Nighttime is a vulnerable time for death from. The Relationship Between Congestive Heart Failure, Sleep Apnea, and Mortality in Older Men*

Transcription

1 The Relationship Between Congestive Heart Failure, Sleep Apnea, and Mortality in Older Men* Sonia Ancoli-Israel, PhD; Einat R. DuHamel, MD; Carl Stepnowsky, PhD; Robert Engler, MD; Mairav Cohen-Zion, MA; and Matthew Marler, PhD Study objectives: To examine the association of sleep apnea with heart disease. Design: Prospective study. Setting: Medical wards at the Veterans Affairs San Diego Healthcare System. Patients: Three hundred fifty-three randomly selected inpatient men. Measurements and results: Sleep was recorded for 2 nights in the hospital. Medical conditions were obtained from hospital medical records. Cox proportional hazards analyses indicated that patients with congestive heart failure (CHF) plus central sleep apnea (CSA) had shorter survival than those with just CHF, just sleep apnea (obstructive or central), or neither. Survival for those with obstructive sleep apnea (OSA) or CSA and no CHF was no different than for those with neither disorder. Follow-up analysis showed that for those with no CHF, neither CSA nor OSA shortened survival (p > 0.80). For those with CHF, having CSA shortened the life span with a hazard ratio of 1.66 (p 0.012), but having OSA had no effect. Patients with CHF had more severe sleep apnea than those with no heart disease. Conclusions: This study does not clarify the issues of cause and effect, but does reinforce the strong associations between sleep apnea and heart disease in elderly men. These data suggest that people with coronary disease should be regarded as a risk group for sleep apnea. (CHEST 2003; 124: ) Key words: central sleep apnea; congestive heart failure; mortality Abbreviations: AHI apnea-hypopnea index; BMI body mass index; CHF congestive heart failure; CPAP continuous positive airway pressure; CSA central sleep apnea; CSR Cheyne-Stokes respiration; ODI oxygen desaturation index; OSA obstructive sleep apnea; VASDHS Veterans Affairs San Diego Healthcare System Nighttime is a vulnerable time for death from cardiovascular disease. 1 Nighttime is also a vulnerable time for death from sleep apnea. 2 Since there is a high prevalence of sleep apnea in the elderly, 3 it has been suggested that sleep apnea may *From the Departments of Psychiatry (Drs. Ancoli-Israel and Marler, and Ms. Cohen-Zion), Reproductive Medicine (Dr. DuHamel), and Medicine (Dr. Engler), University of California, San Diego; and Veterans Affairs San Diego Healthcare System (Dr. Stepnowsky), San Diego, CA. All work was performed at University of California San Diego and VASDHS. Supported by NIA AG02711, NIA AG08415, NCI CA85264, NHLBI HL44915, NHLBI HL36005; the Department of Veterans Affairs VISN-22 Mental Illness Research, Education, and Clinical Center; and the Research Service of the VASDHS. Manuscript received December 28, 2001; revision accepted May 27, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( permissions@chestnet.org). Correspondence to: Sonia Ancoli-Israel, PhD, Department of Psychiatry 116A, VASDHS, 3350 La Jolla Village Dr, San Diego, CA 92161; sancoliisrael@ucsd.edu increase the risk of both cardiovascular morbidity and mortality in that age group. 2,4 The cardiovascular consequences of sleep apnea are of interest since these patients often have severe hypoxemia, and the adverse effects of hypoxemia on the heart are well documented. 5,6 Sleep apnea has been associated with cardiac arrhythmias, cyclic tachycardia, and hypertension in younger clinic patients. 7,8 Schafer et al 9 showed that patients with obstructive sleep apnea (OSA) and coronary heart disease had more apnea-associated ischemia, which in turn activated the CNS and fragmented sleep. Other studies have also found a high prevalence of sleep apnea in patients with heart failure. Work from the Wisconsin cohort 14 and from the Sleep Heart Health Study 15 has suggested sleep apnea may be a risk factor for hypertension and consequent cardiovascular morbidity in the general population. If morbidity is increased, mortality may also be increased. One approach to increasing survival in patients with 1400 Clinical Investigations

