- Email: [email protected]
Cardiovascular risk factors in men and women with obstructive sleep apnoea syndrome
Respiratory Medicine (2010) 104, 1063e1068
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/rmed
Cardiovascular risk factors in men and women with obstructive sleep apnoea syndrome Jean-Pierre Laaban a, Line Mounier b, Olivier Roque d’Orbcastel c, ´ Cornette e, Dan Veale b,*, Jacques Blacher a, Boris Melloni d, Andre Jean-Franc ¸ois Muir f, Edmond Chailleux g, ANTADIR Working group ‘‘CV risk in OSAS’’h, CMTS, ANTADIR a
Hoˆtel-Dieu de Paris APHP, Universite´ Paris Descartes, France ANTADIR, Paris, France c Clinique Les Rieux, Nyons, France d CHU Limoges, France e CHU Nancy, France f CHU Rouen, France g CHU Nantes, France b
Received 29 October 2009; accepted 20 February 2010 Available online 19 March 2010
KEYWORDS Sleep apnoea; Cardiovascular risk; Diabetes; Hypertension; Questionnaire; Continuous positive airway pressure
Summary We wished to evaluate the prevalence of cardiovascular (CV) risk factors in patients with obstructive sleep apnoea syndrome (OSAS) before initiation of continuous positive airway pressure (CPAP), and without any declared or diagnosed pre-existing CV disorder. We wanted to compare the prevalence of these CV risk factors between men and women in an observational study. A questionnaire concerning CV risk factors was submitted to the patients, by a respiratory home-care technician at the time of installation of the CPAP treatment. Patients: The study population consisted of 1117 patients; 834 men, 283 women. Results: The prevalence of arterial hypertension (HT), diabetes, obesity, active smoking, hyperlipidemia and family history of coronary heart disease was 54.1%, 22.8%, 65.8%, 18.3%, 33.8% and 20%, respectively.
* Corresponding author. 66 Bld St Michel, 75006 Paris, France. Tel.: þ33 156 814 065; fax þ33 156 814 061. E-mail addresses: [email protected] (J.-P. Laaban), [email protected] (L. Mounier), [email protected] (O. Roque d’Orbcastel), [email protected] (D. Veale), [email protected] (J. Blacher), [email protected] (B. Melloni), a.cornette@ assistance.du.grand.est (A. Cornette), [email protected] (J.-F. Muir), [email protected] (E. Chailleux). h ANTADIR working group: Dr Franc ¸oise Alluin: Fouquie `res le `s Lens, Pr Jacques Blacher, Paris, Pr Edmond Chailleux; Nantes, Dr Andre ´ Cornette; Nancy, Dr Marie-Odile Debaud; Rochefort, Dr Marie-Pierre Humeau-Chapuis; Nantes, Pr Jean-Pierre Laaban; Paris, Pr Franc ¸ois Lebargy; Reims, Pr Boris Melloni; Limoges, Dr Miloud Merati; Dijon, Line Mounier; Paris, Pr Jean-Franc ¸ois Muir; Rouen, Dr Patrick Orange; Rouen, Dr Jean-Noe ¨l Pre ´vost; Caen, Dr Olivier Roque d’Orbcastel, NYONS, Dr Ziad Rida; La Re ´union, Dr Daniel Veale; Paris. 0954-6111/$ - see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2010.02.021
1064
J.-P. Laaban et al. Women had significantly more HT (62.1 vs 51.4%), diabetes (29.9 vs 20.4%), obesity (77 vs 62%) and family history of coronary disease (25.1 vs 18.2%). The prevalence of active smoking was significantly higher in men (20.4 vs 12%). The prevalence of hyperlipidemia was not different between men and women (34.5 vs 31.8%). Stepwise logistic regression showed that HT and diabetes were both independently associated with BMI and age, while diabetes and not HT was independently associated with female gender. The prevalence of classical CV risk factors was very high in this population with OSAS requiring CPAP, especially in women. There is thus a very elevated CV risk level independent of that directly related to OSAS. It is important to screen for and treat classical CV risk factors in this population. ª 2010 Elsevier Ltd. All rights reserved.
Introduction Cardiovascular (CV) morbidity and mortality are increased in patients with obstructive sleep apnoea syndrome (OSAS).1,2 Cross-sectional studies have shown that OSAS is an independent risk factor for coronary artery disease (CAD) and stroke.3,4 However the relative risk of CAD associated with OSAS was weak and did not increase with increasing severity of OSAS.3 Several observational studies have shown that the incidence of CV complications is increased in untreated OSAS and decreased with continuous positive airway pressure (CPAP).5,6 However, in these studies, the control group consisted of patients refusing CPAP or not complying with it who were also likely to have poor compliance with CV risk reduction treatments. Therefore it is not clear how much of the increased CV risk in apnoeic patients is directly attributable to the OSAS and how much is related to the concomitant presence of other classical CV risk factors. Furthermore studies that have evaluated CV risk factors in OSAS have certain limitations. In most studies, patients with pre-existing CV disorders were not excluded, thereby leading to an overestimation of CV risk factors. The number of patients included in these studies is very variable.7 The proportion of patients with severe OSAS is small in many studies, even in those with large cohorts.3,8 The evaluation of CV risk factors is often incomplete, especially as regards the presence of a family history of CAD which is a major risk factor. Finally, the evaluation of CV risk in OSAS patients has generally been studied in men only; most studies had few or no women.5e7,9,10 Thus we set out to study the prevalence of major CV risk factors in a large population of men and women with OSAS, requiring CPAP, and free of pre-existing known CV disease. We wished to compare the prevalence of these risk factors between men and women.
Patients and methods Consecutive patients with OSAS were included on CPAP initiation, over a period of eighteen months, in 10 homecare associations in the federated ANTADIR network (Association Nationale pour les Traitements A Domicile, les Innovations et la Recherche).11 The diagnosis of OSAS was established by full polysomnography or by nocturnal respiratory polygraphy.
