Assessment of Cardiovascular Coupling during Rest and Postural Stress Using Mutual Information Rate

In this study we applied the Mutual Information Rate (MIR) measure and surrogate data analysis approaches to assess dynamic coupling and the role of nonlinearities in shortterm cardiovascular regulatory mechanisms. The framework of analysis was applied to heart period and systolic arterial pressure time series recorded in 19 healthy normotensive subjects, undergoing an orthostatic stress test. Specifically, the MIR was estimated by using a model-free nearest neighbor approach and decomposed into complexity, namely entropy rate (ER), and causality, namely transfer entropy (TE), measures. The analysis pointed out a statistically significant decrease in the ERs of both heart period and systolic arterial pressure dynamics and a decrease of TE along the branch from heart period to arterial pressure. Moreover, surrogate data analysis highlighted a decrease of the nonlinearity of the involved physiological mechanisms. Our findings suggest an activation of the sympathetic system, an enforcement of the baroreflex mechanism, and a linearization of the cardiovascular dynamics in response to orthostatic stress, which is consistent with the available literature. The proposed approach demonstrated robustness and potentiality for unraveling coupling and nonlinearity properties in cardiovascular dynamics.