Frequency domain algorithm for quantifying atrial fibrillation organization to increase defibrillation efficacy

Understanding and Interpreting Dominant Frequency Analysis of AF Electrograms

Dominant frequency analysis of atrial electrograms has been used to understand the pathophysiology of atrial fibrillation (AF). Although dominant frequency is an effective tool to estimate activation rate during AF, other factors besides activation rate may alter the results. Therefore, an adequate conceptual understanding of frequency domain analysis is required to properly use this technique and interpret the results. This review, while avoiding the use of formulas and equations, aims to explain fundamental theory of how signals can be decomposed into sine waves and how these sine waves relate to the activation rate detected from the electrograms. Through a series of examples and illustrations this relationship can be easily conceptualized. This will in turn allow the strengths and limitations of dominant frequency analysis to be better understood and improve its applicability to potential clinical usages.

Feature extraction of the atrial fibrillation signal using the continuous wavelet transform

Despite advances in cardiac arrhythmia management, atrial fibrillation remains a major cause of cardiovascular morbidity and mortality. Recent data suggests that there are periods of organization within this apparently chaotic arrhythmia. To date, analysis of the rapidly changing atrial fibrillation signal has been limited by a lack of time-frequency resolution. When used to analyze high-density atrial mappings of this arrhythmia, the continuous wavelet transform, with its time-frequency multi-resolution capability, may provide important temporal and spatial information regarding arrhythmia organization and may lead to the development of more effective therapies.

Deterioration of Organization in the First Minutes of Atrial Fibrillation: A Beat-to-Beat Analysis of Cycle Length and Wave Similarity

INTRODUCTION

It has been recently suggested that many episodes of atrial fibrillation (AF) may be partially organized at the onset and thus more suitable for antitachycardia pacing therapy. Nevertheless, the time course of organization in the first minutes of AF has not been quantified yet.

METHODS AND RESULTS

Twenty episodes of paroxysmal AF were studied. Electrograms were recorded from the right atrium (RA), distal (CSd), and proximal coronary sinus (CSp). The time course of AF cycle length (AFCL) and the regularity of wave morphology (similarity index S) were beat-to-beat measured at each recording site during the first 7 minutes of AF. AFCL and S showed a decreasing trend after the onset of AF. AFCL decreased from 208 +/- 31 to 171 +/- 21 msec (P < 0.001), from 206 +/- 40 to 169 +/- 23 msec (P < 0.001) and from 190 +/- 42 to 152 +/- 18 msec (P < 0.05), respectively, in RA, CSd, and CSp. Similarly, the similarity index decreased in CSd from 0.37 +/- 0.27 to 0.12 +/- 0.09 (P < 0.01) and in RA from 0.40 +/- 0.18 to 0.17 +/- 0.16 (P < 0.001). The 80% of the decrease occurred during the first 3 minutes of the arrhythmia, while after this time both cycle length and similarity index did not change significantly anymore. Conversely, the electrical activity in CSp was highly disorganized (S = 0.05 +/- 0.03) even in the first minute of AF, and no decreasing temporal trend was observed.

CONCLUSION

Higher levels of organization and longer fibrillation intervals exist at the onset of AF. The degree of organization of the electrical activity decays within less than 3 minutes. Since antitachycardia pacing success rate increases with high levels of organization, these results suggest an early delivery of pacing treatment.

Structural atrial remodeling alters the substrate and spatiotemporal organization of atrial fibrillation: a comparison in canine models of structural and electrical atrial remodeling