2 heart disease may be treating the sleep apnea. A series of studies and a recent study by Peker et al 22 suggested that using continuous positive airway pressure (CPAP) to treat central sleep apnea (CSA) and Cheyne-Stokes respiration (CSR) in patients with chronic heart failure may also alleviate the heart disease and possibly lengthen survival. Naughton and Bradley 23 treated congestive heart failure (CHF) and CSR with continuous positive airway pressure (CPAP) and found that nightly treatment attenuated the CSA and CSR, improved cardiac function and alleviated symptoms of heart failure. These data confirm that some of the morbidity of heart disease, and therefore perhaps mortality as well, may be worsened by the presence of sleep apnea, particularly CSA. Several investigators have postulated mechanisms of action. The occurrence of frequent episodes of oxygen desaturation and the many arousals caused by apnea may contribute to both left and right ventricular hypertrophy. 24 Noda et al 24 in fact showed that the prevalence of left and right ventricular hypertrophy was higher in patients with severe OSA syndrome than those with mild OSA apnea. They suggested that sleep apnea may influence structural changes in the heart, hypoxia, and elevations in sympathetic activity. It is clear that patients with CHF have shorter survival. 25 In many studies, 26,27 CSA, and particularly CSR, have also been associated with increased mortality. In our own laboratory, in a small study of patients with left ventricular failure, we found that all patients with left ventricular failure and CSR with more than five apneas per hour of sleep died within 6 months, while only 30% of patients without recurrent apnea died within 6 months. 27 Lanfranchi et al 28 followed up 62 patients with chronic heart failure for 2 years and found that those with higher respiratory disturbance indexes, but not higher percentage of CSR, had shorter survival. In a third study, Andreas et al 29 found no association between mortality and CSA, but did find an association between CSR during the day and increased mortality. The studies discussed above focused on fairly small samples followed up for approximately 5 years. The goal of this study was to examine whether having OSA or CSA in addition to heart disease might have a negative effect on survival in older adults. To answer this question, a large sample of older men with and without heart disease and with and without sleep apnea were followed up for up to 17.5 years. Subjects Materials and Methods Three hundred fifty-three men who were inpatients on the medical wards at the Veterans Affairs San Diego Healthcare System (VASDHS) were randomly selected from all new daily admissions to the medical wards. Exclusion criteria included being 60 years old; diagnosis of cancer, dementia, or infectious disease; and being too sick to consent or being in the ICU. Initial data gathering occurred over a 5-year period from 1985 to The project was approved by the University of California, San Diego Institutional Review Board in accordance with the recommendations found in the 1975 Helsinki Declaration. Procedure Sleep was recorded for 2 nights in the hospital using the modified Respitrace/Medilog (Oxford Medical; Oxon, England) portable recording system. 30 At the time of the study, this system was one of the only ambulatory systems available. The system recorded thoracic and abdominal respiration, tibialis electromyogram, and wrist activity to distinguish sleep from wake state. This system was previously validated against the gold standard in-laboratory polysomnograph. 30 Finger pulse oximetry was recorded on 1 of the 2 nights using a Biox finger-pulse oximeter (model 3700; Ohmeda; Boulder, CO). All subjects were interviewed about their sleep. Medical conditions and diagnoses were obtained from hospital medical records. Recordings were scored for number of obstructive apneas (defined as thoracic and abdominal respiration 180 out of phase, and at least a 90% reduction in respiratory effort for a minimum of 10 s), number of central apneas (defined as thoracic and abdominal respiration in phase and at least a 90% reduction in respiratory effort for a minimum of 10 s), number of hypopneas (defined as a 50 to 90% reduction in effort), total sleep time, and wake after sleep onset. The apnea-hypopnea index (AHI), the number of apneas and hypopneas per hour of sleep, was computed for each night, and the mean of the two nights was used in all analyses. Oximetry records were scored with the PROFOX program (PROFOX Associates; Escondido, CA) 31 for mean oxygen saturation, number of desaturations, and percentage of time spent at an arterial oxygen saturation 90%. An oxygen desaturation index (ODI), the number of desaturations per hour of sleep, was computed. A desaturation was defined as a 4% decrease in the saturation level. Follow-up Follow-up was conducted yearly with the last follow-up occurring in the year A