Extracorporeal carbon dioxide removal: Effect of short time infusion of 3 different metabolic acids

Abstract

Objectives: The measures of heart rate variability (HRV) can be classified into multiple domains of variability, depending on their theoretical background and method of computation; however, their physiological interpretation and degree of mutually independent information remain unknown, especially given that multiple measures of variability across domains are correlated with each other. The objective of this study was to investigate independence and ability to detect change of a set of measures of HRV in silico. Methods: To understand the physiological meaning of HRV, we used a physiologically based mathematical model from Kotani et al (2005). By altering its parameters, this model is able to produce R-R interval time series showing the scale-invariant properties of both healthy subjects and patients affected by heart diseases. We used Kotani's model to generate a set of R-R interval time series in the healthy case and in altered physiological conditions (ie, when the values of selected parameters of the model are modified) in order to investigate (1) the degree of independence of a set of measures of variability from multiple domains based on their mutual information and (2) their ability to sense changes in the physiological parameters of Kotani's model and to track them over time. Results: Our results show that, in the healthy cardiovascular system state of the model, the investigated HRV measures present low mutual information and, therefore, high degree of statistical independence. When the physiological conditions are altered by modifying the values of the parameters of Kotani's model, the measures follow those changes either increasing or decreasing their values. Each HRV measure resulted sensitive to multiple changes in the physiological conditions, in particular sympathetic and parasympathetic activity, baroreceptor activity, contractility of the heart, and respiration. Correlation analysis confirmed that the alteration of HRV measures due to changes in the physiological conditions leads to higher levels of correlation among the measures. Conclusions: The results suggest that each measure of HRV is a nonspecific sensor of the cardiovascular system. Because of the independence of HRV measures under baseline conditions as well as their partial correlation following physiological changes, the combination of multiple measures into composite indexes can enhance the “common” information related to physiological changes. This approach could be used to improve our understanding of cardiovascular physiology and translate the acquired knowledge into clinical applications.