This was an observational study. A questionnaire was given to the patient, at home, by paramedical personnel from the homecare association, on the day that the CPAP treatment was initiated. The questionnaire consisted in 38 questions concerning CV risk factors, previous CV events and medications usually used. If the patient had difficulties the paramedical helped him to fill out the questionnaire. The diagnosis of hypertension (HT) and diabetes depended on the patient replying that he had HT and/or diabetes or by the identification of treatment for the condition in the list of medicines taken. Beta-blockers and ACE-inhibitors were regarded as anti-hypertensive drugs and not as treatment for heart failure. Angiotensine IIinhibitors, Ca-channel blockers and diuretics were considered as anti-hypertensive drugs. Obesity was defined by a body mass index (BMI) 30 kg/m2, based on weight and height reported by the patient. Patients were considered as current smokers in the case of active smoking within the year. The diagnosis of hyperlipidemia was based on the patients’ report of the diagnosis and by the identification of treatment for the condition in the list of medications. The presence of a family history of CAD was defined by reported angina, or myocardial infarction or a history of coronary artery bypass operation, or a history of coronary angioplasty before the age of 60 years in any first degree relative (parent or sibling, or child). We excluded from analysis patients with pre-existing coronary artery disease, cerebrovascular disease or peripheral artery disease. Coronary artery disease was defined by reported angina, or myocardial infarction or a history of coronary artery bypass operation, or a history of coronary angioplasty. Cerebrovascular disease was defined as reported stroke or surgery or angioplasty of a carotid artery. Peripheral artery disease was defined by any past history of surgery or angioplasty of a lower limb artery. We have therefore excluded 338 patients in whom the questionnaire revealed a previous cardiovascular disease. ANTADIR is registered with the National Committee of Civil Liberties (CNIL). This study was approved by the Scientific Committee of ANTADIR. All numerical variables were expressed as mean SD. The c2 test was used for a comparison of categorical variables between two groups. For the comparison of numerical variables between two groups, the Student t test was used. A p value of < 0.05 was considered as statistically significant. In the case of finding a difference in the prevalence of a CV risk factor between men and women, and to determine if this
Cardiovascular risk factors in men and women difference was due independently to gender, we carried out a stepwise logistic regression. In this model the CV risk factor concerned was the dependent variable and gender was one of the tested independent variables.
1065 gender (p Z 0.021). Female gender was therefore independently and positively associated with diabetes, with and odds ratio of 1.56 (95% CI:1.07e2.27).
Discussion Results The study population consisted of 1117 patients, 834 men and 283 women, with OSAS requiring CPAP and free of a known CV disease. Age, apnoeaehypopnoea index (AHI) and parameters of nocturnal desaturation did not significantly differ between men and women (Table 1). The prevalence of severe OSAS (AHI 30), moderate OSAS (AHI: 15e29.9) and mild OSAS (AHI: 5e14.9) was respectively 85.6%, 12.2% and 2.2%. The prevalences of CV risk factors are shown in Table 2. The prevalences of HT, diabetes, obesity and family history of CAD were respectively 54.1%, 22.8%, 65.8%, 20% and were significantly higher in women than in men. The prevalence of current smoking was 18.3% and was significantly higher in men than in women. The prevalence of hyperlipidemia was 33.8% and did not differ between men and women. The percentages of patients with at least two or three of the following CV risk factors, i.e. hypertension, diabetes or obesity were significantly higher in women than in men (Table 3). In the female population, there were no differences in the levels of different CV risk factors between those with severe OSAS, and those with moderate to mild OSAS (Table 4). In the male population, the prevalence of HT and obesity was higher in those with severe OSAS than in those with moderate to mild OSAS (Table 4). A stepwise logistic regression was carried out to examine whether the increased prevalence of HT or diabetes was independently associated with female gender. In the first model the dependent variable was HT, and the tested variables were sex, age, BMI, AHI, smoking status and diabetes. The variables associated independently with HT were BMI (p < 0.001), age (p < 0.001) and diabetes (p < 0.001), but gender was not associated independently with HT. In the second model, the dependent variable was diabetes and the tested variables were sex, age, BMI, AHI, smoking status and HT. The variables associated independently with diabetes were BMI (p < 0.001), HT (p < 0.001), age (p < 0.001), smoking status (p Z 0.003) and female
This observational study shows that the prevalence of various CV risk factors was very high in a group of 1117 patents with OSAS, requiring CPAP treatment while not presenting a known CV illness. The prevalence of HT, diabetes, obesity and family history of CAD was greater in women than in men, while age and the severity of OSAS were not. This was an observational study of a large series of patients in a multicenter home-care system for management of patients with CPAP. An auto-questionnaire was used with assistance given to the patient when required. Thus there was no specific medical visit for the study. The patients were interviewed at the time of initiation of CPAP therapy, at home, and thus there was a large recruitment over a short period of time. In our study the prevalence of different CV risk factors, apart from current smoking was greater than in the general population: particularly, diabetes, HT and obesity.12e14 The prevalence of different CV risk factors observed in our study can be compared with data from the literature as regards the CV profile of patients with OSAS. In our study, the prevalence of diabetes was very elevated at 22.8% in patients with OSAS requiring CPAP. Similar figures have been noted in the literature6,9 while other studies have shown a lower prevalence of diabetes in OSAS3,5 which could be related to the lower age of the patients. In our study, the prevalence of HT was high (54.1%), and comparable to that usually found in severe OSAS.6,15e19 The prevalence of obesity was very high in our study (65.8%), but the prevalence of obesity in OSAS is related to the level of the AHI.16 The prevalence of hyperlipidemia was elevated in our study (33.8%) even though the patients did not have known CV disease. Twenty percent of the patients in our study had a family history of CAD. This important CV risk factor20 has not been mentioned in other published studies evaluating CV risk related to OSAS.5,6,21 Furthermore it has been reported that patients with OSAS had a greater family history of premature coronary related death, independently of the presence of CAD associated with the OSAS.22 The prevalence of current smoking in our
Table 1 Age, body mass index (BMI) and nocturnal respiratory data for the overall population with comparison between men and women.
Age (years) BMI (kg/m2) BMI 40 kg/m2 (% pat) AHI (/h) Minimal nocturnal SaO2 (%) Mean nocturnal SaO2 (%) % RT SaO2 < 90 %
All patients (n Z 1117)
Men (n Z 834)
Women (n Z 283)
p
57.7 12 33.6 7.3 15.8 48.6 21.4 70.8 14.6 91.3 4 24.3 25.7
57.7 12 32.4 6.2 10.2 49 20.9 71 14.6 91.3 4.1 23.7 25.2
57.7 12.1 36.8 9.1 32.5 47.2 22.9 70.2 14.8 91.3 3.9 26 27.1
NS <0.001 <0.001 NS NS NS NS
AHI, apnoea-hypopnoea index; RT, recording time.
