ECG-derived cardiopulmonary coupling biomarkers of sleep, sleep-breathing, and ca

描述（由申请人提供）：本申请涉及广泛的挑战领域（03）生物标志物的发现和验证，以及特定的挑战主题，03- hl -101*识别和验证临床相关的，可量化的血液，血管，心脏和呼吸道功能障碍的诊断和治疗反应的生物标志物。量化睡眠和睡眠呼吸的传统方法依赖于人工或计算机辅助的30秒周期评分，将离散的快相脑电图事件标记为觉醒，以及识别病理性呼吸的阈值。评分规则通常依赖于单一的生理流来确定，例如来自脑电图的觉醒。然而，唤醒性刺激可靠地诱导许多生理系统同时发生短暂的变化——皮层电系统、呼吸系统、自主神经系统、血流动力学和运动系统。这些多重关联的生理系统似乎显示出当前分期/评分系统无法识别的耦合活动的重要模式。呼吸化学反射跟踪血液中的氧气（O2）和二氧化碳（CO2）水平。疾病状态可改变呼吸化学反射的设定点或反应斜率，使它们对O2和CO2波动的敏感性降低（如肥胖低通气综合征）或提高（如中枢性睡眠呼吸暂停）。量化和跟踪睡眠期间呼吸化学反射的能力可能具有临床用途，因为1)在某些情况下，如充血性心力衰竭，化学反射敏感性可靠地增加，与疾病的严重程度和结果相关，并有助于睡眠呼吸障碍的高发。2)呼吸化学反射增强可能增加阻塞性睡眠呼吸暂停的严重程度，在使用持续气道正压通气（CPAP）治疗时与中枢呼吸暂停的诱导有关，并可能损害长期疗效和耐受性。阻塞性睡眠呼吸暂停患者由于诱导中枢性呼吸暂停和周期性呼吸（称为“复杂睡眠呼吸暂停”）而导致CPAP治疗失败，否则与CPAP反应性患者无法区分。一种能够追踪睡眠呼吸化学反射调节的生物标志物将为短期和长期动态睡眠生理学提供新的视角，具有重要的临床意义。本文提出的方法是分析耦合睡眠振荡，以数学方法提取状态特征和调节影响。其基本思想是，在多个（2个或更多）生理上不同但生物学上相关的信号流（如电皮层、自主神经、呼吸和运动）中绘制编码的共同主题，从而产生更深层次调节过程的证据，而目前仅用脑电图或气流模式对睡眠进行评分/分级的方法无法证明这一点。我们已经开发了一种方法，只需要一个单通道心电图（ECG），是自动化的，可以有参数变化的检测阈值，并且很容易重复。从心电图中，我们提取了心率变异性（HRV）和与呼吸潮气量变化相关的心电图r波幅度波动（ECG衍生呼吸，EDR）。下一步是用数学方法将HRV和EDR结合起来，产生心肺耦合的交叉乘积一致性，从而产生睡眠频谱图。睡眠谱图显示高（0.1-1 Hz）、低（0.1-0.01）和极低（0.01-0 Hz）耦合谱，显示健康和疾病状态之间的自发变化。高频耦合（High frequency coupling， HFC）是稳定和生理安宁睡眠的生物标志物，低频耦合（low frequency coupling， LFC）是不稳定或生理觉醒睡眠的生物标志物，甚低频耦合（very low frequency coupling， VLFC）是清醒或REM睡眠的生物标志物。健康由HFC主导，睡眠呼吸暂停等疾病由LFC主导。LFC的一个子集与呼吸暂停和呼吸不足相关，即LFC升高（e-LFC）。化学反射调制对e-LFC的影响越强，耦合谱色散越有可能变窄，产生窄带e-LFC（即频率相对固定的节拍振荡）。窄频带e-LFC可由高海拔、心力衰竭诱发，并可预测正压滴定期间中枢性呼吸暂停诱导。心衰的发生和发展与睡眠碎片化和化学反射敏感性增高有关。我们预测HFC会减少，窄频带e-LFC会出现并随着心力衰竭的加重而增加。这些光谱生物标志物应该随着心力衰竭的进展或消退而动态变化——通过睡眠窗口观察心功能。我们的实验将采用以下方法。我们将在健康的成人和儿童以及睡眠呼吸暂停患者中建立光谱稳定和不稳定睡眠的血流动力学相关性，以及心电图衍生生物标志物的夜间稳定性/变异性。接下来，我们将使用海拔诱导的周期性呼吸模型，这是相对纯粹的化学反射介导的睡眠呼吸暂停，来调整谱图的参数，使检测化学反射对睡眠呼吸的影响具有最佳的灵敏度和特异性。我们将平行跟踪心力衰竭患者住院6个月的进展，试图证明HFC的减少和窄频带e-LFC的出现或增加是预测心力衰竭恶化的前哨生物标志物事件（一种早期预警系统）。最后，我们将根据存档数据集的频谱表型评估临床结果，即呼吸暂停正压长期疗效研究。在该奖项为期2年的期限内，我们将验证一种独特的睡眠、睡眠呼吸和心血管生物学生物标志物，该生物标志物可以立即应用于改善健康状况。

DESCRIPTION (provided by applicant): This application addresses broad Challenge Area (03) Biomarker Discovery and Validation, and specific Challenge Topic, 03-HL-101* Identify and validate clinically relevant, quantifiable biomarkers of diagnostic and therapeutic responses for blood, vascular, cardiac, and respiratory tract dysfunction. The traditional approach to quantifying sleep and sleep-respiration relies on manual or computer assisted scoring of 30 second epochs, tagging of discrete fast phasic electroencephalographic events as arousals, and thresholds to identify pathological breathing. The scoring rules are usually reliant on a single physiological stream to make a determination, such as arousals from the electroencephalogram. However, arousing stimuli reliably induce simultaneous transient changes in numerous physiological systems - electrocortical, respiratory, autonomic, hemodynamic, and motor. These multiple linked physiological systems seem to show important patterns of coupled activity that current staging / scoring systems do not recognize. The respiratory chemoreflexes track oxygen (O2) and carbon dioxide (CO2) levels in the blood. Disease states can alter the set-point or response slope of the respiratory chemoreflexes, such that they are less (e.g., obesity hypoventilation syndrome) or more (e.g., central sleep apnea) sensitive to O2 and CO2 fluctuations. An ability to quantify and track the respiratory chemoreflexes during sleep could have clinical use, as 1) In certain conditions like congestive heart failure, chemoreflex sensitivity is reliably increased, correlates with disease severity and outcomes, and contributes to the high prevalence of sleep-disordered breathing. 