Exercise capacity reflects the ability to perform
activities of daily living that require aerobic
metabolism, and its assessment provides diagnostic and
prognostic information for several respiratory and
cardiovascular diseases (1). The impact of obstructive
sleep apnoea (OSA) on exercise capacity remains
unclear, given some studies have shown that this
disorder reduces exercise capacity (2–6), whereas
others have found no differences in exercise capacity
between OSA and non-OSA populations (7–12).
Several factors impact aerobic capacity, as
measured by cardiopulmonary exercise testing (CPET).
Severe obesity reduces exercise capacity because fat
mass interferes with cardiac and respiratory function
(13). Obesity is also considered a major risk factor for
OSA because there is a tendency for OSA to worsen
with significant weight gain (14). A person’s sex also
has an influence on exercise capacity, which could be
at least 20%–40% lower in women than in men (15).
These variables have not always been considered,
however, when interpreting the results of studies that
have evaluated exercise capacity in patients with OSA
(2–10,12). Moreover, these reports have included only
a small number of women; therefore, their results
might not extend to this population. Furthermore, only
one study has included CPET measurements obtained
from subjects with morbid obesity (3).
A delayed heart rate (HR) recovery after CPET is
considered a predictor of overall mortality (16). Some
studies have shown that patients with OSA have an
attenuated HR recovery that is dependent on OSA
severity (3,5,11,17). A possible mechanism involves
reduced parasympathetic activity, which is observed in
patients with OSA during the recovery phase of
exercise (11). However, HR recovery in patients with
OSA has primarily been analysed in men because
previous studies have included only a small number of
women (3,11,17). This bias is of particular importance,
given HR recovery might be decreased in women
compared with men in some populations, such as those
affected with heart disease (18). Because previous
studies have been performed primarily on men, it is
important to also know how OSA impacts HR recovery
in women.
Therefore, the purpose of this study was to study
exercise capacity and cardiovascular response to CPET
in morbidly obese female patients with OSA.
Twenty-four morbidly obese female patients newly
diagnosed with OSA were studied from January 2005
to December 2006. All were on a waiting list for
bariatric surgery due to morbid obesity. Inclusion
criteria were as follows: apnoea-hypopnea index (AHI) >5 h-1 found in the outpatient sleep register, age between 18 and 65 years, body mass index (BMI) ≥40 kg.m-2 and female sex. Exclusion criteria were heart, renal, liver or neurologic disease; pregnancy; unwillingness or inability to perform testing procedures; previous OSA diagnosis; previous CPAP
use; and chronic oxygen therapy. The study was
approved by the Institutional Ethics Committee at La
Paz University hospital and Principe de Asturias
University Hospital, both located in Madrid, Spain. All
the patients provided their written informed consent
prior to any procedure.
At baseline, all the patients’ anthropometric data
was recorded and an Epworth somnolence scale was
completed.
For every patient, an overnight sleep study was
performed using a portable device (Breas SC-20, Breas
Medical S.L., Mölnlyke, Sweden), which is a level 3
sleep monitoring diagnostic tool. Tracings were
initiated at approximately 11:00 PM and were
terminated at 6:00 AM, allowing at least 7 hours of
sleep recording. The device recorded airflow by nasal
pressure cannula, respiratory effort by thoracic and
abdominal bands and oxygenation by pulse oximetry,
as well as snoring and body position. All tracings were
analysed manually by the same pulmonologist, and respiratory events were scored according to the American Academy of Sleep Medicine (19). Apnoea was defined as a complete cessation in airflow ≥10 seconds, and hypopnea as a significant decrease in the amplitude of recorded airflow that lasted ≥10 seconds that was associated with oxygen desaturation >3%. From these measurements, AHI was calculated as the total number of apnoea and hypopnea events divided per hour of study. We also registered the oxygen desaturation index, the lowest oxygen saturation and
the percentage of time spent asleep with oxygen
saturation below 90%.
All the patients completed a symptom-limited
progressive CPET with an Ergobex cycle ergometer
(Bexen, Madrid, Spain). Expired gases and ventilatory
parameters were analysed breath-by-breath (sensor
Oxycon Alpha, Jaëger, Germany). Aerobic capacity
(V’O2), heart rate (HR), 12-lead electrocardiogram and
oxygen saturation were continuously recorded
throughout the test. After one minute of unloaded
cycling, the workload was increased by 20 watt.min-1
until the patient could not continue cycling, at which
moment peak exercise variables were registered.
