January - March 2005: 
Volume 18, Issue 1

Click on the image to download the Issue in PDF format.


Sleep apnea and congestive heartfailure Are we heading towards the end of a vicious circle?
Congestive heartfailure (CHF) and sleep apnea-hypopnea syndrome (SAHS) are two conditions highly prevalent in the general population that often co-exist in the same patient. Patients with CHF often present disordered breathing during sleep that is characterized by episodes of central or obstructive apneas, whereas patients with SAHS suffer significant cardiovascular sequelae. The pathophysiological interactions between the two conditions involve both mechanical and neural effects, implicating the sympathetic nervous system. The application of non-invasive ventilation for the treatment of SAHS has beneficial effects on the concurrent CHF, providing new therapeutic horizons for this syndrome. Pneumon 2005,18(1):26-33.
Full text


Despite recent advance in the treatment of heart failure, the incidence of this condition remains high; furthermore, it is associated with high mortality. It has been demonstrated that patients with heart failure (HF) often suffer disordered breathing during sleep in the form of episodes apnea or hypopnea. Disordered breathing is associated with hypoxia (leading to an imbalance between oxygen demand and oxygen supply to the myocardium), as well as with arousals that stimulate the activity of the sympathetic nervous system (with prominent effects on blood pressure and heart rate). The pathophysiological effects of sleep disordered breathing, which have not as yet been defined as a precipitating factor for the progression of HF or as a consequence of HF, are likely to contribute to the morbidity and mortality of patients with cardiovascular disease.


Obstructive apneas (or hypopneas) are caused by complete (or partial) collapse of pharyngeal walls during sleep,1 whereas central apneas and hypopneas result from reductions in central respiratory drive.2 The mainterms that appear in the text are defined in Table 1.


Congestive heart failure (CHF) affects 1.5-2% of the general population3 and is associated with high morbidity and mortality; from the time of diagnosis the mean life expectancy is 1.7 years in males and 3.2 years in females, as shown in the Framingham Heart Study.4 Despite earlier diagnosis the mean life expectancy and efforts for intensive management, the prognosis of the condition has remained unchanged.5 Sleep apnea-hypopnea syndrome (SAHS) is a common disorder that affects 9- 14% of males and 4-7% of females, according to available reports.6 The incidence of central sleep apnea in the general population, has not been estimated but it appears to be definitely lower compared to that of obstructive sleep apnea.7 The larger study of patients with cardiovascular disease and sleep disordered breathing, the Sleep Heart Health Study, showed that patients with heart failure had a 2.38 times greater risk to present obstructive sleep apnea.8 Similar results were reported in the two larger series of patients with heart failure found in the literature: a prospective study of 81 patients with CHF showed that 40% of the patients had central sleep apnea (CSA) and 11% had obstructive sleep apnea (OSA),9 whereas a larger retrospective series with 450 patients with HF reported that 33% of the patients has CSA and 37% had OSA.10 The latter study indicated that the main risk factors for OSA were obesity in males and older age in females, whereas risk factors for CSA included male gender, hypocapnia when awake, atrial fibrillation and age, with obesity having no implication in the occurrence of CSA.10 Interestingly, however, despite the above described evidence, current guidelines for the management of heart failure refer exclusively to the waking state,11 overlooking the effects of sleep disordered breathing and the potential for treating such a condition. Nevertheless, in recent years, the attention of the investigators has been drawn to the potential role of sleep disordered breathing in pathogenesis of cardiovascular disease8,12,13 and, in particular, congestive heart failure.14,15


SAHS has been associated with systemic hypertension, left ventricular dysfunction, heart rate abnormalities, pulmonary heart disease, myocardial ischemia and sudden death of cardiovascular origin.13,16-18 These effects are more pronounced in patients with underlying coronary artery disease or cardiomyopathy, whereas OSA itself constitutes an independent risk factor for the development of hypertension.19 The fundamental mechanisms of interaction between SAHS and CHF are discussed below.

Pathophysiological effects of OSA on the cardiovascular system

Obstructive apneas are characterized by shallow inspiratory efforts that, due to the pharyngeal occlusion, lead to an abrupt reduction in intrathoracic pressure. This, in turn, increases left ventricular transmural pressure (the difference between intracardiac pressure and intrathoracic pressure) and hence afterload.20 In addition, it enhances venous return and, thus, leads to right ventricular distention and a leftward shift of the intraventricular septum.21 The resulting reduction in preload in combination with the augmented afterload of the left ventricle produce a reduction in stroke volume.22 Furthermore, apnea suppresses the inhibitory action of lung stretch receptors to the sympathetic nervous system, thus augmenting the sympathetic nervous system activity. The ensuing hypoxia and hypercapnia further augment sympathetic activity by stimulating respective chemoreceptors. Sympathetic activity increases heart rate and peripheral vascular resistance, resulting in increased systemic blood pressure. The combination of increased left ventricular afterload and increased heart rate increases myocardial O2 demand in the presence of reduced myocardial O2 supply. This imbalance predisposes to myocardial ischemia and arrhythmias, and, in the long-term, to left ventricular hypertrophy and, ultimately, failure. These changes further reduce stroke volume and perpetuate the vicious circle of the described pathophysiological interactions.