1066 Table 2 women.
J.-P. Laaban et al. Prevalence of different cardiovascular risk factors for the overall population with comparison between men and
Hypertension (% pat) Diabetes (% pat) Obesit (BMI 30 kg/m2) (% pat) Current smoker (% pat) Current or ex-smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat)
All patients (n Z 1117)
Men (n Z 834)
Women (n Z 283)
p
54.1 22.8 65.8 18.3 66.2 33.8 20.0
51.4 20.4 62.0 20.4 77.0 34.5 18.2
62.1 29.9 77.0 12.0 34.3 31.8 25.1
<0.005 <0.002 <0.001 <0.002 <0.001 NS <0.005
BMI, body mass index; CAD, coronary artery disease.
study was 18.3%, while other studies that evaluated CV risk related to OSAS reported comparable or higher or lower figures.5e7,16,17,19 One important result of the current study is that the prevalence of most of the CV risk factors was greater in women than in men. In our study the prevalence of diabetes was greater in women than in men. These results are not in accordance with the data reported in the general population in whom the prevalence of diabetes is higher in men than in women.13 This increased prevalence of diabetes in women may possibly be explained by the greater prevalence of obesity in women, in particular massive obesity. However female gender was in our study an independent predictive factor for the prevalence of diabetes. There are no other data in the literature concerning the prevalence of diabetes in women with severe OSAS, since most studies did not include women5,9 or did not compare the prevalence of diabetes between men and women with OSAS.6 We have no explanation accounting for our finding of an independent association between female gender and diabetes which should be confirmed by further studies. The prevalence of HT was higher in women than in men in our study. This is surprising since male gender is a well known risk factor for HT. This increased prevalence of HT in women was not explained by greater age, nor by more severe OSAS, but most probably by a greater prevalence of obesity, and in particular massive obesity in women in our study. The prevalence of obesity was greater in women than in men in this study, and massive obesity was three times more frequent in women. It has been shown that the level of obesity necessary to cause OSAS is higher in women than in men, aged under 65 years.21 The prevalence of hyperlipidemia was not different between men and women in
this study, whilst in the general population, hyperlipidemia is more frequent in men.12,13 In this study women had a greater level of family history of coronary disease without any evident explanation. We have noted a greater level of smoking in men than in women, as is common in the general population,12 and confirmed in patients with OSAS.21 In this study the prevalence of CV risk factors was not significantly influenced by the severity of OSAS in women, whilst in men the prevalence of HT and obesity was much higher in severe OSAS than in mild-moderate OSAS. These data suggest that the proportion of cardiovascular risk directly attributable to OSAS is greater in men than in women. The prevalence of CV risk factors, although high, may be underestimated in this study. We only took into account patients with previously known diabetes, and a certain number of apnoeic patients may have had occult diabetes.23 Studies using ambulatory control of blood pressure over 24 h showed that HT is often underdiagnosed by standard clinical methods in patients with OSAS.24 We only took into account patients with hyperlipidemia treated by lipid lowering agents. In studies with systematic measurement of blood lipid levels the prevalence of hyperlipidemia was higher.6,7 The high levels of CV risk factors in this study could be explained by the fact that OSAS may favour the development of certain CV risk factors, especially HT. Transverse and longitudinal studies have shown that OSAS is an independent risk factor for HT,8,15e17,25 but the relative risk of HT independently related to OSAS is not very high.26,27 We have found an elevated level of CV risk factors in this population with OSAS. However, this can be considered as a selected population, given that recruited patients have
Table 3 Number of cardiovascular (CV) risk factors per patient and association of CV risk factors for the overall population with comparison between men and women.
No of CV risk factors per patient 3 CV risk factors (3e6) per patient (% pat) At least hypertension and obesity (% pat) At least diabetes and obesity (% pat) At least hypertension and diabetes (% pat) At least hypertension, diabetes and obesity (% pat)
All patients (n Z 1117)
Men (n Z 834)
Women (n Z 283)
p
2.1 1.3 39.8 41.5 18.8 16.7 14.4
2.1 1.3 37.3 38.4 16.1 14.5 12.1
2.4 1.3 47 50.9 26.9 23 21.2
<0.001 <0.005 <0.01 <0.001 <0.001 <0.001
Cardiovascular risk factors in men and women
1067
Table 4 Comparison of the prevalence of different cardiovascular risk factors between patients with mild-moderate OSAS(AHI Z 5e29.9) and patients with severe OSAS(AHI 30) for the overall population with comparison between men and women. All patients
AHI Z 5e29.9 (n Z 161)
AHI 30 (n Z 956)
p
Hypertension (% pat) Diabetes (% pat) Obesity (BMI 30 kg/m2) (% pat) Current smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat) Age (years)
45.0 16.9 50.3 18.0 32.3 20.5 56.6 12.2
55.7 23.8 68.4 18.3 34.1 19.9 57.9 12.0
<0.001 NS <0.001 NS NS NS NS
Women
AHI Z 5e29.9 (n Z 52)
AH 30 (n Z 231)
p
Hypertension (% pat) Diabetes (% pat) Obesity (BMI30 kg/m2) (% pat) Current smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat) Age (years)
60.8 23.5 69.2 13.5 25.0 23.1 58.0 11.7
62.4 31.3 78.8 11.7 33.3 25.5 57.7 12.2
NS NS NS NS NS NS NS
Men
AHI Z 5e29.9 (n Z 109)
AHI 30 (n Z 725)
p
Hypertension (% pat) Diabetes (% pat) Obesity (BMI 30 kg/m2) (% pat) Current smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat) Age (years)
37.6 13.8 41.3 20.2 35.8 19.3 56.0 12.3
53.5 21.4 65.1 20.4 34.3 18.1 58.0 11.9
<0.001 NS <0.001 NS NS NS NS
BMI, body mass index; CAD, coronary artery disease.