letter was sent to each participant asking for current address; in cases where the person was no longer living, information on where and when he died was requested. Death certificate information was obtained either from the hospital where the patient had died, from the San Diego County records office or from the Benefits Information Records Locator System (a Veterans Affairs automated system for identifying and keeping track of veteran beneficiaries with a built-in linkage to the veterans death certificates 32 ). Medical records at the Sleep Apnea Clinic at the VASDHS were examined to determine if any of the patients had received treatment for their sleep-disordered breathing. Although all patients had been sent a letter reporting their condition after their first sleep evaluation, no patients were treated at the VASDHS for sleep apnea. It was not possible to determine if any of the patients received treatment for sleep apnea at another facility. Data Analysis Logistic regression was run for the dependent variable CHF. The continuous independent variables were age, education, body CHEST / 124 / 4/ OCTOBER,

3 mass index (BMI), and years smoked packs per year; dichotomous independent variables were race, income, alcohol intake, history of hypertension, history of diabetes, and AHI. In addition, smoking history, medical conditions (such as hypertension, pulmonary and bronchial disease, and diabetes), and other variables that may confound the association between sleep apnea and CHF were included in the analyses as dichotomous variables. 2 analyses were also run for presence/absence of CHF. The Cox proportional hazards model was used for survival analysis. SPSS (SPSS Version 10.0; SPSS; Chicago, IL) and SAS (SAS Version 9.0; SAS Institute; Cary, NC) were used for all analyses. Demographics Results The mean age of the 353 men was 69 years (SD, 7.7 years; range, 60 to 99 years). Mean BMI was 25.4 (SD, 5.0; range, 14.2 to 47.6). Eighty-five percent of the sample were white, 4% were African American, and 11% were other. Table 1 shows the percentage of the sample with different medical conditions, as listed in medical records. Table 2 shows the medications received by these patients. Prevalence In this sample, at the initial visit, 53% of the men (n 186) had an AHI 15. The mean AHI for the entire sample was 23.2 (SD, 21.3; range, 0 to 112). Thirty-two percent of the sample (n 112) had CHF. Sleep-Disordered Breathing Those with a history of CHF had significantly higher AHIs than those with no heart disease (mean AHI, 27.6 [SD, 24.0] vs mean AHI, 20.0 [SD, 20.0]; p 0.014) and significantly higher ODI (mean ODI, 42.5 [SD, 60.0] vs mean ODI, 27.5 [SD, 41.6]; Table 1 Percentage of Patients With Major Medical Illnesses Illnesses % Hypertension 50 Angina 50 Myocardial infarction 37 CHF 32 Peripheral vascular disease 27 COPD 25 Diabetes 24 Stroke 16 Bronchitis 14 Renal disease 12 Transient ischemic attack 7 History of cancer 3 Cirrhosis 3 Thyroid disease 4 Pancreatitis 1 Table 2 Percentage of Patients Receiving Medications Medication Class % Cardiac drugs 70 Analgesics 54 Diuretics 42 Antihypertensives 36 Sedative-hypnotics 25 Bronchodilators 21 Steroids 16 Minor tranquilizers 9 Antidepressants 6 Antihistamines 6 Home oxygen 5 Narcotics 3 Anticonvulsants 2 Major tranquilizers 2 Vasodilators 2 Anti-Parkinsonian 1 Stimulants 0.6 p 0.030). Forward selection logistic regression indicated that for CHF, age (in years), AHI, and years smoked packs per year, were all independently significant (Table 3). Mortality A prospective study of mortality was performed. Follow-up occurred up to 17.5 years (mean followup, 5.7 years; SD, 4.5 years; range, 0 to 17.5 years); some subjects were studied in their last month of life. As of 2002, of the original 353 men, 86% had died (n 303) and 14% were known to be alive (n 50). None were unavailable for follow-up. The question was asked whether having OSA or CSA in addition to heart disease might be associated with shorter survival. Cox proportional hazards survival analyses were run for those with and without sleep apnea and with and without CHF (Table 4). As expected, those with CHF (with or without OSA or CSA) had shorter survival than those without CHF. However, those subjects with CHF comorbid with either OSA or CSA (AHI 15) [CHF plus sleep apnea] had significantly shorter survival than those with CHF and no OSA or CSA, OSA, or CSA with no CHF, or neither (p 0.001). Survival of Table 3 Logistic Regression Analysis Condition Risk Factors Coefficient Odds Ratio p Value CHF* Constant Age, yr AHI Pack-yr of smoking *Presence or absence. AHI 15 vs AHI Clinical Investigations

