Article

Beyond the Apnoea-Hypopnoea Index - Prognostic Value of Other Elements of Polysomnography to Describe Sleep-disordered Breathing in Heart Failure

Abstract

There is increasing appreciation of the prevalence of sleep-disordered breathing (SDP) in heart failure. As we examine this patient population, the difficulties of determining success in the treatment of SDB are becoming evident. The apnoea–hypopnoea index (AHI) is the standard method of measuring both the severity of the disease and treatment success, but in itself is a composite of multiple components. This article examines both current and developing measurements in the treatment of SDB.

Disclosure: Philip B Adamson has received consulting fees from Medtronic, Inc., St Jude Medical, CardioMems, Inc. and Cardiac Concepts.

Received:

Accepted:

Citation:European Cardiology 2009;5(2):53–6

Correspondence: Philip B Adamson, Oklahoma Foundation for Cardiovascular Research, 4050 W Memorial Road, Oklahoma City, OK 73120, US. E: padamson@ocaheart.com

Open access:

The copyright in this work belongs to Radcliffe Medical Media. Only articles clearly marked with the CC BY-NC logo are published with the Creative Commons by Attribution Licence. The CC BY-NC option was not available for Radcliffe journals before 1 January 2019. Articles marked ‘Open Access’ but not marked ‘CC BY-NC’ are made freely accessible at the time of publication but are subject to standard copyright law regarding reproduction and distribution. Permission is required for reuse of this content.

A growing appreciation of the prevalence of sleep breathing disorders, especially as co-morbidities of disease syndromes such as chronic heart failure, has increased interest in screening, diagnosis and treatment of obstructive and central sleep apnoea. Sleep apnoea diagnosis currently requires an overnight polysomnograhic evaluation in which the disease is defined as the number of times per hour that respiration is inadequate to maintain oxygen saturation. Failure of respiration to maintain oxygen saturation is traditionally quantified using the apnoea–hypopnoea index (AHI), which not only serves to define the presence of treatable sleep apnoea but also provides the basis to measure successful treatment. Although many physiological signals are continuously measured during an overnight sleep study, the AHI is still the most important parameter to establish the presence or absence of significant sleep apnoea.

An apnoea event is defined as cessation of airway flow for greater than 10 seconds and is further classified based on the presence or absence of respiratory effort.1 The event is considered ‘obstructive apnoea’ if respiratory effort continues, without airway flow, for more than 10 seconds. Central apnoea is an absence of airflow without respiratory effort (see Figure 1). However, most of the AHI is composed of hypopnoeas, which are defined as reductions in airflow, usually by 50% or more, with associated reduction in oxyhaemoglobin saturation.1 Critical to disease diagnosis is the ability to accurately detect reduction in airflow, particularly during a hypopnoea event, making it important to understand the methods used to determine airflow. Lack of consensus standards for clinical evaluation of hypopnoea may significantly affect the diagnosis of sleep apnoea and possibly alter assessments of new interventions intended to treat the disease. Further controversy arises when considering the extent of oxyhaemoglobin desaturation that qualifies an event as hypopnoea, with most recent outcome data suggesting that a greater than 4% reduction is required to qualify for a true hypopnoea. Accurate diagnosis and assessment of therapy is very important as it is clear that obstructive sleep apnoea is associated with significantly increased risk of cardiovascular diseases such as heart failure, hypertension, oxidative stress, myocardial infarction, atrial and ventricular arrhythmias and death.

The aim of this article is to understand the importance of other physiological components of the polysomnogram and propose alternative means to measure the severity of pathological sleep disruption, and also to propose additional ways to determine the efficacy of traditional and novel interventions intended to treat the disease syndrome.

Apnoea–Hypopnoea Index

The AHI is the standard method for defining obstructive and central sleep apnoea and is the most common parameter used to establish the consequences of the disease and outcomes of treatment. It is also used to define the severity of the disease from mild (AHI five to 15 events/hour), to moderate (AHI 15–30 events/hour), to severe (AHI >30 events/hour). Obstructive sleep apnoea, defined using the AHI, has a strong association with hypertension, and successful continuous positive airway pressure (CPAP) therapy lowers systolic blood pressure by approximately 10mmHg on average.2 This degree of blood pressure lowering in primary hypertension trials is associated with significantly lower risk of stroke and myocardial infarction.3 This would suggest that similar reductions in such cardiovascular end-points may be seen with successful sleep apnoea treatment, but prospective trials using these end-points are not available. Ruttanaumpawan et al. showed that although AHI was significantly improved in patients with heart failure, sleep efficiency and arousal did not improve.4 Other polysomnographic parameters have even less clear association with outcomes, or have never been tested.