Several animal models of atrial fibrillation (AF) have been developed that demonstrate either atrial structural remodeling or atrial electrical remodeling, but the characteristics and spatiotemporal organization of the AF between the models have not been compared. Thirty-nine dogs were divided into five groups: rapid atrial pacing (RAP), chronic mitral regurgitation (MR), congestive heart failure (CHF), methylcholine (Meth), and control. Right and left atria (RA and LA, respectively) were simultaneously mapped during episodes of AF in each animal using high-density (240 electrodes) epicardial arrays. Multiple 30-s AF epochs were recorded in each dog. Fast Fourier transform was calculated every 1 s over a sliding 2-s window, and dominant frequency (DF) was determined. Stable, discrete, high-frequency areas were seen in none of the RAP or control dogs, four of nine MR dogs, four of six CHF dogs, and seven of nine Meth dogs in either the RA or LA or both. Average DFs in the Meth model were significantly greater than in all other models in both LA and RA except LA DFs in the RAP model. The RAP model was the only one with a consistent LA-to-RA DF gradient (9.5 ± 0.2 vs. 8.3 ± 0.3 Hz, P < 0.00005). The Meth model had a higher spatial and temporal variance of DFs and lower measured organization levels compared with the other AF models, and it was the only model to show a linear relationship between the highest DF and dispersion ( R2 = 0.86). These data indicate that structural remodeling of atria (models known to have predominantly altered conduction) leads to an AF characterized by a stable high-frequency area, whereas electrical remodeling of atria (models known to have predominantly shortened refractoriness without significant conduction abnormalities) leads to an AF characterized by multiple high-frequency areas and multiple wavelets.

Effect of electrogram characteristics on the relationship of dominant frequency to atrial activation rate in atrial fibrillation

BACKGROUND

Dominant frequency analysis has become a widely used tool for studying the pathophysiology of atrial fibrillation (AF).

OBJECTIVES

The purpose of this study was to compare dominant frequency with atrial activation rates in the presence of changing activation intervals and amplitudes and the complex morphologies characteristic of AF electrograms.

METHODS

A combination of atrial electrograms recorded during persistent AF from 10 patients, atrial flutter electrograms with double potentials from 12 patients, and simulated electrograms were used in this study. Dominant frequencies were compared with the mean, median, and mode beat-to-beat activation rates obtained by electrogram marking.

RESULTS

Dominant frequency correlated well with, but did not specifically reflect, mean, median, and mode activation rates. Dominant frequencies were significantly impacted by frequency variation, combined amplitude and frequency variation, and ordering of the activation intervals. Averaging dominant frequency measurement of four consecutive signals improved reproducibility and agreement with mean and median activation rates. Signals with double potentials having longer delays between potentials were associated with harmonics chosen as the dominant frequency.

CONCLUSION

Multiple dominant frequency measurements and scrutiny of the time and frequency domain signals are recommended to obtain accurate and reproducible values.

The Effects of Pulmonary Vein Isolation on the Dominant Frequency and Organization of Coronary Sinus Electrical Activity During Permanent Atrial Fibrillation

BACKGROUND

Pulmonary vein isolation (PVI) has been shown to suppress atrial fibrillation (AF). We examined the effects of PVI on disorganization and dominant frequencies (DF) in patients with permanent AF.

METHODS AND RESULTS

Twenty-eight patients with permanent AF (>6 months) who failed > or =1 antiarrhythmic drugs (AAD) and > or =2 cardioversions (CV) with AF reversion <30 minutes after CV were included. PVI and isolation of DFs in pulmonary veins (PVs) was performed during AF. Fast Fourier transformations of atrial electrograms were performed. Disorganization index (DI) was defined as the percentage of time spent in type III AF during 1-minute continuous recordings. The temporal stability and reproducibility of DIs from the same sites were verified over time prior to ablation. Highly disorganized AF activity concentrated in the posterior left atrium (PLA) including sites at the left atrial (PV-LA) junction (55.7% of sites in PLA, 32.9% in septum, and 11.4% in other sites). DF and DI from the coronary sinus (CS) before and after PVI were analyzed. PVI reduced the DI (14.3 +/- 25.0% before PVI vs 4.6 +/- 8.6% after PVI; P < 0.02). There was significant reduction of DI in 26 of 28 patients. The DF remained unchanged (5.6 +/- 1.3 Hz before PVI vs 5.9 +/- 0.9 Hz after PVI; P = 0.31). After a follow-up of 30 +/- 11 months, 15 (54%) of patients are free of symptomatic AF, 3 (10%) in sinus rhythm on AAD, 5 (18%) with paroxysmal AF, 4 (14%) in chronic AF, and 1 (4%) with atypical flutter.

