temperature sensor. Subjects wore the activity mon-
Self-measured home BP monitoring was performed on 3 itor for the same period as BP monitoring (ie, 72 h), al-
consecutive days using a validated fully automated elec- lowing an estimate of mean daily physical activity by
tronic device (Omron 705CP, Omron Corporation, Tokyo, averaging data over 1 weekday and the weekend.
Japan). Participants were trained to use this device, and The intensity of activity for each minute of wear time
they were instructed to take 3 measures in the morning was calculated and expressed using metabolic equivalents
(between awakening and breakfast) and 3 measures in the (METs) to determine daily time spent in sedentary, light,
evening (between dinner and bedtime) after 5 min sitting or moderate-to-vigorous activity. Sedentary behavior was
at rest and with 1-min intervals between measurements. A defined as activities resulting in energy expenditure ⬍ 1.5
form was supplied to the participants to report all self- METs, which is equivalent to sitting.35 Light-intensity phys-
measured BP values. For each subject, mean of morning ical activity was defined as 1.5–3 METs, moderate phys-
and evening measures was calculated. The circumference ical activity as 3– 6 METs, vigorous physical activity as
at the midpoint of the upper arm was measured, and a 6 –9 METs and very vigorous physical activity as ⬎ 9

1220 RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

METs. This device also allows estimation of sleep dura- Table 1. Anthropometric, Clinical, and Polysomnographic
tion. Characteristics

Clinical Data Mean ⫾ SD (range) (N ⫽ 95)

