Comparison of frequency domain measures based on spectral decomposition for spontaneous baroreflex sensitivity assessment after Acute Myocardial Infarction. (July 2021)

• Record Type: Journal Article
• Authors: Pernice, Riccardo; Sparacino, Laura; Nollo, Giandomenico; Stivala, Salvatore; Busacca, Alessandro; Faes, Luca
• Is Part Of: Biomedical signal processing and control. Volume 68(2021)
• Publication Date: 2021-07
• Subjects: 92C55 -- 92C30; Spectral decomposition -- Frequency domain -- Causality -- Baroreflex -- Head-up tilt -- Cardiovascular control -- Acute myocardial infarction; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.102680
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Parametric approaches to assess spontaneous cardiovascular variability are studied. A new causal local spectral decomposition method is used for baroreflex analysis. Causal approaches allow to differentiate baroreflex and feedforward interactions. Local approaches avoid confounding oscillatory effects outside the band of interest. Causal local measures allow an objective assessment of baroreflex coupling and gain. Abstract: The objective of this study is to present a new method to assess in the frequency domain the directed interactions between the spontaneous variability of systolic arterial pressure (SAP) and heart period (HP) from their linear model representation, and to apply it for studying the baroreflex control of arterial pressure in healthy physiological states and after acute myocardial infarction (AMI). The method is based on pole decomposition of the model transfer function and on the following evaluation of causal measures of coupling and gain from the poles associated to low frequency (0.04−0.15 Hz) oscillatory components. It is compared with traditional non-causal approaches for the spectral analysis of the baroreflex gain, and with causal approaches based on the directed coherence, in a group of AMI patients and in Young and Old healthy controls studied at rest and during head-up tilt. Analysis of feedforward interactions from RR to SAP is also performed. Our results support the importance of using local causal approaches to quantify separately baroreflex and feedforward interactions between RR and SAP, allowing both to confirm known physiological behaviors (e.g., weaker baroreflex effectiveness and lower sensitivity after AMI) and to suggest novel findings (e.g., preserved low frequency baroreflex response to postural stress after AMI). We also find that the postural stress alters feedforward interactions selectively across groups, being related to decreased coupling only in Young and to increased gain mostly in AMI. These results have a significance for the clinical assessment of the baroreflex and the physiological evaluation of cardiovascular interactions.