Predicted maximal heart rate (MHR) was calculated
using equation 220-age (20). After achieving the peak workload, participants spent 90 seconds in a cool-down period at no workload, and HR data were registered
three times (HR1, HR2 and HR3 were registered at 30,
60 and 90 seconds after peak workload, respectively).
HR recovery was assessed as HR1, HR2 and HR3 and
also as a percentage decrease of MHR at HR1, HR2
and HR3.
The patients were classified according to AHI in two groups: mild OSA (AHI 5–15 h-1) and moderateto-severe OSA (AHI ≥15 h-1). Statistical analyses were first performed on the whole population, and then comparisons between groups were performed. The quantitative variables were expressed as means with standard deviation. For the categorical variables, frequencies and proportions were used. To compare two means, an unpaired t-test or the Mann-Whitney U test was used, as appropriate. Homoscedasticity was assessed by the Kolmogorov-Smirnov test. Correlations between the sleep register and CPET variables were evaluated using Pearson’s correlation coefficients. If differences were not normally distributed, Spearman’s correlation was employed. Statistical significance was set at a p-value ≤.05. The data were analysed using SPSS, v.10.1 (Chicago, IL, USA).
A total of twenty-four morbidly obese female
patients were included. Mean age and mean BMI were
respectively. Presented in Table 1 are the sleep data for
the entire study population. Mild, moderate and severe
OSA was detected in 14 (58.30%), 3 (8.60%) and 7
(29.20%) patients, respectively.
Mean peak workload in CPET was 97.72 ± 17.23
watts, and mean peak V’O2 was 15.58 ± 2.44 ml.min
1.kg-1. At maximal workload, the HR was 142 ± 19.14
beats.min-1 and during the recovery period, the HR
decreased 8.08% ± 3.76%, 14.14% ± 9.17% and
after maximal workload, respectively (Table 1).
severity of OSA
No differences were seen between groups with
respect to age and BMI. As expected, patients with
moderate-to-severe OSA had a higher AHI (p<.001)
and oxygen desaturation index than those with mild
OSA (p<.001).
Concerning the CPET results, peak workload was
not different between groups. Interestingly, peak V’O2
measured in moderate-to-severe OSA was not
statistically different from that of patients with mild
OSA (16.11 ± 2.37 vs. 15.19 ± 2.50 ml.min-1.kg-1,
p=.371). Additionally, both groups achieved similar
MHR during CPET and no differences were found in
HR recovery between patients with moderate-to-severe
and mild OSA at 30, 60 and 90 seconds after peak
workload. Nevertheless, the patients with moderate-to
severe OSA had an almost statistically significantly
lower percentage decrease of HR at HR3 compared
with patients with mild OSA (p=.073) (Table 2).
cardiopulmonary exercise testing
We searched for correlations between sleep events
and CPET results. Cross-sectional correlations between
V’O2 and AHI or the oxygen desaturation index were
not significant. With respect to HR recovery, the
percentage decrease of MHR at HR3 was negatively
correlated with the oxygen desaturation index (r=
correlation with AHI (r =-0.655, p=.056) (Table 3).
In morbidly obese female patients with OSA, peak
V’O2 recorded for patients with mild OSA is not
different from that recorded for patients with moderateto-severe OSA. In addition, HR recovery after peak
exercise is similar in both groups, but patients with
severe OSA tend to demonstrate a slower chronotropic
recovery.
Several studies have reported that exercise capacity
in patients with OSA is not different from that
registered in healthy subjects (7–12,21). However,
these results conflict with many other reports that have
described higher maximal V’O2 in non-OSA than in OSA populations (2–6,22). Reinforcing this last
observation, some studies have shown that AHI is
inversely correlated with peak V’O2 during CPET
(3,5,22,23). Nevertheless, other factors have also been
implicated in the exercise capacity of patients with
OSA, such as age, BMI, body weight and sex (2,10,22–
24). Rizzi et al. (10) compared response to CPET in
four groups of patients (OSA/controls and lean/obese
subjects) and found that maximal V’O2 did not differ
between patients with OSA and controls in both obese
and lean patients. However, maximal V’O2 was lower
in patients with obesity (OSA and controls) than in lean
subjects. Also, Butner et al. (22) reported lower peak
V’O2 in patients with moderate-to-severe OSA than in
those with milder OSA or in healthy patients, but these
differences were not detected when this variable was
controlled for body weight and age.
