Exponential distribution of long heart beat intervals during atrial fibrillation and their relevance for white noise behaviour in power spectrum

TL;DR

This study reveals that heart beat intervals during atrial fibrillation follow an exponential distribution with a rate matching the white noise component in the power spectrum, offering new diagnostic parameters.

Contribution

It uncovers the exponential tail in interbeat interval distribution during AF and links it to white noise behavior, providing novel insights for AF classification and diagnosis.

Findings

Exponential tail present in AF heart beat intervals

Rate of exponential decay matches white noise level in power spectrum

Interval decomposition into correlated and uncorrelated processes

Abstract

The statistical properties of heart beat intervals of 130 long-term surface electrocardiogram recordings during atrial fibrillation (AF) are investigated. We find that the distribution of interbeat intervals exhibits a characteristic exponential tail, which is absent during sinus rhythm, as tested in a corresponding control study with 72 healthy persons. The rate of the exponential decay lies in the range 3-12 Hz and shows diurnal variations. It equals, up to statistical uncertainties, the level of the previously uncovered white noise part in the power spectrum, which is also characteristic for AF. The overall statistical features can be described by decomposing the intervals into two statistically independent times, where the first one is associated with a correlated process with 1/f noise characteristics, while the second one belongs to an uncorrelated process and is responsible for the exponential tail. It is suggested to use the rate of the exponential decay as a further parameter for a better classification of AF and for the medical diagnosis. The relevance of the findings with respect to a general understanding of AF is pointed out.