Statistical Analysis
Age (y) 63.3 ⫾ 8.8 (40.0–82.0)
Continuous data were presented as mean ⫾ SD and cat- Gender (men, %) 79 (83.2)
egorical data as percentages (%). Variables with ⬎ 15% of BMI (kg/m2) 29.9 ⫾ 4.9 (21.0–45.0)
missing values were not included in the analysis. Colin- BMI ⬍ 25 kg/m2 (n, %) 12 (12.6)
earity was assessed with Pearson’s or Spearman’s coeffi- 25 ⬍ BMI ⬍ 30 kg/m2 (n, %) 40 (42.1)
cient (depending on validation of normality distribution) 30 ⬍ BMI ⬍ 35 kg/m2 (n, %) 29 (30.5)
35 ⬍ BMI ⬍ 40 kg/m2 (n, %) 11 (11.6)
or Cramer’s V2. Normality of the data were verified using
BMI ⬎ 40 kg/m2 (n, %) 3 (3.1)
Skewness and Kurtosis tests.
Office systolic blood pressure 138.4 ⫾ 18.4 (108.0–200.0)
Relationship between self-measured BP and physical (mm Hg)
activity was analyzed with logistic regressions. A SBP Office diastolic blood pressure 80.7 ⫾ 12.4 (57.0–120.0)
ⱖ 135 mm Hg or a DBP ⱖ 85 mm Hg for nondiabetic sub- (mm Hg)
jects and a SBP ⱖ 125 mm Hg or a DBP ⱖ 80 mm Hg for Type 2 diabetes (%) 34.0
diabetic subjects were considered as high levels of BP.32,33 Smoker (never:current:former; 32 (33.7): 15 (15.8): 48 (50.5)
Two sex-, age-, and BMI-adjusted linear regression mod- n, %)
Alcohol (units/d) 1.2 ⫾ 1.7 (0–7.0)
els, stratified by recruiting centers, were built: one for
Cardiovascular risk score (%) 13.5 ⫾ 3.7 (5.0–27.6)
self-measured morning BP and the other for self-measured
Patients in secondary prevention 57 (60)
evening BP. In the univariate logistic regression, when a (n, %)
continuous variable was not log-linear, it was recoded by Home self-measured blood pressure
creating a new variable from the quartiles or the median Morning systolic blood pressure 135.2 ⫾ 13.7 (105.9–164.0)
depending on the quality of information. Independent pa- (mm Hg)
rameters were included in the multivariate model when Morning diastolic blood pressure 81.3 ⫾ 9.7 (60.8–122.1)
significance was ⱕ 0.1 in the univariate model. Missing (mm Hg)
Evening systolic blood pressure 130.5 ⫾ 15.3 (102.3–174.9)
values were replaced by the median for continuous data
(mm Hg)
and by the most frequent value for categorical data. A Evening diastolic blood pressure 77.1 ⫾ 9.5 (46.8–105.0)
backward selection was employed for multivariate model. (mm Hg)
Results were considered statistically significant when Lipid profile
P value was ⬍ .05. Statistical analysis was performed with Total cholesterol (mmol/L) 4.71 ⫾ 1.13 (2.66–7.79)
the SAS 9.1.3 package (SAS Institute, Cary, North Caro- HDL (mmol/L) 1.29 ⫾ 0.53 (0.37–4.14)
lina). LDL (mmol/L) 2.56 ⫾ 1.06 (0.81–5.33)
Fasting glucose (g/L) 1.13 ⫾ 0.32 (0.51–2.63)
Results Epworth sleepiness score 8.0 ⫾ 4.4 (0–20.0)
Chalder fatigue scale
Study Participant Characteristics Physical symptoms 16.2 ⫾ 3.6 (7.0–28.0)
Mental symptoms 9.1 ⫾ 1.9 (5.0–15.0)
Subjects’ characteristics are presented in Table 1. Sub- Sadoul dyspnea score 41.3 ⫾ 17.5 (0–5.0)
Sleep studies
jects’ mean age was 63.3 ⫾ 8.8 y, and they were predomi-
AHI (No. of events/h of sleep) 41.3 ⫾ 17.5 (10.2–98.6)
nantly male (83.2%) and overweight or obese (BMI of
Mean nocturnal SpO2 (%) 92.4 ⫾ 2.4 (83.0–96.0)
29.9 ⫾ 4.9 kg/m2). By definition, subjects had high 10-y car- Minimum nocturnal SpO2 (%) 78.5 ⫾ 8.5 (49.0–92.0)
diovascular risk scores (13.5 ⫾ 3.7%). They exhibited mod- CT90 (%) 14.6 ⫾ 19.7 (0–85.0)
erate to severe apnea-hypopnea index (41.3 ⫾ 17.5/h) with a
significant amount of nocturnal hypoxia (cumulative time BMI ⫽ body mass index
HDL ⫽ high density lipoprotein
spent with SpO2 below 90%: 14.64 ⫾ 19.73%). Sleepiness as LDL ⫽ low density lioprotein
measured by the Epworth sleepiness scale was at the upper AHI ⫽ apnea hypopnea index
CT90 (%) ⫽ cumulative time spent with SpO2 below 90% (% of time of recording)
limit of the normal range (8.0 ⫾ 4.4). Sleep duration esti-
mated at home on 3 different nights by accelerometer was
353.7 ⫾ 85.1 min (5.9 ⫾ 1.4 h) (Table 2).
2). Average daily MET levels were 1.3 ⫾ 0.3, which is
Physical Activity and Sedentary Behaviors significantly ⬍ 1.5 (ie, the cut-off point below which sed-
entary behavior is defined).35 Time periods spent in sed-
Average number of steps/d was 7,393 ⫾ 3,545, and mean entary behaviors, light physical activity, and moderate phys-
daily energy expenditure was 2,622.6 ⫾ 515.9 kcal (Table ical activity were 736.3 ⫾ 138.6, 224.1 ⫾ 90.9, and 95.1

RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8 1221

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

Table 2. Physical Activity Measurements

Mean ⫾ SD (range)
Measurements
(N ⫽ 95)

Energy expenditure (kcal) 2,622.6 ⫾ 515.9 (1,625.0–3,892.0)

Steps/d (n) 7,393.0 ⫾ 3,544.8 (801.0–17,761.0)
Average METs 1.3 ⫾ 0.3 (0.8–2.4)
Exercise intensity
Total time in sedentary 736.3 ⫾ 138.6 (342.0–1,062.0)
behaviors (0–1.5 METs,
min)
Total time in light physical 224.1 ⫾ 90.9 (48.0–505.0)
activity (1.5–3 METs, min)
Total time in moderate physical 95.07 ⫾ 86.47 (0–416.0)
activity (3–6 METs, min)
Total time in vigorous physical 2.89 ⫾ 7.79 (0–59.0)
activity (6–9 METs, min)
Total time in very vigorous 0.05 ⫾ 0.37 (0–3.0)
physical activity (⬎ 9 METs,
min)
Sleep duration (min) 353.7 ⫾ 85.1 (120.0–516.0)
Total time lying down (min) 453.3 ⫾ 86.6 (213.0–720.0)
Fig. 1. Correlation between body mass index (BMI) and steps/d.
MET ⫽ metabolic equivalent The dotted line represents the recommended level of steps/d for
optimal health.36