been screened by the process of referral to a sleep center and subsequently recommended for treatment. Thus the level of CV risk in a population with suspected OSAS may well be different and likewise in a population with OSAS but not recommended for treatment. Indeed patients with OSAS but with no other concomitant health problems are probably less likely to visit a doctor and may thus be underrepresented in our population. Likewise, women tend to seek medical help more than men, but there may be a counteracting bias against diagnosing OSAS in women. Indeed men without concomitant disease may be induced to visit a doctor at the behest of their partner more than the inverse situation. CPAP treatment is reimbursed in France when the patient has over 30 respiratory events per hour or over 10 respiratory related micro-arousals per hour, with symptoms. Thus in this series the patients could be considered to be more severely affected than in North American. Thus a high proportion of patients can be classed as having severe sleep apnoea, but in contrast we have excluded patients with known CV disease. It could be argued that there is a bias to treat OSAS patients with CV risk factors with CPAP, but no figures exist as to the proportion of known OSAS patients untreated nor as to the relative proportions of patients having other treatment options such as surgery or mandibular advancement prosthesis. Thus this selected population that is physician referred and with severe OSAS, has a high incidence of CV risk
factors. Our findings sustain the idea that such patients merit screening for CV risk factors and health advice aimed to reduce such risks. In order to diminish CV morbidity and mortality it is essential to optimally treat CV risk factors and this is probably as important as treating effectively the sleep apnoea.
Conflict of interest statement The authors have no conflicts of interest.
References 1. McNicholas WT, Bonsignore MRThe Management Committee of EU COST ACTION B26. Sleep apnoea as an independent risk factor for cardiovascular disease: current evidence, basic mechanisms and research priorities. Eur Respir J 2007;29: 156e78. 2. Bradley TD. Obstructive sleep apnoea and its cardiovascular consequences. Lancet 2009;373:82e93. 3. Shahar E, Whitney CW, Redline S, et al. Sleep-disordered breathing and cardiovascular disease. Cross-sectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med 2001; 163:19e25. 4. Arzt M, Young T, Finn L, et al. Association of sleep-disordered breathing and the occurrence of stroke. Am J Respir Crit Care Med 2005;172:1447e51. 5. Marin JM, Carrizo SJ, Vicente E, et al. Long-term cardiovascular outcomes in men with obstructive sleep apnoeaehypopnoea with or without treatment with continuous positive
1068
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
airway pressure: an observational study. Lancet 2005;365: 1046e53. Buchner NJ, Sanner BM, Borgel J, et al. Continuous positive airway pressure treatment of mild to moderate obstructive sleep apnea reduces cardiovascular risk. Am J Respir Crit Care Med 2007;176:1274e80. Kiely JL, McNicholas WT. Cardiovascular risk factors in patients with obstructive sleep apnoea syndrome. Eur Respir J 2000;16: 128e33. Bixler EO, Vgontzas AN, Lin HM, et al. Association of hypertension and sleep-disordered breathing. Arch Intern Med 2000; 160:2289e95. Meslier N, Gagnadoux F, Giraud P, et al. Impaired glucoseinsulin metabolism in males with obstructive sleep apnoea syndrome. Eur Respir J 2003;22:156e60. Peker Y, Carlson J, Hedner J. Increased incidence of coronary artery disease in sleep apnoea: a long term follow-up. Eur Respir J 2006;28:596e602. Stuart M, Weinrich M. Integrated health system for chronic disease management. Lessons learned from France. Chest 2004;125:695e703. ¨ unpuu S, et al. Effect of potentially Yusuf S, Hawken S, O modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004;364:937e52. Rudnichi A, Safar M, Asmar R, et al. Prevalence of cardiovascular risk factors in a French population. J Hypertens 1998; 16(Suppl. 1):S85e90. Ricordeau P, Weill A, Vallier N, et al. The prevalence and cost of diabetes in metropolitan France: what trends between 1998 and 2000? Diabetes Metab 2003;29:497e504. Lavie P, Herer P, Hoffstein V. Obstructive sleep apnoea syndrome as a risk factor for hypertension: population study. BMJ 2000;320:479e82. for the Sleep Heart Health Study Nieto FJ, Young TB, Lind BK, et al. Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. JAMA 2000;283:1829e36.
J.-P. Laaban et al. 17. Young T, Peppard P, Palta M, et al. Population-based study of sleep-disordered breathing as a risk factor for hypertension. Arch Intern Med 1997;157:1746e52. 18. Grote L, Ploch T, Heitmann J, et al. Sleep-related breathing disorder is an independent risk factor for systemic hypertension. Am J Respir Crit Care Med 1999;160:1875e82. 19. Bo ¨rgel J, Sanner BM, Bittlinsky A, et al. Obstructive sleep apnoea and its therapy influence high-density lipoprotein cholesterol serum levels. Eur Respir J 2006;27:121e7. 20. LLoyd-Jones DM, Nam BO, D’Agostino RB, et al. Parental cardiovascular disease as a risk factor for cardiovascular disease in middle-aged adults. A prospective study of parents and offspring. JAMA 2004;291:2204e11. 21. Newman AB, Nieto FJ, Guidry U, et al. Relation of sleepdisordered breathing to cardiovascular disease risk factors. The Sleep Heart Health Study. Am J Epidemiol 2001;154:50e9. 22. Gami AS, Rader S, Svatikova A, et al. Familial premature coronary artery disease mortality and obstructive sleep apnea. Chest 2007;131:118e21. 23. Seicean S, Kirchner HL, Gottlieb DJ, et al. Sleep-disorderd breathing and impaired glucose metabolism in normal-weight and overweight/obese individuals. The Sleep Heart Health Study. Diabetes Care 2008;31:1001e6. 24. Baguet JP, Hammer L, Le ´vy P, et al. Night-time and diastolic hypertension are common and underestimated conditions in newly diagnosed apnoeic patients. J Hypertens 2005;23: 521e7. 25. Peppard PE, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:1378e84. 26. Bazzano LA, Khan Z, Reynolds K, et al. Effect of nocturnal nasal continuous positive airway pressure on blood pressure in obstructive sleep apnea. Hypertension 2007;50:417e23. 27. Haentjens P, Van Meerhaeghe A, Moscariello A, et al. The impact of continuous positive airway pressure on blood pressure in patients with obstructive sleep apnea syndrome. Evidence from a meta-analysis of placebo-controlled randomized trials. Arch Intern Med 2007;167:757e65.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/rmed
Cardiovascular risk factors in men and women with obstructive sleep apnoea syndrome Jean-Pierre Laaban a, Line Mounier b, Olivier Roque d’Orbcastel c, ´ Cornette e, Dan Veale b,*, Jacques Blacher a, Boris Melloni d, Andre Jean-Franc ¸ois Muir f, Edmond Chailleux g, ANTADIR Working group ‘‘CV risk in OSAS’’h, CMTS, ANTADIR a
Hoˆtel-Dieu de Paris APHP, Universite´ Paris Descartes, France ANTADIR, Paris, France c Clinique Les Rieux, Nyons, France d CHU Limoges, France e CHU Nancy, France f CHU Rouen, France g CHU Nantes, France b
Received 29 October 2009; accepted 20 February 2010 Available online 19 March 2010
KEYWORDS Sleep apnoea; Cardiovascular risk; Diabetes; Hypertension; Questionnaire; Continuous positive airway pressure
Summary We wished to evaluate the prevalence of cardiovascular (CV) risk factors in patients with obstructive sleep apnoea syndrome (OSAS) before initiation of continuous positive airway pressure (CPAP), and without any declared or diagnosed pre-existing CV disorder. We wanted to compare the prevalence of these CV risk factors between men and women in an observational study. A questionnaire concerning CV risk factors was submitted to the patients, by a respiratory home-care technician at the time of installation of the CPAP treatment. Patients: The study population consisted of 1117 patients; 834 men, 283 women. Results: The prevalence of arterial hypertension (HT), diabetes, obesity, active smoking, hyperlipidemia and family history of coronary heart disease was 54.1%, 22.8%, 65.8%, 18.3%, 33.8% and 20%, respectively.
