以心電訊號定量睡眠品質之新方法——健康受試者研究

Profiling Objective Sleep Quality in a Healthy Taiwanese Sample: Using a Novel Electrocardiogram-based Cardiopulmonary Coupling Analysis

目的：睡眠與心肺功能及自律神經活性息息相關，而性別和年齡已知會影響心臟生理。本研究應用以心電訊號定量睡眠品質之新方法－－心肺耦合分析(cardiopulmonarycouplinganalysis)，於一群健康台灣受試者來探討年齡及性別兩因素個別對睡眠生理的影響。方法：本研究包含155位健康受試者（41位男性，114位女性，平均年齡37.6±13.0歲，年齡範圍19-67歲）。情緒及睡眠量表評估包括貝氏憂鬱量表，匹茲堡睡眠品質量表及白天嗜睡程度量表。每位受試者接受整晚連續的心電訊號監測。我們以心肺耦合分析來測量心跳與呼吸之相關性並藉以推算睡眠是處於穩定，不穩定或作夢／清醒狀態。結果：本研究有三項主要發現︰(1)相較於小於40歲之成年人，大於40歲者其整晚呼吸心跳之極低頻耦合（作夢／清醒比率）顯著減少；(2)男性較女性有更高比率的呼吸心跳之低頻耦合（不穩定睡眠）；及(3)睡眠生理狀態和量表評估在健康受試者樣本無顯著相關。結論：年齡及性別對於睡眠之循環呼吸生理有顯著影響。本研究也同時提供此心電訊號定量睡眠品質方法於健康受試者之測量標準。我們建議未來此心肺耦合分析可能有應用在臨床評估睡眠的潛力。

Objectives: Sleep affects the regulation of circulatory and respiratory function. The factors of age and gender are also known to have a signifi cant impact on the cardiac physiology. We investigated the impact of the factors of age and gender on sleep-related cardiovascular and respiratory dynamics with a novel, validated cardiopulmonary coupling analysis based solely on the electrocardiogram (ECG) signal. Methods: We recruited 155 healthy subjects (41 males and 114 females, aged 37.6 ± 13.0 years, range being 19-67 years) to participate in this study. We evaluated their mood and sleep with self-reported questionnaires - the Beck Depression Inventory, the Pittsburgh Sleep Quality Index, and the Epworth Sleepiness Scales. Physiologic sleep measures were quantifi ed by analysis of continuous ECG recordings using the cardiopulmonary coupling analysis. Three sleep states were determined, namely stable, unstable, and rapid eye movement (REM)/wake state, by measuring the degree of association between autonomic and respiratory drives during sleep. Results: The key fi ndings in this study included: (A) Compared to subjects under age 40, subjects over age 40 showed to have signifi cantly decreased very-low-frequency coupling, an index of REM/wake state. (B) Compared to female subjects, male subjects revealed to have lower high-frequencycoupling, an index of stable sleep, and higher low-frequency-coupling, an index of unstable sleep. And (C) ECG-based sleep characteristics were not fount to be correlated with self-reported questionnaires in this healthy adult sample. Conclusions: Our study results showed that aging and gender as factors had signifi cant effects on cardiopulmonary coupling dynamics during sleep. This study also provided a profi le of the physiological sleep characteristics of a healthy Taiwanese sample. We suggest that further research may enhance the use of this relatively simple ECG-based method to give a cost-effi cient way to objectively evaluate sleep quality.