Pathophysiological effects of OSA on the cardiovascular system

Heart failure entails increased left ventricular filling pressure leading to chronic congestion in the pulmonary circulation, which, in turn, stimulates lung vagal receptors. As a consequence, patients with heart failure develop chronic hyperventilation.23 This effect is precipitated in the recumbent position during sleep. The occurrence of arousals enhances hyperventilation (hyperpnea) in the nighttime, resulting in reduced PaCO2. If PaCO2 falls below the threshold for the stimulation of ventilation, central respiratory drive is abrogated and a central apnea is triggered.7 This episode of apnea is sustained until PaCO2 in the respiratory center increases above the threshold for ventilation.2 Central sleep apneas (CSA), in contrast with OSA, are not associated with abrupt reductions in intrathoracic pressure, with the relevant consequences in the cardiovascular system. Conversely, in both CSA and OSA, apneas, hypoxia, CO2 retention and subsequent arousals cause periodically augmented sympathetic nervous system activity. Thus, the resultant increases in blood pressure and heart rate increase myocardial O2 demand, but not O2 supply. The ultimate consequence is, once more, a pathophysiological vicious circle. 14 Recurrent central apneas in patients with CHF often have the form of periodic breathing or/and Cheyne- Stokes respiration (CSR), which is considered a typical finding in these patients.


The clinical characteristics of OSA in patients with heart failure are similar to those seen in patients with normal left ventricular function. Typically, patients are male, overweight, with a history of snoring.15 An interesting difference that the clinician should bear in mind, is that most of these patients do not develop the subjective feeling of daytime sleepiness.9,24 Similarly, despite episodes of paroxysmal nocturnal dyspnea, patients with CSA do not report excessive snoring or daytime sleepiness. 9 Many patients with CHF have episodes of both obstructive and central sleep apneas. A progressive transition from mainly obstructive apneas at the beginning of the night to mainly central apneas toward its end has been described in one study.25 This study emphasizes the potential role of breathing disorders to the deterioration of heart function during sleep in these patients.


The classic management for obstructive sleep apnea involves the application of continuous positive airway pressure (CPAP) via a nasal mask (nasal CPAP, nCPAP). 1,8 Alternatively, a nasal-oral or full face mask may be used. The first report of OSA management using CPAP comes from Sullivan et al,26 whereas the application of CPAP in patients with OSA and HF was first reported in 1989 by the team of T. Douglas Bradley in Toronto, Canada.27 The application of CPAP in patients with OSA prevents occlusion of upper airways during sleep, thus disrupting the primary pathophysiological mechanism that generates the syndrome. The benefits of applying CPAP in patients with CHF derive from three potential effects of CPAP: (i) CPAP increases intrathoracic pressure and reduces left ventricular transmural pressure (afterload), thus enhancing cardiac output;28 (ii) in patients with CHF and OSA or CSA, CPAP restrains nocturnal fluctuations of blood pressure by abolishing apneas; and (iii) CPAP reduces sympathetic nervous system activity by eliminating arousals and apnea-related hypoxia.


The utility of CPAP in patients with stable CHF and OSA has been demonstrated in many studies. In a study of patients with CHF and OSA, left ventricular afterload was found to increase during transition from wakefulness to sleep stage 2.29 Effective abolition of obstructive apneas using CPAP led to reduced left ventricular afterload and heart rate, as well as reduced workload of inspiratory muscles in these patients. The same research group found that the application of CPAP evoked a significant reduction in both systolic and diastolic arterial pressure, as well as heart rate.30 The investigators showed that these effects of CPAP on autonomous nervous system are mediated by increased baroreceptor reflex sensitivity. A randomized prospective study showed that patients with CHF and moderate to severe OSA who were effectively managed with CPAP for a month had increased ejection fraction and improved left ventricular function.31 Even more appealing is the finding that the beneficial effects of CPAP are preserved after its discontinuation. In addition, its favorable effects are usually experienced immediately after its first application, even from the first night. In particular, a prospective study of patients with heart failure and obstructive or central sleep apneas showed that abolition of apneas by CPAP resulted in a prompt reduction in ventricular arrhythmias already from the first night of CPAP application.32

The use of CPAP appears to also limit arterial hypertension, one of the primary determinants for the progression to heart failure in patients with OSA. A recent randomized parallel group study examined the effect of nasal CPAP in therapeutic and sub-therapeutic levels on the arterial pressure of patients with OSA.33 Patients who received therapeutic levels of CPAP experienced a significant reduction in both daytime and nighttime blood pressure, as well as improved daytime sleepiness and quality of life. These effects were more profound in patients with more severe OSA.