Several reports have evaluated exercise capacity in obese patients with OSA (2,6–8,10,12), but only one focused on morbidly obese subjects. Vanhecke et al.
(3) compared CPET results from 42 morbidly obese (BMI ≥35 kg.m-2) patients with OSA and 50 of these
without OSA and found a decreased peak V’O2 in the
patients with OSA. It is noteworthy that the V’O2
measured in our population was lower than that
reported by Vanhecke et al. (3) in their OSA group.
These differences might be explained by the fact that
we included only female patients, in contrast to the
study by Vanhecke et al. (3), in which almost one-third
of the enrolled population was male. It is known that
V’O2 is lower in women than in men due to their lesser
height, lower blood haemoglobin concentration and
smaller skeletal muscles, so these sex differences might
have affected the V’O2 in our report (15).
In addition to obesity, other factors could affect
exercise capacity in patients with OSA. For instance,
symptoms such as leg discomfort or dyspnoea during
CPET could limit maximal exercise capacity. The
impairment of glycolytic metabolism and the structural
and metabolic changes in the skeletal muscles of
patients with OSA could explain the leg discomfort
experienced during the test (25). Cardiac dysfunction
could also be responsible for exercise limitation, given
reduced peak V’O2 in patients with OSA has been
linked to lower right ventricular ejection fraction,
reduced cardiac output and reduced oxygen pulse (2,9).
Alonso-Fernández et al. (9) demonstrated that CPAP
therapy for 3 months was associated with significant
improvements in various indexes of left ventricular
systolic function response to exercise such as cardiac
output and stroke volume. Finally, sleep fragmentation
and excessive diurnal somnolence of those with OSA
could be related to decreased daily activity and lack of
fitness, resulting in submaximal efforts during CPET
(2).
The cardiovascular response to exercise in patients
with OSA was reported to be attenuated in several
studies (3,5,7,11). Patients with OSA have lower HR
recovery after peak exercise compared with healthy
individuals, and this impaired chronotropic response is
directly related to the severity of OSA (4,11). Although
our results did not show statistical differences in HR
recovery between both groups, we observed that
patients with moderate-to-severe OSA had a tendency
to show a blunted HR recovery at 90 seconds after
workload and that AHI and DI were almost
significantly correlated to this slower HR recovery.
This impaired HR recovery could be linked to the
increased nocturnal sympathetic activity that patients
with OSA show in relation to sleep fragmentation and
intermittent hypoxia. Thus, sympathetic overactivity
that persists even during wakefulness could reduce
sensitization and produce downregulation of cardiac
beta adrenergic receptors (7,26).
Our study has some limitations. The first is related
to the lack of an obese and a lean control group without
OSA. The absence of a control group makes it difficult
to accurately evaluate the influence of obesity and OSA
on exercise capacity. Another limitation might be
related to the fact that we only used clinical evaluations
to rule out heart disease in our population. Heart
disorders could impair exercise capacity measured during a CPET, and exercise capacity is also
considered a prognostic factor in ischemic and dilated
cardiomyopathies. All the patients in our study had a
nonpathological electrocardiogram because they were
on the waiting list for bariatric surgery, so these
procedures (clinical evaluation and electrocardiogram)
helped us detect important cardiac disorders. As a third
limitation, we are not able to confirm associations
between sympathetic overactivity and HR recovery
because we did not measure sympathetic activity.
In conclusion, in our study, morbidly obese female
patients exercise capacity was not different between
patients with mild OSA and those with moderate-to
severe OSA. However, the patients with moderate-to
severe OSA tended to have a slower HR recovery
response to exercise after they achieved a peak
workload.
[1] Arena R, Myers J, Williams MA, Gulati M,
Kligfield P, Balady GJ, et al. Assessment of
functional capacity in clinical and research
settings: a scientific statement from the American
Heart Association Committee on Exercise,
Rehabilitation, and Prevention of the Council on
Clinical Cardiology and the Council on
Cardiovascular Nursing. Circulation
2007;116:329–343.