4 Table 4 Cox Proportional Hazards Model for CHF and Sleep Apnea Variables Percentage Dead Mean Survival, yr 95% Confidence Interval, Mean Median Survival, yr Hazard Ratio CHF plus CSA CHF Sleep apnea* None ref *OSA or CSA. The hazard ratio takes into account the difference between those who have died and those who are still living at the time last contacted. The mean survival includes the time in the study of everyone whether they have died or are still alive. those with OSA or CSA but no CHF was no different from survival for those with neither CHF nor sleep apnea (Fig 1). Since CHF is known to be associated with CSA, and since close to half (48%) of the men with AHI 15 had CSA, analyses were done to determine if mortality was related to having any type of apnea or primarily to having CSA. The follow-up analysis showed that for those with no CHF, neither CSA or OSA shortened survival (p 0.80). For those with CHF, having CSA shortened the life span with a hazard ratio of 1.66 (p 0.012), but having OSA had no effect. Other variables added into the survival model as controls that were significant included pulmonary disease (p 0.01), age (p 0.01), BMI (p 0.01), use of home oxygen (p 0.01), diabetes (p 0.01), dialysis (p 0.05), and cancer (p 0.01). Although significant in the proportional hazards survival models, they did not account for the other significant results, ie, these effects were additional to the effects Figure 1. Survival curves for those with CHF, and/or sleepdisordered breathing, or neither. Those with CHF plus CSA had significantly shorter survival (p 0.001) than those with just CHF or just OSA or CSA. of CHF and sleep apnea. All other diseases (as listed in Table 1) and saturation variables were not significant. Discussion Our study examined randomly selected Veteran men 60 years old who were inpatients on a medical ward at the VASDHS, and who were followed up for 17.5 years. In this sample, there were significantly greater mortality rates in patients who had both CHF and CSA than in those with just CHF or just CSA or just OSA. Patients with CSA or OSA survived 6.75 years, patients with CHF survived just 4 years, while patients with both CHF and CSA survived just 2.5 years. Survival of patients with just CSA or OSA was no different than that of those patients with neither heart disease nor sleep apnea. The shorter survival for those with CSA and CHF was not explained by any association with pulmonary disease or other comorbidities. The results of this study confirm previous findings but in a larger sample with more years of follow-up and with more covariates included in analyses than some other studies. Findley et al, 27 in a small sample followed up for 6 months, was one of the first to report that men with heart failure and CSA had a greater rate of death than those with heart failure and no sleep apnea. Hanly et al 33 showed that the risk of cardiac death in heart disease was increased in those with CSA. Sin et al 19 found that patients with CSA who received CPAP had a reduction in the combined rate of mortality and heart transplantation compared to patients with CSA who did not receive CPAP. However, among patients with no CSA, there was no difference in mortality transplantation rate between those who received and did not receive CPAP. Those observations suggest that the CSA initially was a risk factor in those patients. The results of this study of larger samples with more years of follow-up also suggest that CSA with CHF is a risk factor. CHEST / 124 / 4/ OCTOBER,

5 We have previously shown that in community dwelling elderly, those with more severe sleep apnea, defined as an AHI 30, had significantly shorter survival (7.9 years) compared to those with AHI 15 (9.4 years). Multiple regression analysis that included age, gender, BMI, and history of pulmonary or cardiovascular disease, however, resulted in only age, pulmonary and cardiovascular disease, but not AHI, being significant predictors of mortality. 