* Corresponding author. 66 Bld St Michel, 75006 Paris, France. Tel.: þ33 156 814 065; fax þ33 156 814 061. E-mail addresses: [email protected] (J.-P. Laaban), [email protected] (L. Mounier), [email protected] (O. Roque d’Orbcastel), [email protected] (D. Veale), [email protected] (J. Blacher), [email protected] (B. Melloni), a.cornette@ assistance.du.grand.est (A. Cornette), [email protected] (J.-F. Muir), [email protected] (E. Chailleux). h ANTADIR working group: Dr Franc ¸oise Alluin: Fouquie `res le `s Lens, Pr Jacques Blacher, Paris, Pr Edmond Chailleux; Nantes, Dr Andre ´ Cornette; Nancy, Dr Marie-Odile Debaud; Rochefort, Dr Marie-Pierre Humeau-Chapuis; Nantes, Pr Jean-Pierre Laaban; Paris, Pr Franc ¸ois Lebargy; Reims, Pr Boris Melloni; Limoges, Dr Miloud Merati; Dijon, Line Mounier; Paris, Pr Jean-Franc ¸ois Muir; Rouen, Dr Patrick Orange; Rouen, Dr Jean-Noe ¨l Pre ´vost; Caen, Dr Olivier Roque d’Orbcastel, NYONS, Dr Ziad Rida; La Re ´union, Dr Daniel Veale; Paris. 0954-6111/$ - see front matter ª 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2010.02.021
1064
J.-P. Laaban et al. Women had significantly more HT (62.1 vs 51.4%), diabetes (29.9 vs 20.4%), obesity (77 vs 62%) and family history of coronary disease (25.1 vs 18.2%). The prevalence of active smoking was significantly higher in men (20.4 vs 12%). The prevalence of hyperlipidemia was not different between men and women (34.5 vs 31.8%). Stepwise logistic regression showed that HT and diabetes were both independently associated with BMI and age, while diabetes and not HT was independently associated with female gender. The prevalence of classical CV risk factors was very high in this population with OSAS requiring CPAP, especially in women. There is thus a very elevated CV risk level independent of that directly related to OSAS. It is important to screen for and treat classical CV risk factors in this population. ª 2010 Elsevier Ltd. All rights reserved.
Introduction Cardiovascular (CV) morbidity and mortality are increased in patients with obstructive sleep apnoea syndrome (OSAS).1,2 Cross-sectional studies have shown that OSAS is an independent risk factor for coronary artery disease (CAD) and stroke.3,4 However the relative risk of CAD associated with OSAS was weak and did not increase with increasing severity of OSAS.3 Several observational studies have shown that the incidence of CV complications is increased in untreated OSAS and decreased with continuous positive airway pressure (CPAP).5,6 However, in these studies, the control group consisted of patients refusing CPAP or not complying with it who were also likely to have poor compliance with CV risk reduction treatments. Therefore it is not clear how much of the increased CV risk in apnoeic patients is directly attributable to the OSAS and how much is related to the concomitant presence of other classical CV risk factors. Furthermore studies that have evaluated CV risk factors in OSAS have certain limitations. In most studies, patients with pre-existing CV disorders were not excluded, thereby leading to an overestimation of CV risk factors. The number of patients included in these studies is very variable.7 The proportion of patients with severe OSAS is small in many studies, even in those with large cohorts.3,8 The evaluation of CV risk factors is often incomplete, especially as regards the presence of a family history of CAD which is a major risk factor. Finally, the evaluation of CV risk in OSAS patients has generally been studied in men only; most studies had few or no women.5e7,9,10 Thus we set out to study the prevalence of major CV risk factors in a large population of men and women with OSAS, requiring CPAP, and free of pre-existing known CV disease. We wished to compare the prevalence of these risk factors between men and women.
Patients and methods Consecutive patients with OSAS were included on CPAP initiation, over a period of eighteen months, in 10 homecare associations in the federated ANTADIR network (Association Nationale pour les Traitements A Domicile, les Innovations et la Recherche).11 The diagnosis of OSAS was established by full polysomnography or by nocturnal respiratory polygraphy.