However, difficulties exist in using the AHI as the primary sleep apnoea marker, as methodologies differ, including how airflow is actually measured (thermistor versus nasal pressure) and what constitutes hypopnoea. These differences are not trivial. The most common airflow measurement technique uses thermistor detection using a change in temperature to determine nasal airflow. This is very sensitive for apnoea events, but is more problematic for hypopnoeas. Since hypopnoeas comprise most of the AHI, significant differences in detection are possible when thermistors are used. Nasal pressure measurements are more robust to detect both apnoea and hypopnoea events, but are not consistently used clinically.

Defining hypopnoea is also variable. Even consensus recommendations from the American Academy of Sleep Medicine (AASM) include two definitions that may constitute ‘hypopnoea’.1 A recent study demonstrated that the recommended definitions for hypopnoea significantly altered the estimation of sleep apnoea in the same group of patients referred for sleep study by up to 40%.5 Therefore, the way airflow is measured and the actual definition of hypopnoea may significantly alter the diagnosis of sleep apnoea. This makes it very difficult to accurately estimate the prevalence of the disease in a given population and may hamper comparison of therapies, whose success may be measured by the AHI. Standardisation in clinical trials may overcome some of these difficulties, but lack of clear consensus in the clinical community may alter appropriate detection and therapeutic intervention for obstructive and central sleep apnoea.

Further disadvantages of using the AHI alone may be an inaccurate assessment of the degree of systemic impact of the disease. Figure 2 is a simplified illustration of many of the pathophysiological aspects of obstructive and central sleep apnoea. Many of the adverse effects of sleep apnoea occur as a result of oxygen desaturation, which directly activates the sympathetic nervous system and renin–angiotensin system and alters chemoreceptor sensitivity. Sympathetic nervous input to peripheral chemoreceptors, compounded by decreased cerebrovascular CO2 reactivity, already present in patients with heart failure, render central breathing control unstable, leading to Cheyne–Stokes respiration patterns and central sleep apnoea, which is much more common in heart failure patients.6 To complicate diagnosis and treatment assessments further, many patients with chronic heart failure spontaneously alternate the predominant form of sleep breathing disorder between a central and obstructive mechanism.

While the AHI may quantify the end result of these systemic events, other markers may provide important insights into the true pathophysiology and mechanisms responsible for disordered breathing. The AHI will certainly remain the mainstay of defining sleep apnoea syndromes, but attention to alternative markers routinely assessed during polysomnography may provide useful information to judge the potential clinical benefit of novel interventions that may provide different approaches to treatment than just airway support. Most of the alternative markers have no outcome-driven evidence base for use, but have direct correlation to sleep apnoea pathophysiology.

Oxygen Desaturation

Oxygen desaturation is the obvious element of inadequate respiration that likely leads to systemic pathology. Therefore, an alternative marker of disease severity is the time spent with an oxygen saturation level below 90%. This marker is mostly linked to the AHI, but theoretically an intervention that converts an apnoea event to a hypopnoea may result in better oxygen saturation, but may not change the AHI. Improving oxygen saturation and decreasing the time exposure to desaturation are both theoretically beneficial and may favourably alter the underlying mechanisms responsible for adverse outcomes in patients with this disease syndrome. Time spent with an oxygen saturation level below 90% may be adjunctive to the AHI to improve an assessment of sleep apnoea pathophysiology. However, supplemental oxygen alone has not been shown to improve survival in patients with sleep apnoea.

Oxygen desaturation is often measured as the oxygen desaturation index, which is the number of times that the oxygen level drops by three or four percentage points per hour. While a measure of oxygenation, it is usually directly correlated with AHI and may not offer additional insight to the AHI.7 However, it is possible that new interventions designed to treat either central or obstructive sleep apnea may have an impact on overall oxygen saturation with an avoidance of significant desaturations. The end-point of reducing the deoxygenation burden improves the ‘lesion’ of sleep apnoea pathophysiology and should be considered an important effect, even if it is independent of the AHI.

Sleep Efficiency

An intervention that increases sleep efficiency, defined as sleep latency as well as sleep duration, coupled with improvement in sleep architecture may improve clinical outcomes. Short sleep duration predicted the risk of hypertension, cardiovascular disease and type II diabetes in the Sleep Heart Health Study.8,9 Sleep duration may alter sleep stages achieved, which may further alter cardiac and systemic autonomic control, likely resulting in increased sympathetic activation with vagal withdrawal.10 Sympathetic activation leads to renin–angiotensin activation and significantly changes vasomotor control, leading to hypertension and other cardiovascular diseases (see Figure 2).