2) Heightened respiratory chemoreflexes may contribute to obstructive sleep apnea severity, be associated with induction of central apneas when continuous positive airway pressure (CPAP) is used for treatment, and possibly impair long term efficacy and tolerance. Patients with obstructive sleep apnea who fail CPAP therapy due to induction of central apneas and periodic breathing (called "complex sleep apnea") are not otherwise distinguishable from CPAP-responsive patients. A biomarker that can track chemoreflex modulation of sleep respiration will provide a new view of short and long-term dynamic sleep physiology with important clinical implications. The approach proposed here is to analyze coupled sleep oscillations to mathematically extract state characteristics and modulatory influences. The fundamental idea is that mapping common themes encoded within multiple (2 or more) physiologically distinct but biologically linked signal streams (such as electrocortical, autonomic, respiratory and motor) yields evidence of deeper regulatory processes not evident by the current approach of scoring / staging sleep with electroencephalogram or airflow patterns alone. We have developed a method that needs only a single channel electrocardiogram (ECG), is automated, can have parametrically varied detection thresholds, and is readily repeatable. From the ECG, we extract heart rate variability (HRV) and ECG R-wave amplitude fluctuations associated with respiratory tidal volume changes (the ECG-derived respiration, EDR). The next step is to mathematically combine the HRV and EDR to generate the cross-product coherence of cardiopulmonary coupling, which yields the sleep spectrogram. The sleep spectrogram shows high (0.1-1 Hz, low (0.1-0.01) and very low (0.01-0 Hz) coupling spectra that show spontaneous shifts between states in health and disease. High frequency coupling (HFC) is the biomarker of stable and physiologically restful sleep, low frequency coupling (LFC) is unstable or physiologically aroused sleep, and very low frequency coupling (VLFC) is wake or REM sleep. Health is dominated by HFC, diseases such as sleep apnea by LFC. A subset of LFC that correlate with apneas and hypopneas is elevated LFC (e-LFC). The stronger the chemoreflex modulatory influence on e-LFC, the more likely the coupling spectral dispersion narrows, yielding narrow band e-LFC (i.e., metronomic oscillations with a relatively fixed frequency). Narrow band e-LFC is induced by high altitude, heart failure, and predicts central apnea induction during positive pressure titration. The development and progression of heart failure is associated with fragmented sleep and heightened chemoreflex sensitivity. We predict that HFC will decrease and narrow band e-LFC will emerge and increase with worsening heart failure. These spectral biomarkers should change dynamically with heart failure progression or regression - viewing cardiac function through the window of sleep. Our experiments will take the following approach. We will establish the hemodynamic correlates of spectrographic stable and unstable sleep and night-to-night stability / variability of the ECG-derived biomarkers in adults and children in health, and in those with sleep apnea. Next, we will use a model of altitude-induced periodic breathing, which is relatively pure chemoreflex-mediated sleep apnea, to adjust the spectrogram's parameters that allow the best sensitivity and specificity for detecting chemoreflex influences on sleep respiration. We will in parallel track the progress of heart failure patients from a hospitalization episode for 6 months, attempting to show that reductions of HFC and emergence or increases in narrow band e-LFC are sentinel biomarker events that predict worsening of heart failure (an early warning system). Finally, we will assess clinical outcomes based on spectral phenotyping of an archived data set, the Apnea Positive Pressure Long-term Efficacy Study. In the 2-year duration of the award, we will validate a unique biomarker of sleep, sleep-breathing, and cardiovascular biology that can be applied immediately to improve health outcomes. PUBLIC HEALTH RELEVANCE: Simple measures of sleep, sleep-breathing and heart function that are cheap, readily repeatable, and which can track disease fluctuations would be useful, for clinical and research purposes. A new method based on a single channel of electrocardiogram (ECG) has been developed; it uses changes in the speed of the heart beat and breathing-related size modifications of the ECG, to create a "picture of the "music of sleep". We propose to show its usefulness as a monitor of sleep in health, in those with simple and complicated forms of sleep apnea, and in patients with heart failure.