This was an observational study. A questionnaire was given to the patient, at home, by paramedical personnel from the homecare association, on the day that the CPAP treatment was initiated. The questionnaire consisted in 38 questions concerning CV risk factors, previous CV events and medications usually used. If the patient had difficulties the paramedical helped him to fill out the questionnaire. The diagnosis of hypertension (HT) and diabetes depended on the patient replying that he had HT and/or diabetes or by the identification of treatment for the condition in the list of medicines taken. Beta-blockers and ACE-inhibitors were regarded as anti-hypertensive drugs and not as treatment for heart failure. Angiotensine IIinhibitors, Ca-channel blockers and diuretics were considered as anti-hypertensive drugs. Obesity was defined by a body mass index (BMI) 30 kg/m2, based on weight and height reported by the patient. Patients were considered as current smokers in the case of active smoking within the year. The diagnosis of hyperlipidemia was based on the patients’ report of the diagnosis and by the identification of treatment for the condition in the list of medications. The presence of a family history of CAD was defined by reported angina, or myocardial infarction or a history of coronary artery bypass operation, or a history of coronary angioplasty before the age of 60 years in any first degree relative (parent or sibling, or child). We excluded from analysis patients with pre-existing coronary artery disease, cerebrovascular disease or peripheral artery disease. Coronary artery disease was defined by reported angina, or myocardial infarction or a history of coronary artery bypass operation, or a history of coronary angioplasty. Cerebrovascular disease was defined as reported stroke or surgery or angioplasty of a carotid artery. Peripheral artery disease was defined by any past history of surgery or angioplasty of a lower limb artery. We have therefore excluded 338 patients in whom the questionnaire revealed a previous cardiovascular disease. ANTADIR is registered with the National Committee of Civil Liberties (CNIL). This study was approved by the Scientific Committee of ANTADIR. All numerical variables were expressed as mean SD. The c2 test was used for a comparison of categorical variables between two groups. For the comparison of numerical variables between two groups, the Student t test was used. A p value of < 0.05 was considered as statistically significant. In the case of finding a difference in the prevalence of a CV risk factor between men and women, and to determine if this
Cardiovascular risk factors in men and women difference was due independently to gender, we carried out a stepwise logistic regression. In this model the CV risk factor concerned was the dependent variable and gender was one of the tested independent variables.
1065 gender (p Z 0.021). Female gender was therefore independently and positively associated with diabetes, with and odds ratio of 1.56 (95% CI:1.07e2.27).
Discussion Results The study population consisted of 1117 patients, 834 men and 283 women, with OSAS requiring CPAP and free of a known CV disease. Age, apnoeaehypopnoea index (AHI) and parameters of nocturnal desaturation did not significantly differ between men and women (Table 1). The prevalence of severe OSAS (AHI 30), moderate OSAS (AHI: 15e29.9) and mild OSAS (AHI: 5e14.9) was respectively 85.6%, 12.2% and 2.2%. The prevalences of CV risk factors are shown in Table 2. The prevalences of HT, diabetes, obesity and family history of CAD were respectively 54.1%, 22.8%, 65.8%, 20% and were significantly higher in women than in men. The prevalence of current smoking was 18.3% and was significantly higher in men than in women. The prevalence of hyperlipidemia was 33.8% and did not differ between men and women. The percentages of patients with at least two or three of the following CV risk factors, i.e. hypertension, diabetes or obesity were significantly higher in women than in men (Table 3). In the female population, there were no differences in the levels of different CV risk factors between those with severe OSAS, and those with moderate to mild OSAS (Table 4). In the male population, the prevalence of HT and obesity was higher in those with severe OSAS than in those with moderate to mild OSAS (Table 4). A stepwise logistic regression was carried out to examine whether the increased prevalence of HT or diabetes was independently associated with female gender. In the first model the dependent variable was HT, and the tested variables were sex, age, BMI, AHI, smoking status and diabetes. The variables associated independently with HT were BMI (p < 0.001), age (p < 0.001) and diabetes (p < 0.001), but gender was not associated independently with HT. In the second model, the dependent variable was diabetes and the tested variables were sex, age, BMI, AHI, smoking status and HT. The variables associated independently with diabetes were BMI (p < 0.001), HT (p < 0.001), age (p < 0.001), smoking status (p Z 0.003) and female
This observational study shows that the prevalence of various CV risk factors was very high in a group of 1117 patents with OSAS, requiring CPAP treatment while not presenting a known CV illness. The prevalence of HT, diabetes, obesity and family history of CAD was greater in women than in men, while age and the severity of OSAS were not. This was an observational study of a large series of patients in a multicenter home-care system for management of patients with CPAP. An auto-questionnaire was used with assistance given to the patient when required. Thus there was no specific medical visit for the study. The patients were interviewed at the time of initiation of CPAP therapy, at home, and thus there was a large recruitment over a short period of time. In our study the prevalence of different CV risk factors, apart from current smoking was greater than in the general population: particularly, diabetes, HT and obesity.12e14 The prevalence of different CV risk factors observed in our study can be compared with data from the literature as regards the CV profile of patients with OSAS. In our study, the prevalence of diabetes was very elevated at 22.8% in patients with OSAS requiring CPAP. Similar figures have been noted in the literature6,9 while other studies have shown a lower prevalence of diabetes in OSAS3,5 which could be related to the lower age of the patients. In our study, the prevalence of HT was high (54.1%), and comparable to that usually found in severe OSAS.6,15e19 The prevalence of obesity was very high in our study (65.8%), but the prevalence of obesity in OSAS is related to the level of the AHI.16 The prevalence of hyperlipidemia was elevated in our study (33.8%) even though the patients did not have known CV disease. Twenty percent of the patients in our study had a family history of CAD. This important CV risk factor20 has not been mentioned in other published studies evaluating CV risk related to OSAS.5,6,21 Furthermore it has been reported that patients with OSAS had a greater family history of premature coronary related death, independently of the presence of CAD associated with the OSAS.22 The prevalence of current smoking in our
Table 1 Age, body mass index (BMI) and nocturnal respiratory data for the overall population with comparison between men and women.
Age (years) BMI (kg/m2) BMI 40 kg/m2 (% pat) AHI (/h) Minimal nocturnal SaO2 (%) Mean nocturnal SaO2 (%) % RT SaO2 < 90 %
All patients (n Z 1117)
Men (n Z 834)
Women (n Z 283)
p
57.7 12 33.6 7.3 15.8 48.6 21.4 70.8 14.6 91.3 4 24.3 25.7
57.7 12 32.4 6.2 10.2 49 20.9 71 14.6 91.3 4.1 23.7 25.2
57.7 12.1 36.8 9.1 32.5 47.2 22.9 70.2 14.8 91.3 3.9 26 27.1
NS <0.001 <0.001 NS NS NS NS
AHI, apnoea-hypopnoea index; RT, recording time.