As novel therapies emerge, it will be important to assess the impact on sleep efficiency, including increased sleep time and acquisition of appropriate sleep stages. This effect would likely be reflected in the AHI as fewer arousals are expected, but theoretically could alter sleep physiology without significant changes in the AHI. Again, if apnoea is prevented but replaced by hypopnoea, an alteration in the AHI would not be seen, but improved sleep efficiency might be present. Improved sleep efficiency may also affect other symptoms of sleep apnoea, including daytime sleepiness and fatigue.

Heart Rate Variability

An important aspect of sleep apnoea pathophysiology is chronic sympathetic nervous activation to levels above those seen in individuals without disturbed sleep. The cascade of problems that develop with sympathetic activation include systemic hypertension, increased risk of ventricular arrhythmias, adverse ventricular and vascular remodelling and renin–angiotensin activation. Changes in autonomic activity can be measured directly from nerve fibre recordings or indirectly from analysis of heart rate variability. Heart rate variability, especially using the spectral analysis approach, has the benefit of providing complete autonomic information to include cardiac parasympathetic influences.

The importance of cardiac events during sleep apnoea was recently recognised by the AASM with the recommendation to include electrocardiogram (ECG) recordings as standard data in a polysomnogram. Processing of the ECG signal allows quantification of heart rate variability and ventricular and atrial arrhythmias during apnoeic periods or desaturation events. Analysing and quantifying information from ECG recordings during polysomnography has not routinely been included in previous sleep apnoea studies, but may provide an excellent means to judge successful treatment interventions. Specifically, the variability in heart period is directly correlated with vagal control of the heart. Heart rate variability can be measured in so-called ‘time–domain’ methods, such as standard deviation of the inter-beat (RR) intervals, or ‘frequency–domain’ measurements, which identify periodic changes in heart rate. Frequency–domain measurements can be applied to shorter segments of heart rate recordings and offer a more detailed view of specific input to the heart from both the parasympathetic nervous system, which exerts high-frequency influences (~0.15Hz), and the sympathetic nervous system, which exerts lower-frequency signals (>0.2Hz). Heart rate variability measurements in either domain correlate with improved morbidity and mortality in heart failure patients.10 Changes in cardiac autonomic control occur naturally during the transition from non-rapid eye movement (REM) to REM sleep stages and can be quantified by using heart rate variability analyses.11 Myocardial infarction changes the autonomic characteristics of discrete sleep stages11 and would be expected to have further derangement if sleep apnoea and other cardiovascular diseases are present. Cardiac autonomic control is important and can be used to evaluate existing and novel therapies for sleep apnoea. Reducing sympathetic activation and vagal withdrawal would be expected to favourably influence prognosis in any cardiovascular disease, especially in the cardiovascular complications associated with sleep apnoea.

Arrhythmias

Sleep apnoea has been associated with both atrial and ventricular arrhythmias. Ventricular arrhythmias have been reported in up to 66% of patients with sleep-disordered breathing.12 This is very important, as the presence of sustained ventricular tachycardia in heart failure patients correlates with mortality and leads to consideration of implantable cardioverter–defibrillator placement. Additionally, atrial arrhythmias, such as atrial fibrillation and flutter, are common co-morbidities found in patients with heart failure. The mechanisms of atrial arrhythmias in heart failure are complex, but the presence of sleep-disordered breathing correlates with a higher risk of atrial rhythm disturbances.

This is probably true for both systolic and diastolic heart failure patients. Atrial arrhythmias interrupt atrial contribution to ventricular filling and may cause worsening heart failure symptoms. Atrial rhythm disorders also increase the risk of cerebrovascular accidents, which often lead to the clinical recommendation for chronic anticoagulation. Both obstructive and central apnoea have been associated with atrial fibrillation.13,14 Atrial fibrillation is associated with increased morbidity and mortality in heart failure, although it is not clear whether decreasing the burden of atrial fibrillation improves morbidity in heart failure. Quantifying atrial and ventricular arrhythmias during sleep studies is very important and should be considered when evaluating therapies.

