Session P74.2

An Investigation on Autonomic Effects by Using PR Intervals

TW Shen*, YT Tsao

Tzu Chi University
Hualien, Taiwan

Objectives: Many experts are looking for the relation between the heart and the automatic nervous system (ANS). Most commonly, physicians nowadays use the heart rate variability (HRV) to observe ANS more than others. However, the PR interval represents the atrio-ventricular conduction time (AVCT), and the AVCT is coupled to the sympathetic / parasympathetic activities of the ANS. Unfortunately, little is known about the PR autonomic modulations. In this research, a method is provided to obtain a location of a P-wave fractal point in surface ECGs recorded with the current technology. In addition, the research would like to investigate PR intervals and to compare with HRV analysis.
Methods: Malik et al. showed that the spectral analysis of RR interval variability will change from rest to 90° head-up tilt. The same experiment is done to observe PR interval variability. Our experiment collected lead-I and lead-III ECG signals from 32 young normal healthy subjects at the sampling rate 500 Hz. Two five-minute ECG signals were recorded both at rest and during 90° head-up tilt. Many digital signal processing technologies applied on collected ECG signals to detect PQRST points. Here are digital filtering, the Pan & Tompkins' method, first derivative ECG (dECG), and the zero-crossing method. After PQRST points are correctly detected. The general HRV, the PR interval variability, the TR interval, and the combined voting system analysis were studied.
Results: This research chose five characteristics to analysis for distinguishing two poses. There are the mean of RR intervals, the ratio LF/HF by AR model, the ratio LF/HF by FFT model, the mean of PR intervals, and the mean of TR intervals. The results of HRV analysis showed that only 62.5% subjects followed the principle that the ratio LF/HF decreased during tilt. Then, the same database was used to analysis PR interval variability, and we discovered that 84.4% of the mean of PR intervals increased when subjects tilted their head. Finally, a voting model which combined all characteristics has 81.3% accuracy to distinguish two poses.
Conclusions: Based on our experiment, standard HRV analysis seems not the only choice to distinguish from resting position to 90° head-up tilt position. Hence, we make a hypothesis to explain why the PR segment influenced more than others – The head-up tilt position may suppress parasympathetic activity, the heart rate increases. However, the sympathetic system would like to restrain in the current state, so the PR intervals prolong and tried to decrease the heart rate. Our future work will study more on PR interval variability to obtain further knowledge on automatic nervous system. Moreover, our developed technology may be contributed on the pacemaker algorithm, doctor diagnosis, HRV analysis, and ANS physiological modeling.

(Abstract Control Number: 85)