1066 Table 2 women.
J.-P. Laaban et al. Prevalence of different cardiovascular risk factors for the overall population with comparison between men and
Hypertension (% pat) Diabetes (% pat) Obesit (BMI 30 kg/m2) (% pat) Current smoker (% pat) Current or ex-smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat)
All patients (n Z 1117)
Men (n Z 834)
Women (n Z 283)
p
54.1 22.8 65.8 18.3 66.2 33.8 20.0
51.4 20.4 62.0 20.4 77.0 34.5 18.2
62.1 29.9 77.0 12.0 34.3 31.8 25.1
<0.005 <0.002 <0.001 <0.002 <0.001 NS <0.005
BMI, body mass index; CAD, coronary artery disease.
study was 18.3%, while other studies that evaluated CV risk related to OSAS reported comparable or higher or lower figures.5e7,16,17,19 One important result of the current study is that the prevalence of most of the CV risk factors was greater in women than in men. In our study the prevalence of diabetes was greater in women than in men. These results are not in accordance with the data reported in the general population in whom the prevalence of diabetes is higher in men than in women.13 This increased prevalence of diabetes in women may possibly be explained by the greater prevalence of obesity in women, in particular massive obesity. However female gender was in our study an independent predictive factor for the prevalence of diabetes. There are no other data in the literature concerning the prevalence of diabetes in women with severe OSAS, since most studies did not include women5,9 or did not compare the prevalence of diabetes between men and women with OSAS.6 We have no explanation accounting for our finding of an independent association between female gender and diabetes which should be confirmed by further studies. The prevalence of HT was higher in women than in men in our study. This is surprising since male gender is a well known risk factor for HT. This increased prevalence of HT in women was not explained by greater age, nor by more severe OSAS, but most probably by a greater prevalence of obesity, and in particular massive obesity in women in our study. The prevalence of obesity was greater in women than in men in this study, and massive obesity was three times more frequent in women. It has been shown that the level of obesity necessary to cause OSAS is higher in women than in men, aged under 65 years.21 The prevalence of hyperlipidemia was not different between men and women in
this study, whilst in the general population, hyperlipidemia is more frequent in men.12,13 In this study women had a greater level of family history of coronary disease without any evident explanation. We have noted a greater level of smoking in men than in women, as is common in the general population,12 and confirmed in patients with OSAS.21 In this study the prevalence of CV risk factors was not significantly influenced by the severity of OSAS in women, whilst in men the prevalence of HT and obesity was much higher in severe OSAS than in mild-moderate OSAS. These data suggest that the proportion of cardiovascular risk directly attributable to OSAS is greater in men than in women. The prevalence of CV risk factors, although high, may be underestimated in this study. We only took into account patients with previously known diabetes, and a certain number of apnoeic patients may have had occult diabetes.23 Studies using ambulatory control of blood pressure over 24 h showed that HT is often underdiagnosed by standard clinical methods in patients with OSAS.24 We only took into account patients with hyperlipidemia treated by lipid lowering agents. In studies with systematic measurement of blood lipid levels the prevalence of hyperlipidemia was higher.6,7 The high levels of CV risk factors in this study could be explained by the fact that OSAS may favour the development of certain CV risk factors, especially HT. Transverse and longitudinal studies have shown that OSAS is an independent risk factor for HT,8,15e17,25 but the relative risk of HT independently related to OSAS is not very high.26,27 We have found an elevated level of CV risk factors in this population with OSAS. However, this can be considered as a selected population, given that recruited patients have
Table 3 Number of cardiovascular (CV) risk factors per patient and association of CV risk factors for the overall population with comparison between men and women.
No of CV risk factors per patient 3 CV risk factors (3e6) per patient (% pat) At least hypertension and obesity (% pat) At least diabetes and obesity (% pat) At least hypertension and diabetes (% pat) At least hypertension, diabetes and obesity (% pat)
All patients (n Z 1117)
Men (n Z 834)
Women (n Z 283)
p
2.1 1.3 39.8 41.5 18.8 16.7 14.4
2.1 1.3 37.3 38.4 16.1 14.5 12.1
2.4 1.3 47 50.9 26.9 23 21.2
<0.001 <0.005 <0.01 <0.001 <0.001 <0.001
Cardiovascular risk factors in men and women
1067
Table 4 Comparison of the prevalence of different cardiovascular risk factors between patients with mild-moderate OSAS(AHI Z 5e29.9) and patients with severe OSAS(AHI 30) for the overall population with comparison between men and women. All patients
AHI Z 5e29.9 (n Z 161)
AHI 30 (n Z 956)
p
Hypertension (% pat) Diabetes (% pat) Obesity (BMI 30 kg/m2) (% pat) Current smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat) Age (years)
45.0 16.9 50.3 18.0 32.3 20.5 56.6 12.2
55.7 23.8 68.4 18.3 34.1 19.9 57.9 12.0
<0.001 NS <0.001 NS NS NS NS
Women
AHI Z 5e29.9 (n Z 52)
AH 30 (n Z 231)
p
Hypertension (% pat) Diabetes (% pat) Obesity (BMI30 kg/m2) (% pat) Current smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat) Age (years)
60.8 23.5 69.2 13.5 25.0 23.1 58.0 11.7
62.4 31.3 78.8 11.7 33.3 25.5 57.7 12.2
NS NS NS NS NS NS NS
Men
AHI Z 5e29.9 (n Z 109)
AHI 30 (n Z 725)
p
Hypertension (% pat) Diabetes (% pat) Obesity (BMI 30 kg/m2) (% pat) Current smoker (% pat) Hyperlipidemia (% pat) Family history of CAD (% pat) Age (years)
37.6 13.8 41.3 20.2 35.8 19.3 56.0 12.3
53.5 21.4 65.1 20.4 34.3 18.1 58.0 11.9
<0.001 NS <0.001 NS NS NS NS
BMI, body mass index; CAD, coronary artery disease.