⫾ 86.5 min, respectively (12.7 ⫾ 2.3; 3.7 ⫾ 1.5, and 1.6 cardiovascular SCORE, and cumulative time spent with
⫾ 1.5 h, respectively) (Table 2). SpO2 below 90% as significant determinants for morning
self-measured BP (see Table S1 in the supplementary ma-
Association Between Obesity and Sleep Apnea terials at http://www.rcjournal.com). In multivariate anal-
Severity and Physical Activity and Sedentary ysis, secondary prevention was associated with morning self-
Behaviors measured BP in the normal range (odds ratio [OR], 0.22,
95% CI, 0.05– 0.93, P ⫽ .039), whereas alcohol consumption
Body mass index was inversely correlated with steps/d was associated with higher values of self-measured BP (OR,
(r ⫽ ⫺0.28, P ⫽ .007) (Fig. 1) and daily METs (r ⫽ ⫺0.43, 10.43, 95% CI, 2.36 – 46.04, P ⫽ .002).
P ⬍ .001). Steps/d did not correlate with sleepiness or
other indices of OSA severity (Fig. 2, ESS in panel A and Evening BP. In univariate analysis, cardiovascular risk
mean nocturnal SpO2 in panel B). SCORE, cumulative time spent with SpO2 below 90%, num-
ber of steps/d, and time spent in very vigorous physical
Association of Physical Activity and Sedentary activity (⬎ 9 METs) were linked with evening self-mea-
Behaviors With BP sured BP in the normal range (see Table S1 in the supple-
mentary materials at http://www.rcjournal.com). On the
Total energy expenditure was inversely correlated with other hand, time spent in sedentary behaviors, BMI, type
office DBP (r ⫽ ⫺0.22, P ⫽ .037). Steps/d were inversely 2 diabetes, and high scores for Sadoul and Chalder scales
correlated with evening self-measured SBP (r ⫽ ⫺0.27, were identified as determinants of elevated evening self-
P ⫽ .01) (Fig. 3A) but not morning self-measured BP. measured BP. Smoking was not identified as a predictor of
Time spent in sedentary behaviors (0 –1.5 METs) was pos- elevated morning or evening BP.
itively associated with self-measured morning and evening Multivariate analysis identified cardiovascular score ⬎ 15%
SBP (Fig. 3B) and DBP (r ⫽ 0.32, P ⫽ .002; r ⫽ 0.28, (OR, 0.20, 95% CI, 0.04 – 0.96, P ⫽ .044) and number of
P ⫽ .006 and r ⫽ 0.29, P ⫽ .004; r ⫽ 0.27, P ⫽ .009, re- steps/d (for 6,932.5 ⬍ steps/d ⬍ 10,012 vs ⬍ 4,974.5 steps/d:
spectively). OR, 0.15, 95% CI, 0.02– 0.89, P ⫽ .037) and (for steps/d
ⱖ 10,012 vs ⬍ 4,974.5: OR, 0.14, 95% CI, 0.02– 0.86,
Determinants of Blood Pressure P ⫽ .034) as significant determinants of evening self-mea-
sured BP in the normal range (Fig. 4). Presence of diabetes
Morning BP. Univariate analysis, adjusted for age, sex, was associated with elevated evening self-measured BP (OR,
and BMI and stratified for recruiting center, identified 7.37, 95% CI, 1.71⫺31.73, P ⫽ .007).

1222 RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

Fig. 2. A: Epworth sleepiness score; B: mean nocturnal oxygen saturation (SapO2) plotted against steps/d.

Fig. 3. A: correlation between steps/d and self-measured evening systolic blood pressure (SBP); B: correlation between time spent in
sedentary behavior and self-measured evening SBP.