[2] Lin C-C, Hsieh W-Y, Chou C-S, Liaw S-F.
Cardiopulmonary exercise testing in obstructive
sleep apnea syndrome. Respir Physiol Neurobiol
2006;150:27–34.
[3] Vanhecke TE, Franklin BA, Zalesin KC, Sangal
RB, deJong AT, Agrawal V, et al.
Cardiorespiratory fitness and obstructive sleep
apnea syndrome in morbidly obese patients. Chest
2008;134:539–545.
[4] Grote L, Hedner J, Peter JH. The heart rate
response to exercise is blunted in patients with
sleep-related breathing disorder. Cardiology 2004;102:93–99.
[5] Mansukhani MP, Allison TG, Lopez-Jimenez F,
Somers VK, Caples SM. Functional aerobic
capacity in patients with sleep-disordered breathing. Am J Cardiol 2013;111:1650–1654.
[6] Beitler JR, Awad KM, Bakker JP, Edwards BA,
DeYoung P, Djonlagic I, et al. Obstructive sleep
apnea is associated with impaired exercise
capacity: a cross-sectional study. J Clin Sleep Med 2014;10:1199–1204.
[7] Kaleth AS, Chittenden TW, Hawkins BJ, Hargens
TA, Guill SG, Zedalis D, et al. Unique
cardiopulmonary exercise test responses in
overweight middle-aged adults with obstructive sleep apnea. Sleep Med 2007;8:160–168.
[8] Tryfon S, Stanopoulos I, Dascalopoulou E,
Argyropoulou P, Bouros D, Mavrofridis E. Sleep apnea syndrome and diastolic blood pressure
elevation during exercise. Respiration
2004;71:499–504.
[9] [Alonso-Fernández A, García-Río F, Arias MA,
Mediano O, Pino JM, Martínez I, et al.
Obstructive sleep apnoea-hypoapnoea syndrome
reversibly depresses cardiac response to exercise.
Eur Heart J 2006;27:207–215.
[10] Rizzi CF, Cintra F, Mello-Fujita L, Rios LF,
Mendonca ET, Feres MC, et al. Does Obstructive
Sleep Apnea Impair the Cardiopulmonary
Response to Exercise? Sleep 2013;36:547–553.
[11] Hargens TA, Guill SG, Zedalis D, Gregg JM, Nickols-Richardson SM, Herbert WG.
Attenuated heart rate recovery following exercise
testing in overweight young men with untreated
obstructive sleep apnea. Sleep 2008;31:104–110.
[12] Maeder MT, Ammann P, Münzer T, Schoch OD,
Korte W, Hürny C, et al. Continuous positive
airway pressure improves exercise capacity and
heart rate recovery in obstructive sleep apnea. Int
J Cardiol 2009;132:75–83.
[13] Serés L, López-Ayerbe J, Coll R, Rodríguez O,
Manresa JM, Marrugat J, et al. Cardiopulmonary
function and exercise capacity in patients with
morbid obesity. Rev Esp Cardiol 2003;56:594–
600.
[14] Newman AB. Progression and regression of sleep-disordered breathing with changes in weight. Arch Intern Med 2005;165:2408-2413.
[15] Shephard RJ. Exercise and training in women,
Part I: Influence of gender on exercise and
training responses. Can J Appl Physiol
2000;25:19–34.
[16] Cole CR, Blackstone EH, Pashkow FJ, Snader CE, Lauer MS.Heart-rate recovery immediately
after exercise as a predictor of mortality. N Engl J
Med 1999;341:1351–1357.
[17] Maeder MT, Münzer T, Rickli H, Schoch OD,
Korte W, Hürny C, et al. Association between
heart rate recovery and severity of obstructive
sleep apnea syndrome. Sleep Med 2008;9:753–
761.
[18] Kligfield P, McCormick A, Chai A, Jacobson A,
Feuerstadt P, Hao SC. Effect of age and gender on
heart rate recovery after submaximal exercise
during cardiac rehabilitation in patients with
angina pectoris, recent acute myocardial
infarction, or coronary bypass surgery. Am J
Cardiol 2003;92:600–603.
