ReviewMeasuring therapeutic efficacy in the treatment of central sleep apnoea in patients with heart failure
Introduction
Approximately 50% of heart failure patients have sleeping disorders and recognition of the importance of these co-morbidities has led to inclusion of sleep in cardiology guidelines worldwide [1], [2]. Understanding which patients are at risk and how to identify them is of growing concern in the care of these patients. Sleep medicine has developed into a distinct specialty with specialist physicians, dedicated clinics and specialist societies publishing guidelines related to the diagnosis, assessment and management of affected patients [3], [4]. The spectrum of sleep disordered breathing (SDB) includes two major subtypes, obstructive sleep apnoea (OSA) and central sleep apnoea (CSA); many patients have some component of each disorder. With 10% of the adult male population diagnosed with OSA, this form of SDB alone makes up the majority of patients seen in sleep clinics [4]. Thus, only a small proportion of patients seen within the sleep clinic are diagnosed with CSA [3]. In contrast, approximately 1/3 of patients with heart failure have CSA [5]. These patients have higher morbidity and mortality than their counterparts without SDB [6]. As most of these patients are often seen within the cardiology practice identification and treatment of these patients should be a focus of cardiologists. In addition, a focus on developing relationships between sleep and HF specialists is important to improve the care of these patients. Cardiologists therefore are presented with the problem of how to manage a condition (CSA) which is epidemiologically largely secondary to CHF whose patients attend specialist heart failure services, but which is usually identified and treated by a separate specialist community (Respiratory Sleep Physicians) whose more usual patient has the OSA type of SDB. It is therefore crucial to do two things: to differentiate between OSA and CSA and to educate CHF specialists in how to identify, assess and manage CSA as they have the patients who are more likely to have the CSA pattern of disorder.
It may seem surprising that so few patients with CSA are typically identified in sleep clinics. However, the typical presentation of sleep apnea is absent in patients with heart failure. Patients with heart failure typically deny daytime sleepiness and questionnaires typically used with healthy patients are not predictive in patients with HF. Recently a sleep apnea “score” was developed to identify heart failure patients at high risk for sleep apnea. This test uses information commonly found in a patient's chart such as NYHA class, age and gender to identify risk and trigger testing for the patients [7].
Since 1985 when a very small study first suggested a high prevalence of CSA in CHF patients and a heightened risk of death associated with CSA, multiple studies have confirmed the importance of CSA on prognosis in CHF [8], [9], [10], [11]. An AHI above a critical value of > 30 episodes per hour has been repeatedly shown to be a powerful marker of adverse outcome in SDB. A recent meta-analysis has confirmed that CSA confers an increased mortality outcome on CHF patients (comparing an AHI of less than compared to more than 30/h) whereas no significant effects were seen for OSA in the CHF population [12]. Eleven studies were identified recruiting 1944 patients (1399 in the SDB group and 545 in the no-SDB group). Patients with SDB showed significantly increased all-cause mortality compared to controls [RR 1.66 (1.19–2.31). This was driven almost entirely by an increased risk associated with CSA [RR 1.48 (1.15–1.91) with no significant effect of OSA being seen.
We need to evaluate treatment aims for CSA both because it is a high-risk co-morbidity of CHF but also because of the added symptom burden of sleep-disorder in CHF. To do so we need to evaluate markers of treatment efficacy short of requiring large scale mortality trials for every new treatment modality and variant. This article will review the efficacy metrics we should utilise in measuring “success” in treating CSA in HF patients as it is clearly not feasible to evaluate mortality effects for every co-morbidity affecting HF patients.
Section snippets
Differentiating CSA and OSA
There are many detailed reviews of the differing underlying pathophysiology of OSA and CSA; to briefly summarise, OSA is caused by upper airway obstruction, usually associated with upper airway collapse and CSA by fluctuating central respiratory drive [13], [14], [15]. OSA manifests as intermittent airway occlusions, accompanied by consequent loss of airflow, arterial deoxygenation and arousal due to the stimulatory effects of chemoreflex firing secondary to the disturbed arterial blood gases.
What efficacy metric should be used to measure “success” in treating CSA in HF patients?
