Homeostasis from a time-series perspective: An intuitive interpretation of the variability of physiological variables

Abstract

Homeostasis implies the approximate constancy of specific regulated variables, where the independence of the internal from the external environment is ensured by adaptive physiological responses carried out by other so-called effector variables. The loss of homeostasis is the basis to understand chronicdegenerative disease and age-associated frailty. Technological advances presently allow to monitor a large variety of physiological variables in a non-invasive and continuous way and the statistics of the resulting physiological time series is thought to reflect the dynamics of the underlying control mechanisms. Recent years have seen an increased interest in the variability and/or complexity analysis of physiological time series with possible applications in pathophysiology. However, a general understanding is lacking for which variables variability is an indicator of good health (e.g., heart rate variability) and when on the contrary variability implies a risk factor (e.g., blood pressure variability). In the present contribution, we argue that in optimal conditions of youth and health regulated variables and effector variables necessarily exhibit very different statistics, with small and large variances, respectively, and that under adverse circumstances such as ageing and/or chronicdegenerative disease these statistics degenerate in opposite directions, i.e. towards an increased variability in the case of regulated variables and towards a decreased variability for effector variables. We demonstrate this hypothesis for a simple mathematical model of a thermostat, and for blood pressure and body temperature homeostasis for healthy controls and patients with metabolic disease, and suggest that this scheme may explain the general phenomenology of physiological variables of homeostatic regulatory mechanisms.