[19] Sleep-related breathing disorders in adults:
recommendations for syndrome definition and
measurement techniques in clinical research. The
Report of an American Academy of Sleep
Medicine Task Force. Sleep 1999;22:667–689.
[20] American Thoracic Society, American College of
Chest Physicians. ATS/ACCP Statement on
cardiopulmonary exercise testing. Am J Respir
Crit Care Med 2003;167:211–277.
[21] Rizzi CF, Cintra F, Risso T, Pulz C, Tufik S, de
Paola A, et al. Exercise capacity and obstructive
sleep apnea in lean subjects. Chest 2010;137:109–
14.
[22] Butner KL, Hargens TA, Kaleth AS, Miller LE,
Zedalis D, Herbert WG. Association of
obstructive sleep apnea severity with exercise
capacity and health-related quality of life. N Am J Med Sci 2013;5:362–366.
[23] Ucok K, Aycicek A, Sezer M, Genc A, Akkaya M,
Caglar V, et al. Aerobic and anaerobic exercise
capacities in obstructive sleep apnea and
associations with subcutaneous fat distributions.
Lung 2009;187:29–36.
[24] Cintra F, Poyares D, Rizzi CF, Risso TT, Skomro
R, Montuori E, et al. Cardiorespiratory response
to exercise in men and women with obstructive
sleep apnea. Sleep Med 2009;10:368–373.
[25] Sauleda J, García-Palmer FJ, Tarraga S, Maimó
A, Palou A, Agustí AGN. Skeletal muscle changes
in patients with obstructive sleep apnoea
syndrome. Respir Med 2003;97:804–810.
[26] Grote L, Kraiczi H, Hedner J. Reduced alpha-and beta(2)-adrenergic vascular response in patients with obstructive sleep apnea. Am J Respir Crit Care Med 2000;162:1480–1487.
Table 1. Baseline characteristics of the study population
| Variables | Value | |
| (n=24) | ||
| Demographics | ||
| Age, years | 36.17 ± 9.20 | |
| Weight, kg | 120.17 ± 15.74 | |
| Height, m | 1.63 ± 5.92 | |
| BMI, kg/m2 | 45.20 ± 5.41 | |
| ESS | 6.25 ± 4.94 | |
| Sleep registers | ||
| AHI, h-1 | 21.59 ± 23.41 | |
| AI, h-1 | 7.80 ± 17.86 | |
| DI, h-1 | 22.97 ± 23.61 | |
| Lowest SpO2, % | 79.13 ± 7.40 | |
| Time spent <90%, % | 9.83 ± 16.40 | |
| Peak exercise | ||
| Breathing reserve, % | 46.58 ± 10.77 | |
| Breathing frequency, min-1 | 35.20 ± 7.18 | |
| Tidal volume, ml | 1821.21 ± 467.85 | |
| V’E/V’CO2 slope | 28.94 ± 4.48 | |
| HR slope, 1.ml-1.kg-1 | 9.48 ± 1.76 | |
| V’O2, ml.min-1.kg-1 | 15.58 ± 2.44 | |
| Work, watts | 97.72 ± 17.23 | |
| HR, beats.min-1 | 142 ± 19.14 | |
| Predicted maximal HR, % | 74.91 ± 10.70 | |
| HR recovery at min 0.5, beats.min-1 | 130.54 ± 18.08 | |
| %HR recovery at min 0.5, % | 8.08 ± 3.76 | |
| HR recovery at min 1, beats.min-1 | 122.82 ± 23.74 | |
| %HR recovery at min 1, % | 14.14 ± 9.17 | |
| HR recovery at min 1.5, beats.min-1 | 123.57± 16.62 | |
| %HR recovery at min 1.5, % | 14.83 ± 4.11 |
Data are presented as mean ± SD
%HR: per cent decrease of maximal heart rate; AHI: apnoea-hypopnea index; AI: apnoea index; BMI: body mass index; DI: oxygen