The goal of treating SDB has traditionally focused on improving daytime sleepiness and fatigue [4]. In HF patients with SDB, this is not as easy to ascertain as their symptoms overlap with HF [7]. Thus, improvement in treating SDB in HF patients must focus more on overall quality of life. Over the past 5 years, there has been a shift in sleep medicine from only improving symptoms in SDB, to preventing the long term consequences [16]. Much of the harm associated with SDB (increased CV risk,
Disease metrics
The traditional metric used to look at disease severity in CSA is the apnoea/hypopnea index or AHI. AHI measures the average number of disordered breathing events (apnoeas of at least 10 s or hypopnoeas that lead to at least a 4% oxygen desaturation) per hour. Typically the severity of both CSA and OSA were defined based on the number of episodes per hour with < 5 normal, 5–15 per hour classified as mild, 16–30 events per hour classified as moderate and > 30 severe [17]. In OSA, it has been
Oxygenation
Another variant on total apnoeic burden may be to consider the effect of apnoeas on arterial oxygenation and consider the total integrated hypoxaemic burden during sleep. This measure would be quite logical, for many of the adverse consequences of an apnoeic episode are the result of the combined effects of hypoxia on critical organs and tissues (including the myocardium) coincident with the surge in sympathetic drive that occurs reflexly through hypoxia-driven chemoreflex induced
Arousals/sympathetic activation
Sympathetic arousals and the consequent cerebral arousal that frequently accompanies it are additional important metrics as this contributes to overall sympathetic activation and downstream effects of sleep disordered breathing. Measures such as single nerve sympathetic nerve firing or integrated measures of sympathetic activity (such as radiolabelled norepinephrine spill over) could be measured, but as these are largely in the realms of special laboratory research tests, they would have little
Heart rate variability (HRV)
Sympathetic/parasympathetic imbalance, despite being an important complication of both SDB and CHF and being of major prognostic significance, is not something that is easy to measure or summarise in patients. Indirect measures need to be employed. The most popular of these in clinical use has been heart rate variability (HRV). HRV can be measured simply as a measure of dispersion of RR intervals, such as SDNN extracted by processing from 24 h Holter ECG recordings or can be measured in more
Quality of life
CHF and CSA are both conditions associated with a considerable symptom burden, including dyspnoea, fatigue, sleepiness, depression and reduced mental acuity. Improvement in such symptoms is valued highly by patients, and effective therapy of CSA and other types of SDB has been established to improve symptoms in several trials [19], [31], [32]. Many patients' value improvements in quality of life over prolonged life and it is important to keep this in mind when treating co-morbidities such as
Exercise tolerance
Exercise tolerance is one of the hallmarks of the effects of CHF on patients and there is some evidence that SDB can further impair exercise tolerance in patients. It is also one of the most common complaints of sufferers of both conditions and is tied to improvements in quality of life. Improving exercise capacity would be of very important clinical value, and as such if a treatment for CSA could be shown to improve exercise capacity, objectively measured, then this would be a very valuable
Morbidity/mortality
A mortality/morbidity reduction has become the main route for approving new treatments designed to improve CHF. Diverse treatments such as ACE inhibitors, beta-blocker, mineralocorticoid receptor antagonists, and CRT/ICD and in selected patients LVAD devices have passed this threshold [3]. Other interventions such as exercise training, disease management programmes and the use of diuretics to treat congestion have been recommended on the basis of symptom improvement or improved exercise
How do you assess safety of a new therapy?
Safety in treating patients with heart failure is of paramount importance. Therapeutic options which are quite safe in the healthy population, such as ibuprofen for mild pain and metformin for diabetes, have been shown to have detrimental effects in patients with heart failure [38], [39]. The adverse effects recently demonstrated with positive airway pressure masks in SERVE-HF make us maintain considerable caution in interpreting results without randomised, controlled trials. “First do no harm”
Risk versus benefit in the treatment of CSA
The adverse effects recently demonstrated with positive airway pressure masks in SERVE-HF make us maintain considerable caution in interpreting early results. It remains unclear why mortality was increased in patients on PAP therapy. More recently developed treatment options for CSA in the CHF population including implantable phrenic nerve stimulators have quite a different mode of action [40]. It has as yet received less study than the older options, but does appear to be able to reduce
Conclusions
Regulators and clinicians alike have been worried about the increasing time it is taking to bring innovative therapies to market and the fact that a treatment delayed too long is at risk of being a treatment forever lost, even if effective. As a result, there are signs that new treatment approval is moving to one of adaptive testing and adaptive licencing where proof of principle and efficacy on certain patient benefits could be established in much smaller trials, leading to early uptake and
Conflict of interest
Dr. Coats reports receiving honoraria from Resmed and Respicardia. Dr. Shewan: None to declare.
Acknowledgements
We declare we have adhered to the statement of ethical publishing as appears in the International Journal of Cardiology [42].
References (42)
- et al.
2013 ACCF/AHA guideline for the management of heart failure
J Am Coll Cardiol
(2013) - et al.
Central sleep apnea, right ventricular dysfunction, and low diastolic blood pressure are predictors of mortality in systolic heart failure
J Am Coll Cardiol
(May 22 2007) - et al.
State-of-the-art review: mechanisms and clinical consequences of untreated central sleep apnea in heart failure
J Am Coll Cardiol
(2015) - et al.
Treatment of obstructive sleep apnea reduces the risk of atrial fibrillation recurrence after catheter ablation
J Am Coll Cardiol
(2013) - 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) - et al.
Heart rate variability in risk stratification of cardiac patients
Prog Cardiovasc Dis
(2013) - et al.
Quantitative evaluation of drug or device effects on ventricular remodeling as predictors of therapeutic effects on mortality in patients with heart failure and reduced ejection fraction: a meta-analytic approach
J Am Coll Cardiol
(2010) - et al.
Phrenic nerve stimulation for the treatment of central sleep apnea
JCHF
(2015) - et al.
Design of the remede system pivotal trial: a prospective randomized study in the use of respiratory rhythm management to treat central sleep apnea
J Card Fail
(2015 Nov) - et al.
A statement on ethical standards in publishing scientific articles in the international journal of cardiology family of journals
Int J Cardiol
(2014)