The use of CPAP appears to improve both central apneas and periodic breathing during sleep in patients with CHF, even though the implicated mechanisms have not been clearly defined yet.19 The first randomized controlled studies of CPAP administration for 1-3 months in patients with heart failure showed significant limitation of Cheyne-Stokes respiration, as well as improved oxygenation and less arousals during the night.34-36 In a more recent prospective randomized study, the use of CPAP was found to be more beneficial for those patients with HF and co-existing CSA-CSR compared to patients who did not experience periodic breathing.37 Patients with CHF and CSA-CSR had significantly improved left ventricular ejection fraction after 3 months of CPAP therapy, as well as improved overall survival after 2.2 years of follow-up. The positive effect of CPAP on periodic breathing in patients with CHF appears to result from the increase in PaCO2 above apneic threshold. A proposed mechanism for the explanation of this effect involves the reduction of minute ventilation, due to the redistribution of intravascular volume away from the lungs that reduces the stimulation of lung vagal receptors and, ultimately, increases PaCO2. Furthermore, it has been shown that in patients with CSA some upper airway obstruction during sleep also occurs,38 but is potentially reversible with the use of positive airway pressure.

The effect of CPAP on central apneas and periodic breathing is apparently less prompt compared to its effect on patients with OSA. In early studies of patients with CHF and CSR, short-term CPAP therapy (for up to 2 weeks) did not improve CSR.39,40 Furthermore, administration of CPAP in patients with CSA and CHF did not limit the number of central apneas during the first night of its application.32 A possible explanation is that CPAP primarily improves heart function and, then, through this improvement, produces a secondary reduction in central apneas and periodic breathing. It has also been found that abolition of CSA in patients with left ventricular systolic dysfunction requires high CPAP levels (9-16.5 cm H2O),32 which were not administered in most of the above-mentioned studies. Hence, severe CSA is probably a determinant of non-response to CPAP therapy in some patients with heart failure. To define the role of CPAP therapy in patients with heart failure and CSA a large 5-year follow-up prospective multicenter study, the Canadian Positive Airway Pressure (CANPAP) trial, is currently in progress; the first results of the study are expected in about 2 years.42


The administration of CPAP in patients with heart failure is associated with some drawbacks. The potential adverse effects of CPAP therapy include reduced cardiac output and hypotension. These effects may be particularly profound with the use of high levels of CPAP, such as those required to abolish Cheyne-Stokes respiration, as mentioned above. Patients with CHF who may have reduced intravascular volume or/and take vasodilators or â-blockers, may find it extremely difficult to compensate for the abrupt increase in intrathoracic pressure evoked by high CPAP levels, and are thus more susceptible to the adverse effects of CPAP therapy.

A potential solution to the problem of administering the "optimal" CPAP level in patients with heart failure and CSA-CSR may be the relatively new method of Adaptive Pressure Support Servo-Ventilation, delivered by the Autoset CS device (ResMed, Sydney, Australia). This device uses a specially designed algorithm to achieve the hemodynamic benefits of CPAP in patients with heart failure in concert with the prompt abolition of CSA-CSR and prevention of hyperventilation and hypocapnia by the ventilator. When the patient breathes normally, the device provides a basic ventilatory support with a CPAP of 5 cm H2O; when respiratory efforts diminish or stop (as is the case in central apnea), it provides an additional variable inspiratory pressure (usually between 3-10 cm H2O) aiming to maintain a ventilation of about 90% of the average patient ventilation. When the patient resumes respiratory efforts, ventilatory assistance is reduced. In addition, in the event of apnea, the device provides a basic respiratory rate of 15 respirations per minute. The use of Autoset CS in a population of patients with CSA and CSR has led to a greater reduction in central apneas and CSR, as well as to improved sleep quality compared to the use of "standard" CPAP.43 Particularly interesting is the fact that improvement was reported even from the first day of the use of the device. A more recent randomized controlled double blind study of Autoset CS in patients with CHF and CSR showed a significant reduction in sleep apneas after one month of therapy, as well as improved daytime sleepiness in patients who received therapeutic pressures using this device. 44

Another recent study examined the effect of high frequency atrial pacing (with overdrive) in patients with CSA or OSA.45 This study included patients without CHF who had pacemakers due to symptomatic bradyarrhythmias. In patients who were found to have CSA or OSA, pacing at 15 beats per minute above basic nocturnal sinus rhythm was applied. This type of pacing reduced episodes of apnea. The findings of this research group are particularly interesting, although further studies are required to elucidate the underlying mechanisms for this effect and determine the potential role of such a therapeutic approach in patients with heart failure and heart rate disturbances.14


Despite their high prevalence among patients with heart failure, disorders of breathing during sleep remain underdiagnosed and undertreated in this patient population. Although there are as yet no guidelines concerning the performance of sleep studies (polysomnography) in patients with heart failure, clinicians treating such patients should be alert and aware of the potential coexistence of sleep disordered breathing. In particular, patients with the above-discussed risk factor profile or typical symptoms (snoring, excessive daytime sleepiness, paroxysmal nocturnal dyspnea, headaches in the morning), should be evaluated using polysomnography, in order to examine whether episodes of central or obstructive sleep apnea occur. The appropriate management using either CPAP or newer forms of non-invasive mechanical ventilation may improve disease outcome, life expectancy and quality of life in these patients.