desaturation index; ESS: Epworth somnolence scale; HR: heart rate; SpO2: pulse oximetry oxygen saturation; VE: minute ventilation;
V’CO2: carbon dioxide output; V’O2 oxygen uptake
Table 2. Comparison between mild and moderate-to-severe obstructive sleep apnoea
| Variable | AHI <15 | AHI ≥15 | 95% CI |
|---|---|---|---|
| (n = 14) | (n = 10) | ||
| Age, years | 35.57 ± 11.37 | 37.00 ± 5.31 | -9.48 – 6.62 |
| Weight, kg | 117.07 ± 17.70 | 124.50 ± 12.03 | -20.85 – 6.00 |
| BMI, kg.m-2 | 44.30 ± 6.02 | 46.47 ± 4.40 | -6.82 – 2.48 |
| ESS | 7 ± 4,50 | 5.50 ± 5.53 | -3.91 – 6.91 |
| AHI, h-1 | 7.26 ± 2.20 | 41.65 ± 25.06 | -52.32 – -16.44** |
| AI, h-1 | 0,83 ± 1.27 | 17.56 ± 25.14 | -30.55 – -2.89* |
| DI, h-1 | 8.8 ± 4.82 | 42.74 ± 25.43 | -52.19 – -15.61** |
| Lowest SpO2, % | 82.70 ± 4.92 | 74.50 ± 7.71 | 2.71 – 13.68** |
| Time spent <90%, % | 2.14 ± 20.60 | 4.32 ± 20.96 | -33.54 – -3.37* |
| Breathing reserve, % | 47.86 ± 12.06 | 44.80 ± 8.98 | -6.31 – 12.42 |
| Breathing frequency, min-1 | 34.07 ± 5.40 | 36.80 ± 9.21 | -8.92 – 3.46 |
| Tidal volume, l | 1837 ± 526.47 | 1799.10 ± 397.59 | -372.51 – 448.31 |
| V’E/V’CO2 slope | 28.63 ± 4.20 | 29.35 ± 5.09 | -4.65 – 3.21 |
| HR slope , 1.ml-1.kg-1 | 9.92 ± 1.96 | 8.90 ± 1.36 | -0.47 – 2.51 |
| V’O2, ml.min-1.kg-1 | 15.19 ± 2.50 | 16.11 ± 2.37 | -3.03 – 1.17 |
| Peak workload, watts | 94.37 ± 18.40 | 106.67 ± 11.54 | -38.51 – 13.93 |
| HR, beats.min-1 | 142.86 ± 22.04 | 139.67 ± 11.06 | -27.62 – 34.02 |
| Predicted maximal HR, % | 75.75 ± 12.65 | 72.67 ± 1.54 | -14.02 – 20.18 |
| HR recovery at min 0.5, beats.min-1 | 129.88 ± 20.71 | 132.33 ± 11.24 | -31.59 – 26.67 |
| %HR recovery at min 0.5, % | 9.13 ± 3.91 | 5.28 ± 1.00 | -1.48 – 9.17 |
| HR recovery at min 1, beats.min-1 | 121.12 ± 27.30 | 127.33 ± 12.90 | -44.24 – 31.83 |
| %HR recovery at min 1, % | 16.09 ± 10.11 | 8.93 ± 2.45 | -6.63 – 20.93 |
| HR recovery at min 1.5, beats.min-1 | 123.17 ± 16.17 | 124.34 ± 14.19 | -27.29 – 24.96 |
| %HR recovery at min 1.5, % | 16.69 ± 3.23 | 11.14 ± 3.26 | -0.82 – 11.89 |
Data are presented as mean ± SD. A t-test was used for comparisons
%HR: per cent decrease of maximal heart rate; AHI: apnoea-hypopnea index; AI: apnoea index; BMI: body mass index; DI: oxygen desaturation
index; ESS: Epworth somnolence scale; HR: heart rate; SpO2: pulse oximetry oxygen saturation; VE: minute ventilation; V’CO2: carbon dioxide
output; V’O2 oxygen uptake
*p<.05, **p<.01
Table 3. Correlations between sleep registers and cardiopulmonary exercise testing
| Variable AHI r p Predicted maximal HR, % -0.067 0.844 %HR recovery at min 0.5, % -0.468 0.147 %HR recovery at min 1, % -0.406 0.216 %HR recovery at min 1.5, % -0.655 0.056 | DI r -0.062 -0.414 -0.381 -0.679 | P 0.856 0.206 0.248 0.044 |
|---|---|---|
| %HR: per cent decrease of maximal heart rate; AHI: apnoea-hypopnea index; DI: oxygen desaturation index |