Aims: It is well-known that the classical heart rate variability (HRV) analysis might lead to the wrong assessment of the autonomic nervous system (ANS) when respiratory influences are not taken into account. In this context, this study proposes an approach to quantify these (linear and nonlinear) influences by means of least-squares support vector machines. These influences can then be separated from the HRV signal, and an improved estimation of the respiratory sinus arrhythmia (RSA) and the sympathovagal balance (SB) can be achieved.
Methods: To validate the proposed approach, a dataset with different phases was used. These phases include an orthostatic maneuver (supine-standing) and a single pharmacological blockade of the sympathetic and the parasympathetic branches of the ANS. During these phases, ECG and respiratory effort signals were recorded from 13 male volunteers (ages 19-38 years, 21±4.4 years) with no history of cardiopulmonary disease. The estimations of RSA and SB were obtained for each phase after extracting the linear and nonlinear respiratory influences from HRV using the proposed methodology. These estimates were then compared against the classical HRV analysis and the significance of the nonlinear estimates was assessed using surrogate data analysis. Differences between the phases for each estimate were evaluated using the Friedman test for repeated measures with α=0.05.
Results: Results indicate that the classical HRV underestimate the cardiorespiratory interactions. Moreover, the proposed indices can better identify vagal and sympathetic withdrawal when going from supine to standing and during pharmacological blockade. Additionally, the linear and nonlinear interactions respond differently to vagal and sympathetic withdrawal.
Conclusions: The proposed approach allows to quantify both linear and nonlinear cardiorespiratory interactions, hence improving the assessment of the ANS. Finally, the findings of this study suggest that the linear and nonlinear interactions are mediated by two different cardiorespiratory control mechanisms.