Discussion entary behaviors are the most important predictors of

evening self-measured BP. Thus, in individuals with
In adults with OSA presenting high cardiovascular risk, OSA, daily physical activity impacts BP levels but with
physical activity is reduced and below consensus recom- a different magnitude in the morning and evening. Self-
mendations for optimal health. The level of physical ac- measured BP measures were performed, as this type of
tivity is significantly related to the degree of obesity but measure seems to have a stronger predictive power for
not linked with the severity of sleep apnea or sleepiness. future cardiovascular events.37 Both morning and eve-
The main finding of our study was that steps/d and sed- ning home BP have been shown to be positively asso-

RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8 1223

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

Fig. 4. Multivariate analysis for self-measured evening blood pressure. Odds ratios and 95% CI are depicted. BMI ⫽ body mass index.

ciated with the incidence of stroke,37,38 indicating that predictor of target organ damages and even incident
evening BP has clinical relevance. Thus, in the present cardiovascular events compared with office measure-
study, we created 2 univariate models: for morning and ments of BP.46
evening BP. Study participants were representative of a typical mid-
Our study is the first to examine the relationship be- dle-aged, male OSA population with co-morbidities.5,47
tween objective measures of physical activity and seden- Sleepiness for the whole group was at the upper limit of
tary behaviors and home self-measured BP in high cardio- the normal range. OSA patients also suffering from asso-
vascular risk OSA subjects. Only 2 previous studies in the ciated cardiovascular diseases48 or diabetes49 tend to not
field have reported interesting but more limited data on complain from sleepiness but rather of fatigue or dyspnea.
higher energy expenditure during sleep39 or mean activity Only 25% of the included subjects reached recommended
in arbitrary units.39 Interestingly, physical activity was not levels of physical activity (10,000 steps/d). There was an
increased by CPAP treatment, suggesting that OSA per se inverse correlation between the number of steps/d and BMI
is not the main contributor to reduced physical activity consistent with the well-documented inverse relationship
levels.40 In our study, we measured physical activity pat- between physical activity and weight gain.50 A previous
terns with an accelerometer validated in healthy sub- study that assessed physical activity by questionnaires
jects41 and in diabetics.34 This device, which has also found that physical activity was a stronger predictor of
been recently validated in OSA during sleep,39 is con- perceptions of fatigue and energy than OSA severity.27
sidered one of the most reliable means of estimating These results contrast with our study, in which subjective
physical activity and its patterns in free living condi- sleepiness was not associated with spontaneous physical
tions.42,43 In a study comparing the validity of 6 activity activity. As questionnaires are known to be subject to
monitors, the SenseWear Armband (a biaxial monitor) widely varying bias,51 our study has the advantage of pro-
was 1 of the 2 most valid monitors during standardized viding objectively measured spontaneous physical activity
physical activities.44 A major strength of our study was in individuals with OSA.
to provide objective measures of sedentary behaviors
concurrently to home self-measurement of BP. The rec-
Blood Pressure: Respective Role of Physical Activity
ommended duration for such a BP assessment is 72 h,45
and OSA
and we decided to correlate physical activity and BP
measurement in this time window. We do, however,
acknowledge that averaging measures from a 7-d period Approximately 60% of individuals with OSA exhibit
would have been ideal. Another strength and originality daytime hypertension.52 The dose-response relationship be-
of this study is that BP was assessed by home self- tween the severity of OSA and the incidence of hyperten-
measurement of BP, which is now accepted as a better sion has been clearly demonstrated both in the general