been screened by the process of referral to a sleep center and subsequently recommended for treatment. Thus the level of CV risk in a population with suspected OSAS may well be different and likewise in a population with OSAS but not recommended for treatment. Indeed patients with OSAS but with no other concomitant health problems are probably less likely to visit a doctor and may thus be underrepresented in our population. Likewise, women tend to seek medical help more than men, but there may be a counteracting bias against diagnosing OSAS in women. Indeed men without concomitant disease may be induced to visit a doctor at the behest of their partner more than the inverse situation. CPAP treatment is reimbursed in France when the patient has over 30 respiratory events per hour or over 10 respiratory related micro-arousals per hour, with symptoms. Thus in this series the patients could be considered to be more severely affected than in North American. Thus a high proportion of patients can be classed as having severe sleep apnoea, but in contrast we have excluded patients with known CV disease. It could be argued that there is a bias to treat OSAS patients with CV risk factors with CPAP, but no figures exist as to the proportion of known OSAS patients untreated nor as to the relative proportions of patients having other treatment options such as surgery or mandibular advancement prosthesis. Thus this selected population that is physician referred and with severe OSAS, has a high incidence of CV risk
factors. Our findings sustain the idea that such patients merit screening for CV risk factors and health advice aimed to reduce such risks. In order to diminish CV morbidity and mortality it is essential to optimally treat CV risk factors and this is probably as important as treating effectively the sleep apnoea.
Conflict of interest statement The authors have no conflicts of interest.
References 1. McNicholas WT, Bonsignore MRThe Management Committee of EU COST ACTION B26. Sleep apnoea as an independent risk factor for cardiovascular disease: current evidence, basic mechanisms and research priorities. Eur Respir J 2007;29: 156e78. 2. Bradley TD. Obstructive sleep apnoea and its cardiovascular consequences. Lancet 2009;373:82e93. 3. Shahar E, Whitney CW, Redline S, et al. Sleep-disordered breathing and cardiovascular disease. Cross-sectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med 2001; 163:19e25. 4. Arzt M, Young T, Finn L, et al. Association of sleep-disordered breathing and the occurrence of stroke. Am J Respir Crit Care Med 2005;172:1447e51. 5. Marin JM, Carrizo SJ, Vicente E, et al. Long-term cardiovascular outcomes in men with obstructive sleep apnoeaehypopnoea with or without treatment with continuous positive
1068
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
airway pressure: an observational study. Lancet 2005;365: 1046e53. Buchner NJ, Sanner BM, Borgel J, et al. Continuous positive airway pressure treatment of mild to moderate obstructive sleep apnea reduces cardiovascular risk. Am J Respir Crit Care Med 2007;176:1274e80. Kiely JL, McNicholas WT. Cardiovascular risk factors in patients with obstructive sleep apnoea syndrome. Eur Respir J 2000;16: 128e33. Bixler EO, Vgontzas AN, Lin HM, et al. Association of hypertension and sleep-disordered breathing. Arch Intern Med 2000; 160:2289e95. Meslier N, Gagnadoux F, Giraud P, et al. Impaired glucoseinsulin metabolism in males with obstructive sleep apnoea syndrome. Eur Respir J 2003;22:156e60. Peker Y, Carlson J, Hedner J. Increased incidence of coronary artery disease in sleep apnoea: a long term follow-up. Eur Respir J 2006;28:596e602. Stuart M, Weinrich M. Integrated health system for chronic disease management. Lessons learned from France. Chest 2004;125:695e703. ¨ unpuu S, et al. Effect of potentially Yusuf S, Hawken S, O modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004;364:937e52. Rudnichi A, Safar M, Asmar R, et al. Prevalence of cardiovascular risk factors in a French population. J Hypertens 1998; 16(Suppl. 1):S85e90. Ricordeau P, Weill A, Vallier N, et al. The prevalence and cost of diabetes in metropolitan France: what trends between 1998 and 2000? Diabetes Metab 2003;29:497e504. Lavie P, Herer P, Hoffstein V. Obstructive sleep apnoea syndrome as a risk factor for hypertension: population study. BMJ 2000;320:479e82. for the Sleep Heart Health Study Nieto FJ, Young TB, Lind BK, et al. Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. JAMA 2000;283:1829e36.
J.-P. Laaban et al. 17. Young T, Peppard P, Palta M, et al. Population-based study of sleep-disordered breathing as a risk factor for hypertension. Arch Intern Med 1997;157:1746e52. 18. Grote L, Ploch T, Heitmann J, et al. Sleep-related breathing disorder is an independent risk factor for systemic hypertension. Am J Respir Crit Care Med 1999;160:1875e82. 19. Bo ¨rgel J, Sanner BM, Bittlinsky A, et al. Obstructive sleep apnoea and its therapy influence high-density lipoprotein cholesterol serum levels. Eur Respir J 2006;27:121e7. 20. LLoyd-Jones DM, Nam BO, D’Agostino RB, et al. Parental cardiovascular disease as a risk factor for cardiovascular disease in middle-aged adults. A prospective study of parents and offspring. JAMA 2004;291:2204e11. 21. Newman AB, Nieto FJ, Guidry U, et al. Relation of sleepdisordered breathing to cardiovascular disease risk factors. The Sleep Heart Health Study. Am J Epidemiol 2001;154:50e9. 22. Gami AS, Rader S, Svatikova A, et al. Familial premature coronary artery disease mortality and obstructive sleep apnea. Chest 2007;131:118e21. 23. Seicean S, Kirchner HL, Gottlieb DJ, et al. Sleep-disorderd breathing and impaired glucose metabolism in normal-weight and overweight/obese individuals. The Sleep Heart Health Study. Diabetes Care 2008;31:1001e6. 24. Baguet JP, Hammer L, Le ´vy P, et al. Night-time and diastolic hypertension are common and underestimated conditions in newly diagnosed apnoeic patients. J Hypertens 2005;23: 521e7. 25. Peppard PE, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:1378e84. 26. Bazzano LA, Khan Z, Reynolds K, et al. Effect of nocturnal nasal continuous positive airway pressure on blood pressure in obstructive sleep apnea. Hypertension 2007;50:417e23. 27. Haentjens P, Van Meerhaeghe A, Moscariello A, et al. The impact of continuous positive airway pressure on blood pressure in patients with obstructive sleep apnea syndrome. Evidence from a meta-analysis of placebo-controlled randomized trials. Arch Intern Med 2007;167:757e65.