CONCLUSIONS

In the vast majority of patients with chronic AF, PVI reduces AF disorganization without affecting the DF as measured in the CS.

Wave Similarity Mapping Shows the Spatiotemporal Distribution of Fibrillatory Wave Complexity in the Human Right Atrium During Paroxysmal and Chronic Atrial Fibrillation

INTRODUCTION

The complexity of waveforms during atrial fibrillation may reflect critical activation patterns for the arrhythmia perpetuation. In this study, we introduce a novel concept of map, based on the analysis of the wave morphology, which gives a direct evidence in the human right atrium on the spatiotemporal distribution of fibrillatory wave complexity in paroxysmal (PAF) and chronic (CAF) atrial fibrillation.

METHODS AND RESULTS

Electrograms were recorded from a 64-electrode catheter in the right atrium of 15 patients during PAF (n = 8) and CAF (n = 7). Wave similarity maps were constructed by calculating the degree of morphological similarity of activation waves (S) at each atrial site and by following its temporal evolution. During PAF the spatiotemporal distribution of the waveforms was highly consistent across the subjects and was determined by the anatomic location. Wave similarity maps showed the existence of an extended area with low similarity index, which covered the low posteroseptal atrium (S = 0.28 +/- 0.09) and the septal region (S = 0.22 +/- 0.04), and the presence of a large tongue with high similarity index, which penetrated the lateral wall (S = 0.55 +/- 0.08) starting from the high anterolateral atrium (S = 0.54 +/- 0.06). A completely different spatiotemporal pattern was seen during CAF. No distinct regions with different similarity indexes were recognized, but a uniformly distributed low similarity index (S = 0.27 +/- 0.07) was found. The spatial pattern was highly stable in time with fluctuations of S < 0.04.

CONCLUSION

Quantification of the spatiotemporal distribution of fibrillatory wave complexity is feasible in humans by wave similarity mapping. Anatomic anchoring of waveforms during PAF and pattern destruction during CAF was determined.

Quantification of synchronization during atrial fibrillation by Shannon entropy: validation in patients and computer model of atrial arrhythmias

Atrial fibrillation (AF), a cardiac arrhythmia classically described as completely desynchronized, is now known to show a certain amount of synchronized electrical activity. In the present work a new method for quantifying the level of synchronization of the electrical activity recorded in pairs of atrial sites during atrial fibrillation is presented. A synchronization index (Sy) was defined by quantifying the degree of complexity of the distribution of the time delays between sites by Shannon entropy estimation. The capability of Sy to discriminate different AF types in patients was assessed on a database of 60 pairs of endocardial recordings from a multipolar basket catheter. The analysis showed a progressive and significant decrease of Sy with increasing AF complexity classes as defined by Wells (AF type I Sy = 0.73 +/- 0.07, type II Sy = 0.56 +/- 0.07, type III Sy = 0.36 +/- 0.04, p < 0.001). The extension of Sy calculation to the whole right atrium showed the existence of spatial heterogeneities in the synchronization level. Moreover, experiments simulated by a computer model of atrial arrhythmias showed that propagation patterns with different complexity could be the basis of different synchronization levels found in patients. In conclusion the quantification of synchronization by Shannon entropy estimation of time delay dispersion may facilitate the identification of different propagation patterns associated with AF, thus enhancing our understanding of AF mechanisms and helping in its treatment.

Presence of left-to-right atrial frequency gradient in paroxysmal but not persistent atrial fibrillation in humans

BACKGROUND

Recent studies have demonstrated spatiotemporal organization in atrial fibrillation (AF), with a left-to-right atrial frequency gradient during AF in isolated sheep hearts. We hypothesized that human AF would also manifest a left-to-right atrial frequency gradient.