1224 RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

population2 and in clinical cohorts.1 OSA is now acknowl- related to a number of cardiovascular health outcomes.58
edged as a cause of hypertension in European and Amer- Pedometers may prove a more financially viable option to
ican guidelines on hypertension management.31,53 How- increase physical activity levels and concomitantly reduce
ever, it has been shown that CPAP intervention only has a sedentary behavior. Simpler recommendations such as in-
limited effect on BP control with pooled mean changes creasing steps/d, ideally to 10,000/d, can be easily inte-
of ⫺2.46 mm Hg in SBP and ⫺1.83 mm Hg in 24-h DBP.7 grated in daily life activities (transport, domestic tasks)
This suggests that OSA per se, which is frequently asso- and should be implemented as a combined therapy with
ciated with obesity, is only one of the many potential CPAP treatment. Moreover, as the effects of CPAP on
contributors to uncontrolled BP. In the multivariate anal- cardiometabolic markers are minimal,58,59 any increase in
ysis of morning self-measured BP, only alcohol consump- physical activity will positively impact on lipid and glu-
tion and diabetes were identified as determinants for hy- cose metabolism abnormalities.
pertension. Secondary cardiovascular prevention was
associated with lower morning self-measured BP. The re- Conclusions
lation between hypertension and excess alcohol consump-
tion and diabetes is well established.54 Among the poten-
Thus, for individuals with OSA presenting high cardio-
tial contributors to elevated BP are sedentary behaviors,
vascular risk, a combined approach integrating CPAP and
which have emerged as an independent risk factor for
lifestyle interventions including physical activity programs
cardiovascular disease.55 In the present study, sedentary
would improve BP control and, in turn, potentially reduce
behaviors were positively associated with self-measured
mortality. Regular physical activity can also enhance the
morning and evening SBP and DBP. Furthermore, univar-
effects of CPAP therapy, as it has been associated with
iate analysis identified time spent in sedentary behaviors
reduced incidence and severity of sleep-disordered breath-
as a determinant of elevated evening self-measured BP.
ing in longitudinal and cross-sectional studies.60-63
Our results are supported by a prospective cohort study in
which self-reported total sedentary behaviors were directly
associated with a higher risk of incident hypertension.17 REFERENCES
In our study, multivariate analysis identified steps/d as
1. Marin JM, Agusti A, Villar I, Forner M, Nieto D, Carrizo SJ, et al.
a determinant of lower evening self-measured BP. A pop- Association between treated and untreated obstructive sleep apnea
ulation-based study showed that leisure activity at differ- and risk of hypertension. JAMA 2012;307(20):2169-2176.
ent life stages was associated with lower SBP and DBP in 2. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the
mid-adulthood by ⬃1–2 mm Hg and with reduced risk of association between sleep-disordered breathing and hypertension.
hypertension by ⬃23% (active vs nonactive), although this N Engl J Med 2000;342(19):1378-1384.
3. Lévy P, Pépin JL, Arnaud C, Baguet JP, Dematteis M, Mach F.
reduction was influenced by age.56 Obstructive sleep apnea and atherosclerosis. Prog Cardiovasc Dis
2009;51(5):400-410.
Study Limitations 4. Tamisier R, Pépin JL, Rémy J, Baguet JP, Taylor JA, Weiss JW,
Lévy P. 14 nights of intermittent hypoxia elevate daytime blood
This study could have benefited from a control group, pressure and sympathetic activity in healthy humans. Eur Respir J
2011;37(1):119-128.
which would have allowed the study to establish whether 5. Cano-Pumarega I, Durán-Cantolla J, Aizpuru F, Miranda-Serrano E,
the same determinants impact BP in individuals presenting Rubio R, Martínez-Null C, et al. Obstructive sleep apnea and sys-
high cardiovascular risk without OSA. It would have been temic hypertension: longitudinal study in the general population; the
interesting to measure dietary patterns of participants in Vitoria Sleep Cohort. Am J Respir Crit Care Med 2011;184(11):
the present study to establish whether certain dietary hab- 1299-1304.
6. O’Connor GT, Caffo B, Newman AB, Quan SF, Rapoport DM,
its contributed to elevated BP, as several epidemiological Redline S, et al. Prospective study of sleep-disordered breathing and
surveys have suggested that diet is an important factor in hypertension: the Sleep Heart Health Study. Am J Respir Crit Care
modulating BP.57 Med 2009;179(12):1159-1164.
7. Bazzano LA, Khan Z, Reynolds K, He J. Effect of nocturnal nasal
Practical Implications in the Management of OSA continuous positive airway pressure on blood pressure in obstructive
sleep apnea. Hypertension 2007;50(2):417-423.
Patients 8. Haentjens P, Van Meerhaeghe A, Moscariello A, De Weerdt S,
Poppe K, Dupont A, Velkeniers B. The impact of continuous posi-
In our final multivariate model, only steps/d were iden- tive airway pressure on blood pressure in patients with obstructive
tified as a determinant for lower evening self-measured sleep apnea syndrome: evidence from a meta-analysis of placebo-
BP. Steps per day can be measured simply with a pedom- controlled randomized trials. Arch Intern Med 2007;167(8):757-764.
9. Barbé F, Durán-Cantolla J, Capote F, de la Peña M, Chiner E, Masa
eter, which is far less expensive than the accelerometer we JF, et al. Long-term effect of continuous positive airway pressure in
used in the present study. The emergent body of evidence hypertensive patients with sleep apnea. Am J Respir Crit Care Med
suggests that pedometer-determined physical activity is 2010;181(7):718-726.

RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8 1225

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

10. Pépin JL, Tamisier R, Barone-Rochette G, Launois SH, Lévy P, 27. Hong S, Dimsdale JE. Physical activity and perception of energy and
Baguet JP. Comparison of continuous positive airway pressure and fatigue in obstructive sleep apnea. Med Sci Sports Exerc 2003;35(7):
valsartan in hypertensive patients with sleep apnea. Am J Respir Crit 1088-1092.
Care Med 2010;182(7):954-960. 28. Conroy RM, Pyo K, Backer GD, Bacquer DD, Ducimetie P, Keil U,
11. Haskell WL, Lee IM, Pate RR, Powell KE, Blair SN, Franklin BA, et al. Estimation of ten-year risk of fatal cardiovascular disease in
et al. Physical activity and public health: updated recommendation Europe: the SCORE project. Science 2003;24(1):987-1003.
for adults from the American College of Sports Medicine and the 29. Chalder T, Berelowitz G, Pawlikowska T, Watts L, Wessely S, Wright
American Heart Association. Circulation 2007;116(9):1081-1093. D, Wallace EP. Development of a fatigue scale. J Psychosom Res
12. Wen CP, Wai JPM, Tsai MK, Yang YC, Cheng TY, Lee MC, et al. 1993;37(2):147-153.
Minimum amount of physical activity for reduced mortality and 30. Sadoul P. Assessment of functional respiratory deficit. Bull Eur Phys-
extended life expectancy: a prospective cohort study. Lancet 2011; iopathol Resp 1983;19(1):3-6.
378(9798):1244-1253. 31. Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R,
13. Bassett DR Jr, Fitzhugh EC, Crespo CJ, King GA, McLaughlin JE. Germano G, et al. 2007 Guidelines for the management of arterial
Physical activity and ethnic differences in hypertension prevalence hypertension: the task force for the management of arterial hyper-
in the United States. Prev Med 2002;34(2):179-186. tension of the European Society of Hypertension (ESH) and of the
14. Pereira MA, Folsom AR, McGovern PG, Carpenter M, Arnett DK, European Society of Cardiology (ESC). J Hypertens 2007;25(6):
Liao D, et al. Physical activity and incident hypertension in black 1105-1187.
and white adults: the Atherosclerosis Risk in Communities Study. 32. NICE. NICE guideline CG66: type 2 diabetes: the management of
Prev Med 1999;28(3):304-312. type 2 diabetes (updated May 2009). http://wwwniceorguk/CG66.
15. Owen N, Healy GN, Matthews CE, Dunstan DW. Too much sitting: Accessed March 17, 2012.
the population health science of sedentary behavior. Exerc Sport Sci 33. Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter
Rev 2010;38(3):105-113. JF, et al. Guidelines for management of hypertension: report of the
16. Aadahl M, Kjaer M, Jørgensen T. Influence of time spent on TV fourth working party of the British Hypertension Society, 2004-BHS
viewing and vigorous intensity physical activity on cardiovascular IV. J Hum Hypertens 2004;18(3):139-185.
biomarkers: the Inter 99 study. Eur J Cardiovasc Prev Rehabil 2007; 34. Mignault D, St-Onge M, Karelis AD, Allison DB, Rabasa-Lhoret R.
14(5):660-665. Evaluation of the Portable HealthWear Armband: a device to mea-
17. Beunza JJ, Martínez-González MA, Ebrahim S, Bes-Rastrollo M, sure total daily energy expenditure in free-living type 2 diabetic
Núñez J, Martínez JA, Alonso A. Sedentary behaviors and the risk of individuals. Diabetes Care 2005;28(1):225-227.
incident hypertension: the SUN Cohort. Am J Hypertens 2007;20(11): 35. Sedentary Behaviour Research N. Standardized use of the terms
1156-1162. “sedentary” and “sedentary behaviours.” Appl Physiol Nutr Metab
18. Cintra F, Poyares D, Rizzi CF, Risso TT, Skomro R, Montuori E, et 2012;37(3):540-542.