2 In a review of sleep apnea and cardiovascular disease, Young and Peppard 34 noted that this should not be interpreted as a negative finding. They presented a model of sleep apnea and cardiovascular disease that suggested three components: nightly exposure to sleep apnea as a direct cause of cardiovascular disease, nightly exposure to sleep apnea causing hypertension and thus indirectly causing of cardiovascular disease, or sleep apnea modifying the effect of cardiovascular disease on mortality. Young and Peppard 34 suggested that the data of Ancoli- Israel et al 2 could be explained by their model, ie, one would not expect AHI to be predictive when cardiovascular disease was controlled for, since the AHI would lead to the cardiovascular disease, which would then lead to death. The same theory could be applied to the data in this study. The fact that men with both CSA and CHF had significantly shorter survival, by several years, than men with just CHF, may reflect the sleep apnea modifying effect of cardiovascular disease on mortality or it may reflect the severity of the heart disease; ie, it is possible that sleep apnea is a marker for the severity of the heart disease rather than a causal factor. Somers, 35 in an editorial, also suggested that the pathophysiology of CHF and of sleep apnea may be linked and in fact may potentiate each other. While additional research is needed to more fully understand these relationships, the results of this study confirm previous findings with a larger sample and longer follow-up. Sleep apnea is found in approximately 60% of patients with heart failure, 23,36 and CSA and CSR account for approximately 30 to 40% of those, 13,37 with many of them having occult sleep apnea. 12 In this sample, patients with CHF had more severe sleep apnea than those without CHF. Sin et al, 38 in a large study of 450 men and women referred to a sleep laboratory, determined that the risk factors for CSA were male gender, having atrial fibrillation and hypocapnia, and being 60 years old. Shahar et al 39 reported that sleep apnea, even in the mild-tomoderate states, was strongly associated with selfreported heart failure and stroke in the 6,000 participants of the Sleep Heart Health Study. Two recent large epidemiologic studies, the Wisconsin cohort 14 and the Sleep Heart Health Study, 15 showed that sleep apnea may be a risk factor for hypertension and consequent cardiovascular morbidity in general population. The authors suggested that sleep apnea should be treated even at low levels to avoid the development of these comorbidities. Our data support this recommendation and suggest extra caution when patients, particularly older patients with CHF, also have CSA. If the mortality rate of those with CSA and CHF is higher than the rate of those with just CHF, it is imperative that older patients at risk for these disorders be treated. This study does not clarify the issues of cause and effect, but does reinforce the strong associations between sleep apnea and heart disease, CHF, in elderly hospitalized men. It is unknown whether the same results would hold in nonhospitalized elderly men. Nevertheless, these data suggest that older men with coronary disease should be regarded as a risk group for sleep apnea. Sleep apnea may adversely affect CHF and may induce a higher mortality rate among patients with CHF. Other investigators have shown that treating sleep apnea in patients with chronic heart failure results in improved leftventricular ejection fraction and decreased daytime plasma norepinephrine, atrial natriuretic peptide, and heart rate. 40,41 While controlled treatment trials are still needed, given the shorter survival of patients with CHF and CSA, there is an urgent need to screen patients with CHF for sleep disorders and to initiate treatment. ACKNOWLEDGMENT: These data were originally collected in collaboration with Daniel F. Kripke, MD, and Melville R. Klauber, PhD. References 1 Mitler MM, Hajdukovic RM, Shafor R, et al. When people die: cause of death versus time of death. Am J Med 1987; 82: Ancoli-Israel S, Kripke DF, Klauber MR, et al. Morbidity, mortality and sleep disordered breathing in community dwelling elderly. Sleep 1996; 19: Ancoli-Israel S, Kripke DF, Klauber MR, et al. Sleep disordered breathing in community-dwelling elderly. Sleep 1991; 14: Sanner B, Sturm A, Konermann M. Coronary heart disease in patients with obstructive sleep apnea. Dtsch Med Wochenschr 1996; 121: Orr WC. Sleep apnea, hypoxemia, and cardiac arrhythmias. Chest 1986; 89:1 2 6 Shepard JW, Garrison MW, Grither DA, et al. Relationship of ventricular ectopy to oxyhemoglobin desaturation in patients with obstructive sleep apnea. Chest 1985; 88: Guilleminault C, Connolly SJ, Winkle RA. Cardiac arrhythmia and conduction disturbances during sleep in 400 patients with sleep apnea syndrome. Am J Cardiol 1983; 52: Williams AJ, Houston D, Finberg S, et al. Sleep apnea syndrome and essential hypertension. Am J Cardiol 1985; 55: Schafer H, Koehler U, Hasper E, et al. Sleep apnea and cardiovascular risk. Z Kardiol 1995; 84: Clinical Investigations

