Session P7A.2

Multiscale Information Analysis of the Autonomous Nervous System during Myocardial Ischemia

JF Valencia, M Vallverdú, P Gomis*, GS Wagner, P Caminal

Universidad Politécnica de Catalonia
Barcelona, Spain

Heart Rate Variability is used to study the autonomous nervous system’s (ANS) control behavior, since it modulates the cardiovascular system during percutaneous transluminal coronary angioplasty (PTCA). The purpose of this study is to provide a new methodology to characterize ANS changes as a response over three periods of a PTCA procedure using a multiscale information analysis (MIA) of HRV. The present study describes the ANS’s response before, during and after a PTCA by MIA based on the concept of entropy rates. The study group consisted of 66 patients undergoing elective prolonged balloon occlusion during PTCA in one of their major coronary arteries (STAFF-III study): left main in 1 patient, left anterior descending artery (LAD) in 21 patients, right coronary artery (RCA) in 30 patients and left circumflex artery (LCX) in 14 patients. Each patient has 3 min. pre-PTCA, 3 min. PTCA occlusion, and up to 3 min. post-inflation of RR series obtained from ECG recording. The RR sequence was interpolated using cubic splines in order to obtain a series of uniformly sampled data. The re-sampling rate was set to 5 Hz. Then the RR series were filtered in the LF [0.04-0.15 Hz] and HF [0.15-0.4 Hz] bands. Finally, the RR(LF), RR(HF) and RR series were transformed to a suitable symbolic description of four symbols and words of different scales (length-L-words) were constructed. MIA was defined as a rate of entropies measured on the series of words calculated on different time scales. The entropies were based on Shannon and Renyi definitions. The methodology considered different entropy orders, q of 0.1, 0.15, 0.25, 1, 2 and 4. To assess multiscale information measures, statistically significant differences with p-value<0.01 were taken into account. Results have showed less regularity behavior in pre-PTCA than during PTCA for q>=1 and less regularity during PTCA than in post-PTCA for q<=0.25 when RR series were analyzed. In RR(HF) series, higher regularity was presented in pre-PTCA than during PTCA for q<=1. In RCA group when LF band was considered, higher regularity behavior was shown in pre-PTCA than during PTCA for low scales and q<=0.25 and a reverse behavior was presented for high scales. These findings suggest that MIA provides a powerful tool allowing assessment of ANS´s response during PTCA.

(Abstract Control Number: 177)