al. Cardiorespiratory response to exercise in men and women with 36. Tudor-Locke C. Steps to better cardiovascular health : how many
obstructive sleep apnea. Sleep Med 2009;10(3):368-373. steps does it take to achieve good health and how confident are we
19. Kato M, Phillips BG, Sigurdsson G, Narkiewicz K, Pesek CA, Som- in this number? Curr Cardiovasc Risk Rep 2010;4(4)271-276.
ers VK. Effects of sleep deprivation on neural circulatory control. 37. Niiranen TJ, Hänninen MR, Johansson J, Reunanen A, Jula AM.
Hypertension 2000;35(5):1173-1175. Home-measured blood pressure is a stronger predictor of cardiovas-
20. Tochikubo O, Ikeda A, Miyajima E, Ishii M. Effects of insufficient cular risk than office blood pressure: the Finn-Home study. Hyper-
sleep on blood pressure monitored by a new multibiomedical re- tension 2010;55(6):1346-1351.
corder. Hypertension 1996;27(6):1318-1324. 38. Asayama K, Ohkubo T, Kikuya M, Obara T, Metoki H, Inoue R, et
21. Dean E, Bloom A, Cirillo M, Hong Q, Jawl B, Jukes J, et al. Asso- al. Prediction of stroke by home “morning” versus “evening” blood
ciation between habitual sleep duration and blood pressure and clin- pressure values: the Ohasama study. Hypertension 2006;48(4):737-
ical implications: a systematic review. Blood Press 2012;21(1):45- 743.
57. 39. O’Driscoll DM, Turton AR, Copland JM, Strauss BJ, Hamilton GS.
22. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration Energy expenditure in obstructive sleep apnea: validation of a mul-
is associated with reduced leptin, elevated ghrelin, and increased tiple physiological sensor for determination of sleep and wake. Sleep
body mass index. PLoS Med 2004;1(3):e62. Breath 2013;17(1):139-146.
23. Vorona RD, Winn MP, Babineau TW, Eng BP, Feldman HR, Ware 40. West SD, Kohler M, Nicoll DJ, Stradling JR. The effect of contin-
JC. Overweight and obese patients in a primary care population uous positive airway pressure treatment on physical activity in pa-
report less sleep than patients with a normal body mass index. Arch tients with obstructive sleep apnoea: a randomised controlled trial.
Intern Med 2005;165(1):25-30. Sleep Med 2009;10(9):1056-1058.
24. Aghamohammadzadeh R, Heagerty AM. Obesity-related hyperten- 41. Jakicic JM, Marcus M, Gallagher KI, Randall C, Thomas E, Goss
sion: epidemiology, pathophysiology, treatments, and the contribu- FL, Robertson RJ. Evaluation of the SenseWear Pro Armband to
tion of perivascular adipose tissue. Ann Med 2012; 44 Suppl 1:S74- assess energy expenditure during exercise. Med Sci Sports Exerc
S84. 2004;36(5):897-904.
25. Eckel RH, Wassef M, Chait A, Sobel B, Barrett E, King G, et al. 42. Pitta F, Troosters T, Probst VS, Spruit MA, Decramer M, Gosselink
Prevention Conference VI: diabetes and cardiovascular disease. Writ- R. Quantifying physical activity in daily life with questionnaires and
ing Group II: pathogenesis of atherosclerosis in diabetes. Circulation motion sensors in COPD. Eur Respir J 2006;27(5):1040-1055.
2002;105(18):e138-e143. 43. Van Remoortel H, Giavedoni S, Raste Y, Burtin C, Louvaris Z,
26. Schmid SM, Hallschmid M, Jauch-Chara K, Wilms B, Benedict C, Gimeno-Santos E, et al. Validity of activity monitors in health and
Lehnert H, et al. Short-term sleep loss decreases physical activity chronic disease: a systematic review. Int J Behav Nutr Phys Act
under free-living conditions but does not increase food intake under 2012;9(1):84.
time-deprived laboratory conditions in healthy men. Am J Clin Nutr 44. Van Remoortel H, Raste Y, Louvaris Z, Giavedoni S, Burtin C,
2009;90(6):1476-1482. Langer D, et al. Validity of six activity monitors in chronic obstruc-

