Photoplethysmography Analysis to Assess Cardio-Cerebrovascular Impact of Sleep

点击翻译按钮获取中文摘要

Project Summary Obstructive sleep apnea (OSA) is common in the aging population and is associated with increased risk of cardiovascular (CV) disease, including cerebrovascular injury. Repetitive exposure to acute CV response, such as sympathetic surge and blood pressure (BP) rise following obstructive respiratory events, is an important mediating mechanism linking OSA to CV disease and cerebrovascular injury. However, such acute CV responses are not effectively captured by conventional polysomnography metrics, such as the apnea hypopnea index (AHI) commonly used in OSA evaluation. This results in imprecise classification of patients in terms of their CV risk and may be responsible for inconsistent results in epidemiologic and clinical studies. More importantly, the uncertainty of the effectiveness of continuous positive airway pressure (CPAP) therapy in reducing the CV risk poses a significant challenge in therapeutic decision-making in older individuals. Therefore, the identification of additional phenotypic markers that better quantify the unfavorable CV effects of OSA and provide improved prediction of CV outcomes is crucial to improving risk stratification and clinical therapeutic decision-making. Herein, we propose to study novel physiologic measurements that can readily be retrieved from a single photoplethysmography (PPG) sensor and investigate whether PPG features are associated with markers of subclinical, clinical CV disease, and cerebrovascular injury. We will extract PPG features from polysomnography obtained in the Multi-Ethnic Study of Atherosclerosis (MESA) Sleep study and examine whether PPG features are associated with CV outcomes, including left ventricular mass, aortic stiffness, and incident cardiovascular events, as well as markers of cerebrovascular injury, including brain structural abnormalities by MRI and cognitive impairment by Cognitive Abilities Screening Instrument in the older men and women. We will compare the association of PPG features with these outcomes with that of conventional OSA assessment metrics (such as apnea hypopnea index). The main PPG feature of interest is slope transit time. To assess the utility of the various PPG features for use in estimating and classifying outcomes, we will use novel machine learning methods, such as are based on the GNOSIS (Generalized Networks for the Optimal Synthesis of Information Systems) information-theory-based modeling tool. Subsequently, we will determine how PPG features predict BP treatment response to CPAP and O2 therapy using data from the HeartBEAT randomized controlled trial. This study attempts to identify novel PSG metrics that are cardiovascular-centric and to assess their clinical utility in the older adults. The findings of the study will provide a basis for further development of a simpler method by which to assess and monitor OSA. Considering the high prevalence of OSA in older populations, coupled with its impact on adverse health outcomes, the proposal has significant public health implications.