CARDIOPULMONARY TIME SERIES ANALYSIS--DRUG AND AGING EFFECTS

点击翻译按钮获取中文摘要

We propose that bioelectrical time series such as electrocardiograms (ECG), blood pressure recordings, and respiratory recordings, contain information regarding the state of a human, and, the effect of a drug on the state of a patient that univariate analytic techniques and simple mean effect measures are inadequate to characterize. We have shown that Fast Fourier Transformation (FFT)-based methods to analyze beat-to-beat variability of heart rate (ECG data) during administration of verapamil to man demonstrated effects (decreased variability in frequencies associated with respiratory or vagally-mediated parasympathetic modulation) which were not detected in average heart rate responses or other time domain analyses of heart rate responses. In addition, we and others have shown that frequency domain analysis of ECG data can identify activation of baroreflex responses - potentially allowing identification and quantitation of indirect drug effects via reflex response activation which may markedly influence net drug effect in vivo. We have also used these analytic techniques to show that aging alters basal heart rate variability and frequency content of heart rate responses to postural maneuvers. In these initial/pilot studies, it became apparent that method of respiration (as well as rate) alters the frequency content of heart rate data, and needs to be considered in evaluating heart rate data, and, that optimal methods for interpretation of frequency domain data were not yet developed. We have also demonstrated age-related changes in pharmacodynamics as well as pharmacokinetics. In particular, baroreflex responses appear to be altered (diminished) with aging. Since altered responses to drugs can result from differing drug demonstrations, tissue sensitivities, differing homeostatic reflex responses, or a combination of these factors, it is important to evaluate the contributions of altered reflex responses to the altered pharmacodynamic responses of elderly individuals to drugs. This is especially important with vasoactive drugs which may activate baroreceptor responses and are the most frequently prescribed drugs in the elderly population. Therefore, we now propose to further develop and validate methods for analysis of bioelectric time series information. In addition to refining methods for analysis of beat-to-beat variation in heart rate, we will develop a method for analysis of beat-to-beat variability in blood pressure, and breath-to-breath variability in respiration. We will then develop a model in which heart rate, blood pressure, and respiration time series are interrelated by transfer functions and in which the interactions between variables can be evaluated and quantitated as to time course, magnitude, and contribution to in vivo effects. We will then use this combined bioelectric time series model to (1) analyze effects during drug administration to identify and quantitate contributions of reflex (parasympathetic or beta-adrenergic) responses in combination with drug dose or concentration data to net drug effect in vivo and (2) test the hypotheses that aging changes time series content of blood pressure and respiration and the interrelationship between heart rate, blood pressure and respiration; and, that age-related changes in reflex responses are a major determinant of altered drug responses in the elderly that may be predictable and quantifiable. Finally, based on results from these investigations, we will reduce the model to the simplest form that adequately describes and predicts physiologic and drug responses in humans.