1226 RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8

 PHYSICAL ACTIVITY AND BLOOD PRESSURE IN OSA

tive pulmonary disease: a comparison with indirect calorimetry. PLoS 54. Menotti A, Blackburn H, Kromhout D, Nissinen A, Adachi H, Lanti
One 2012;7(6):e39198. M. Cardiovascular risk factors as determinants of 25-year all-cause
45. Parati G, Stergiou GS, Asmar R, Bilo G, de Leeuw P, Imai Y, et al. mortality in the seven countries study. Eur J Epidemiol 2001;17(4):
European Society of Hypertension practice guidelines for home blood 337-346.
pressure monitoring. J Hum Hypertens 2010;24(12):779-785. 55. Thorp AA, Owen N, Neuhaus M, Dunstan DW. Sedentary behaviors
46. Bobrie G, Chatellier G, Genes N, Clerson P, Vaur L, Vaisse B, et al. and subsequent health outcomes in adults a systematic review of
Cardiovascular prognosis of “masked hypertension” detected by blood longitudinal studies, 1996-2011. Am J Prev Med 2011;41(2):207-
pressure self-measurement in elderly treated hypertensive patients. 215.
JAMA 2004;291(11):1342-1349. 56. Pouliou T, Ki M, Law C, Li L, Power C. Physical activity and
47. Basta M, Lin HM, Pejovic S, Sarrigiannidis A, Bixler E, Vgontzas sedentary behaviour at different life stages and adult blood pressure
AN. Lack of regular exercise, depression, and degree of apnea are in the1958 British cohort. J Hypertens 2012;30(2):275-283.
predictors of excessive daytime sleepiness in patients with sleep 57. Savica V, Bellinghieri G, Kopple JD. The effect of nutrition on blood
apnea: sex differences. J Clin Sleep Med 2008;4(1):19-25. pressure. Annu Rev Nutr 2010;30:365-401.
48. Garrigue S, Pépin JL, Defaye P, Murgatroyd F, Poezevara Y, Clé-
58. Drager LF, Lorenzi-Filho G. CPAP for obstructive sleep apnea and
menty J, Lévy P. High prevalence of sleep apnea syndrome in pa-
the metabolic syndrome. N Engl J Med 2012;366(10):964; author
tients with long-term pacing: the European Multicenter Polysomno-
reply 965-966.
graphic Study. Circulation 2007;115(13):1703-1709.
59. Hoyos CM, Killick R, Yee BJ, Phillips CL, Grunstein RR, Liu PY.
49. Foster GD, Sanders MH, Millman R, Zammit G, Borradaile KE,
Cardiometabolic changes after continuous positive airway pressure
Newman AB, et al. Obstructive sleep apnea among obese patients
for obstructive sleep apnoea: a randomised sham-controlled study.
with type 2 diabetes. Diabetes Care 2009;32(6):1017-1019.
50. Hill JO, Wyatt HR. Role of physical activity in preventing and Thorax 2012;67(12)1081-1089.
treating obesity. J Appl Physiol 2005;99(2):765-770. 60. Awad KM, Malhotra A, Barnet JH, Quan SF, Peppard PE. Exercise
51. Luke A, Dugas LR, Durazo-Arvizu RA, Cao G, Cooper RS. Assess- is associated with a reduced incidence of sleep-disordered breathing.
ing physical activity and its relationship to cardiovascular risk fac- Am J Med 2012;125(5):485-490.
tors: NHANES 2003-2006. BMC Public Health 2011;11:387. 61. Giebelhaus V, Strohl KP, Lormes W, Lehmann M, Netzer N. Phys-
52. Pedrosa RP, Drager LF, Gonzaga CC, Sousa MG, de Paula LKG, ical exercise as an adjunct therapy in sleep apnea: an open trial. Sleep
Amaro ACS, et al. Obstructive sleep apnea: the most common sec- Breath 2000;4(4):173-176.
ondary cause of hypertension associated with resistant hypertension. 62. Peppard PE, Young T. Exercise and sleep-disordered breathing:
Hypertension 2011;58(5):811-817. an association independent of body habitus. Sleep 2004;27(3):
53. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, 480-484.
Izzo JL, et al. The Seventh Report of the Joint National Committee 63. Sengul YS, Ozalevli S, Oztura I, Itil O, Baklan B. The effect of
on Prevention, Detection, Evaluation, and Treatment of High Blood exercise on obstructive sleep apnea: a randomized and controlled
Pressure: the JNC 7 report. JAMA 2003;289(19):2560-2572. trial. Sleep Breath 2011;15(1):49-56.

RESPIRATORY CARE • AUGUST 2014 VOL 59 NO 8 1227