Nonhomogeneous Local Atrial Activity During Acute Atrial Fibrillation: Spectral and Dynamic Analysis

Authors

  • HRAYR S. KARAGUEUZIAN,

    Corresponding author
    1. The Division of Cardiology, Cedars-Sinai Medical Center, and Department of Medicine, UCLA School of Medicine, Los Angeles, California
    Search for more papers by this author
  • STEVEN S. KHAN,

    1. The Division of Cardiology, Cedars-Sinai Medical Center, and Department of Medicine, UCLA School of Medicine, Los Angeles, California
    Search for more papers by this author
  • WERNER PETERS,

    1. The Division of Cardiology, Cedars-Sinai Medical Center, and Department of Medicine, UCLA School of Medicine, Los Angeles, California
    Search for more papers by this author
  • WILLIAM J. MANDEL,

    1. The Division of Cardiology, Cedars-Sinai Medical Center, and Department of Medicine, UCLA School of Medicine, Los Angeles, California
    Search for more papers by this author
  • GEORGE A. DIAMOND

    1. The Division of Cardiology, Cedars-Sinai Medical Center, and Department of Medicine, UCLA School of Medicine, Los Angeles, California
    Search for more papers by this author

2 Division of Cardiology, Halper Bldg. #216, Cedars-Sinai Medical Center, 8700 Beverly Blvd., Los Angeles, CA 90048

Abstract

KARAGUEUZIAN, H.S., ET AL.: Nonhomogeneous Local Atrial Activity During Acute Atrial Fibrillation: Spectral and Dynamic Analysis. Atrial fibrillation (A Fib) has been categorized into four different types (I-IV) based on the morphology of the epicardial bipolar electrogram. In the present study, we hypothesized that these same types of A Fib also exist at endocardial sites. Simultaneous high, mid, and low right atrial endocardial bipolar electrograms were analyzed during acute A Fib induced by a rapid train of stimuli (20–40 Hz] for 1–3 seconds in anesthetized closed-chest dogs (N = 7, total of 72 episodes). A Fib lasted between 3 seconds and a few minutes (22.3 ± 22.8 sec). During A Fib, bipolar electrograms (0.5–500 Hz) were both discrete (types I and II) on electrograms recorded at one site and at the same time irregular (type III) on electrograms recorded at another site. The three simultaneously recorded electrograms encompassed all combinations of the four types of A Fib. When A Fib had a discrete electrogram morphology (types I andlor II), the mean rate of the A Fib was 494 ± 93 beats/min. At a given site, electrogram morphology also changed type over time. Fast Fourier transform (FFT) of the digitized electrograms (8–10 sec, 800 Hz digitization) showed peaks mostly below 15 Hz [range 0–30 Hz), that were either discrete (narrow band) with clear harmonic components, or had continuous (broad band) spectra, that changed in a time and site dependent manner. Phase plane plots (PPP), a plot of voltage versus rate of change of voltage, varied with respect to time and location. However, the morphology of these PPP often inscribed well defined structure suggesting dynamics compatible with deterministic chaos, rather than random dynamics. We conclude that A Fib is both temporally and spatially heterogeneous and that all combinations of the four different types of A Fib occur simultaneously. These findings may be helpful in developing robust algorithms for A Fib recognition for antitachycardia devices.

Ancillary