Heart Rate Variability (HRV): Sleep Disordered Breathing

Abstract

Sleep‐related breathing disorders are a common but often underdiagnosed condition both in adults and in children. It is now evident that abnormal ventilation during sleep can cause a range of adverse physiological sequela including autonomic dysfunction. Extensive research effort over the past two decades has demonstrated that patterns of cyclic variation in the heart rate (HR) or heart rate variability (HRV) are closely linked to partial reduction or complete cessation of airflow during abnormal breathing events (hypopnea or apnea) during sleep. These efforts culminated in the year 2000, when Thomas Penzel from the Hospital of Phillips‐University in Germany and George Moody from MIT, with their collaborators, coordinated a global competition to stimulate rapid research advances in deploying the HRV signal to detect sleep apnea (Computers in Cardiology Apnea Challenge, 2000). The outcome of these efforts have motivated a number of prominent international research groups to use advanced signal processing methods to analyze the HRV signal and to develop dedicated instrumentation or software packages with the objective to detect sleep disordered breathing in symptomatic subjects at an early stage. A comprehensive review of the literature in this field reveals the value, reliability, sensitivity, specificity, and limitations of these methods and devices. Even though the ultimate goal of simplified diagnosis of sleep disordered breathing from HRV signal analysis is not fully realized yet, it is clearly established that the HRV signal carries a wealth of measures that are sensitive to sleep stages, sleep disordered breathing (SDB), and other disorders influencing the cardiovascular system. Thus, careful analysis of the readily accessible electrocardiogram (ECG) and its derived HRV signal, along with new and sophisticated nonlinear signal analysis methods provide the exciting promise that accurate separation of sleep stages and reliable detection of SDB as well as other disorders influencing the autonomic control of the heart is within reach. In this chapter, we first present a simplified and distilled overview of regulation of the HR by the autonomic nervous system (ANS), sleep physiology, and pathophysiology in children and adults. This is followed by a brief review of the HRV signal derivation and processing methods applicable to its analysis for detection of SDB. We conclude the chapter with an example graphical user interface and consider some issues associated with development of integrated software packages to facilitate the simplified detection of SDB from HRV signals.

Comparison of Heart Rate Variability Signal Features Derived from Electrocardiography and Photoplethysmography in Healthy Individuals

“…The RR tachogram allows us to perform time-domain, frequency-domain and nonlinear dynamics analyses to extract useful features to investigate the autonomic control of the heart rate in a quantitative fashion [3]. ECG signals are acquired by placing Ag/AgCl electrodes on clearly defined anatomical positions.…”

Sleep stage classification based on multi-level feature learning and recurrent neural networks via wearable device

“…During REM sleep, in contrast, sympathetic activity shows more predominate influence. Accordingly, heart rate increases and becomes more unstable [14]. The spectral components of heart rate also exhibit different characteristics in sleep transition [9], [15].…”