6 10 Lafaso R, Verschueren P, Rande JLD, et al. Prevalence of sleep disorder breathing in patients on a heart transplant waiting list. Chest 1994; 116: Chan J, Sanderson J, Chan JW, et al. Prevalence of sleepdisordered breathing in diastolic heart failure. Chest 1997; 111: Javaheri S, Parker TJ, Wexler L, et al. Occult sleep-disordered breathing in stable congestive heart failure. Ann Intern Med 1995; 122: Javaheri S, Parker TJ, Liming JD, et al. Sleep apnea in 81 ambulatory male patients with stable heart failure: types and their prevalences, consequences, and presentations. Circulation 1998; 97: Peppard P, Young T, Palta M, et al. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000; 342: Nieto FJ, Young T, Lind B, et al. Association of sleepdisordered breathing, sleep apnea, and hypertension in a large community-based study. JAMA 2000; 283: Takasaki Y, Orr D, Popkin J, et al. Effect of nasal continuous positive airway pressure on sleep apnea in congestive heart failure. Am Rev Respir Dis 1989; 140: Naughton MT, Liu PP, Bernard DC, et al. Treatment of congestive heart failure and Cheyne-Stokes respiration during sleep by continuous positive airway pressure. Am J Respir Crit Care Med 1995; 151: Yan AT, Bradley TD, Liu PP. The role of continuous positive airway pressure in the treatment of congestive heart failure. Chest 2001; 120: Sin DD, Logan AG, Fitzgerald FS, et al. Effects of continuous positive airway pressure on cardiovascular outcomes in heart failure patients with and without Cheyne-Stokes respiration. Circulation 2000; 102: Bradley TD. Continuous positive airway pressure for congestive heart failure. Can Med Assoc J 2000; 162: Tkacova R, Rankin F, Fitzgerald FS, et al. Effects of continuous positive airway pressure on obstructive sleep apnea and left ventricular afterload in patients with heart failure. Circulation 1998; 98: Peker Y, Hedner J, Norum J, et al. Increased incidence of cardiovascular disease in middle-aged men with obstructive sleep apnea: a 7-year follow-up. Am J Res Crit Care Med 2002; 166: Naughton MT, Bradley TD. Sleep apnea in congestive heart failure. Clin Chest Med 1998; 19: Noda A, Okada T, Yasuma F, et al. Cardiac hypertrophy in obstructive sleep apnea syndrome. Chest 1995; 107: Massie BM, Conway M. Survival of patients with congestive heart failure: past, present and future prospects. Circulation 1987; 75(suppl 4): Hanly PJ, Zuberi-Khokhar NS. Increased mortality associated with Cheyne-Stokes respiration in patients with congestive heart failure. Am J Respir Crit Care Med 1996; 153: Findley LJ, Zwillich CW, Ancoli-Israel S, et al. Cheyne- Stokes breathing during sleep in patients with left ventricular heart failure. South Med J 1985; 78: Lanfranchi P, Braghiroli A, Bosimini E, et al. Prognostic value of nocturnal Cheyne-Stokes respiration in chronic heart failure. Circulation 1999; 99: Andreas S, Hagenah G, Moller C, et al. Cheyne-Stokes respiration and prognosis in congestive heart failure. Am J Cardiol 1996; 78: Ancoli-Israel S, Kripke DF, Mason W, et al. Comparisons of home sleep recordings and polysomnograms in older adults with sleep disorders. Sleep 1981; 4: Timms RM, Dawson A, Taft R, et al. Oxygen saturation by oximetry: analysis by microcomputer. J Polysomnographic Technol 1988; Spring: Page WF, Gee SC, Kuntz AJ. Protocol for the Vietnam Veteran Mortality Study. Washington, DC: Office of Reports and Statistics, VA, 1983; Hanly PJ, Millar TW, Steljes DG, et al. Respiration and abnormal sleep in patients with congestive heart failure. Chest 1989; 96: Young T, Peppard P. Sleep-disordered breathing and cardiovascular disease: epidemiologic evidence for a relationship. Sleep 2000; 23(suppl 4):S122 S Somers VK. To sleep, perchance to breathe: implications for the failing heart. Am J Respir Crit Care Med 1999; 160: Javaheri S. Central sleep apnea-hypopnea syndrome in heart failure: prevalence, impact, and treatment. Sleep 1996; 19(10 Suppl):S229 S Andreas S. Central sleep apnea and chronic heart failure. Sleep 2000; 23: Sin DD, Fitzgerald F, Parker JD, et al. Risk factors for central and obstructive sleep apnea in 450 men and women with congestive heart failure. Am J Respir Crit Care Med 1999; 160: 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: Tkacova R, Liu P, Naughton MT, et al. Effect of continuous positive airway pressure on mitral regurgitant fraction and atrial natriuretic peptide in patients with heart failure. J Am Coll Cardiol 1997; 30: Naughton MT, Benard DC, Liu PP, et al. Effects of nasal CPAP on sympathetic activity in patients with heart failure and central sleep apnea. Am J Respir Crit Care Med 1995; 152: CHEST / 124 / 4/ OCTOBER,