Mortality

With any disease process, mortality is a significant end-point with both the course of the disease and any treatment. Sleep apnoea, defined as an AHI greater than five, is associated with increased mortality.15 Patients with sleep-disordered breathing are more likely to have an increase in cardiovascular disease, including coronary artery disease and stroke.16,17 Although mortality trials are difficult and expensive, certainly any treatment of sleep-disordered breathing that improved mortality would be accepted above any other outcome measurement available.

Conclusions

Sleep breathing disorders in patients with chronic heart failure represent an important source of pathology that theoretically promotes adverse ventricular remodelling leading to progressive heart failure and death. Clarity of diagnosis is essential in order to appropriately offer therapies that may attenuate the effects of obstructive and central sleep apnoea. Lack of consensus about the definition of what constitutes a hypopnoeic event, coupled with difficulties in measuring airflow, makes comparison of the AHI across clinical trials difficult. Theoretically, the AHI alone may not be the best means to judge the effect of novel therapies under development. Changes in other parameters that directly link to the pathophysiology of heart disease should be considered as important end-points, which may at least develop hypotheses about an impact on outcomes such as morbidity and mortality.

References

  1. Iber C, ed., The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specification. Westchester III: The American Academy of Sleep Medicine, 2007.
  2. Becker HF, et al., Effect of nasal continuous positive airway pressure treatment on blood pressure in patients in obstructive sleep apnea, Sleep, 2003;107(1):68–73.
    Crossref | PubMed
  3. MacMahon S, et al., Blood pressure, stroke, and coronary heart disease. Part 1. Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias, Lancet, 1990;335:765–74.
    Crossref | PubMed
  4. Ruttanaumpawan P, Logan AG, Floras JS, et al., Effect of continuous chronic airway pressure on sleep structure in heart failure patients with central sleep apnea, Sleep, 2009;32:91–8.
    PubMed
  5. Ruehland WR, Rochford PD, O’Donoghue FJ, et al., The new AASM criteria for scoring hypopneas: impact on the apnea hypopnea index, Sleep, 2009;32(2):150–7.
    PubMed
  6. Ainslie PN, Duffin J, Integration of cerebrovascular CO2 reactivity and chemoreflex control of breathing: mechanisms of regulation, measurement, and interpretation, Am J Physiol Regul Integr Comp Physiol, 2009;296:R1473–95.
    Crossref | PubMed
  7. Fietze I, Dingli K, Diefenbach K, et al., Night-to night variation of the oxygen desaturation index in sleep apnoea syndrome, Eur Resp J, 2004;24:987–93.
    Crossref | PubMed
  8. Seicean S, Kirchner HL, Gottlieb DJ, et al., Sleep disordered breathing and impaired glucose metabolism in normal-weight and overweight/obese individuals, Diabetes Care, 2008;31:1001–6.
    Crossref | PubMed
  9. Gottlieb DJ, Redline S, Nieto FJ, et al., Association of usual sleep duration with hypertension: the sleep heart health study, Sleep, 2006;29:1009–14.
    PubMed
  10. Adamson PB, Smith AL, Abraham WT, et al., Continuous autonomic assessment in patients with symptomatic heart failure, Circulation, 2004;110:2389–94.
    Crossref | PubMed
  11. Vanoli E, Adamson PB, Ba-Lin, et al., Heart rate variability during specific sleep stages. A comparison of healthy subjects with patients after myocardial infarction, Circulation, 1995;91:1918–22.
    Crossref | PubMed
  12. Guilleminault C, Connolly SJ, Winkle RA, Cardiac arrhythmia and conduction disturbances during sleep in 400 patients with sleep apnea syndrome, Am J Cardiol, 1983;52:490–94.
    Crossref | PubMed
  13. Gami AS, et al., Association of atrial fibrillation and obstructive sleep apnea, Circulation, 2004;110(4):364–7.
    Crossref | PubMed
  14. Mehra R, Benjamin EJ, Shahar E, et al., Association of nocturnal arrhythmias with sleep-disordered breathing: the sleep heart health study, Am J Respir Crit Care Med, 2006:173:910–16.
    Crossref | PubMed
  15. Javaheri S, Shukla R, Zeigler H, 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, 2007;49(20):2028–34.
    Crossref | PubMed
  16. Valham F, Mooe T, Rabben T, et al., Increased risk of stroke in patients with coronary artery disease and sleep apnea, Circulation, 2008;188:955–60.
    Crossref | PubMed
  17. Peker Y, Kraiczi H, Hedner J, et al., An independent association between obstructive sleep apnea and coronary artery disease, Eur Resp J, 1999;14:179–84.
    Crossref | PubMed