Assessing baroreflex gain from spontaneous variability in conscious dogs: role of causality and respiration

1 Dipartimento di Scienze Precliniche, Laboratorio Interdisciplinare Tecnologie Avanzate di Vialba, 3 Centro Ricerche Cardiovascolari, Consiglio Nazionale della Ricerca, Medicina Interna II, and 4 Medicina Interna I, Ospedale L. Sacco, Universitá degli Studi di Milano, 20157 Milan; and 2 Dipartimento di Bioingegneria, Politecnico di Milano, 20133 Milan, Italy A double exogenous autoregressive (XXAR) causal parametric model was used to estimate the baroreflex gain ( XXAR ) from spontaneous R-R interval and systolic arterial pressure (SAP) variabilities in conscious dogs. This model takes into account 1 ) effects of current and past SAP variations on the R-R interval (i.e., baroreflex-mediated influences), 2 ) specific perturbations affecting R-R interval independently of baroreflex circuit (e.g., rhythmic neural inputs modulating R-R interval independently of SAP at frequencies slower than respiration), and 3 ) influences of respiration-related sources acting independently of baroreflex pathway (e.g., rhythmic neural inputs modulating R-R interval independently of SAP at respiratory rate, including the effect of stimulation of low-pressure receptors). Under control conditions, XXAR = 14.7 ± 7.2 ms/mmHg. It decreases after nitroglycerine infusion and coronary artery occlusion, even though the decrease is significant only after nitroglycerine, and it is completely abolished by total arterial baroreceptor denervation. Moreover, XXAR is comparable to or significantly smaller than (depending on the experimental condition) the baroreflex gains derived from sequence, power spectrum [at low frequency (LF) and high frequency (HF)], and cross-spectrum (at LF and HF) analyses and from less complex causal parametric models, thus demonstrating that simpler estimates may be biased by the contemporaneous presence of regulatory mechanisms other than baroreflex mechanisms. cardiovascular variability; linear parametric modeling; identification techniques