METHODS AND RESULTS

Thirty-one patients aged 56.7+/-10.5 years with a history of paroxysmal or persistent (>1 month) AF were included. Recordings were made at each pulmonary vein (PV) ostium and simultaneously from the coronary sinus (CS) and posterior right atrium (RA) during AF. Sequential fast Fourier transforms (FFTs) were performed. FFT profiles were analyzed to determine the dominant frequency (DF). There were 18 patients with paroxysmal AF and 13 with persistent AF. In the paroxysmal group, there was a significant left-to-right atrial DF gradient, with DF highest at the PV/left atrial (LA) junction, intermediate at the CS, and lowest in the RA (6.2+/-0.8, 5.5+/-0.7, and 5.1+/-0.6 Hz, respectively; P<0.001). There were no patients in whom DF was greater at the RA than the PV/LA junction. In the persistent group, there was no significant difference between DF recorded from the LA/PV junction, CS, and RA (6.1+/-0.7, 5.8+/-0.6, and 5.8+/-0.6 Hz, respectively; P=NS).

CONCLUSIONS

In humans with paroxysmal AF, DFs are highest at the PV/LA junction, intermediate in the CS, and slowest in the posterior RA. These findings agree with animal models that suggest that the posterior LA may play an important role in maintaining paroxysmal AF. The role of the posterior LA in persistent AF requires further study.

Temporal and Spatial Phase Analyses of the Electrocardiogram Stratify Intra-Atrial and Intra-Ventricular Organization

We hypothesized that electrocardiogram (ECG) spatial phase analysis would define a spectrum of intracardiac organization from atrial fibrillation (AF), nonisthmus-dependent and isthmus-dependent atrial flutter (AFL) to supraventricular tachycardias (SVT), and similarly for ventricular arrhythmias. We analyzed arrhythmia ECGs of 33 patients with isthmus (n = 9) and nonisthmus (n = 5) dependent AFL and SVT: atrial (n = 3), atrioventricular nodal (n = 3), and orthodromic reciprocating (n = 3) tachycardias, as well as AF (n = 5), ventricular tachycardia (monomorphic, VT-MM; n = 7), and fibrillation (VF; n = 3). ECG spatial phase was considered coherent when the correlation coefficient of an atrial (or ventricular) template to its ECG over time maintained a constant relationship in XY, XZ, and YZ planes. Regularity was quantified spectrally from ECG and correlation series. Spatial coherence occurred in 9/9 cases of isthmus--but only 1/5 of cases of nonisthmus-dependent AFL (p < 0.01; chi2). All showed one dominant spectral peak (temporal coherence). In AF, spatial phase was inconsistent in all planes and spectra were broad band. Temporal and spatial coherence occurred in other SVT. VT-MM maintained spatial phase and a single spectral peak, while VF displayed neither. Our conclusions are that temporal and spatial phase analysis from the ECG stratifies intra-atrial and intra-ventricular organization and reveals subtle variability lost on visual inspection.

Defibrillation shock success estimation by a set of six parameters derived from the electrocardiogram

It is well known that in some cases defibrillator shocks cannot terminate ventricular fibrillation (VF). Repeated failed shocks often may worsen subsequent response to therapy. This study assesses the ability of six parameters derived from the surface electrocardiogram (ECG) to predict defibrillation shock outcome. Using stepwise discriminant analysis, we obtained several discriminant functions, yielding different combinations of sensitivity and specificity for detection of pre-shock ECG segments corresponding to successful versus unsuccessful shocks. The study was performed consecutively for 3, 4 and 5 s ECG time intervals. The prediction accuracy of 72.3% (61.8% sensitivity and 79.6% specificity) with five parameters and 3 s VF segment analysis prior to defibrillation shock could be considered acceptable for possible practical application in automatic external defibrillators.

Left atrial dilatation resulting from chronic mitral regurgitation decreases spatiotemporal organization of atrial fibrillation in left atrium

Atrial conduction properties have been shown to differ among animal atrial fibrillation (AF) models of rapid atrial pacing (RAP), chronic mitral regurgitation (MR), and control. We hypothesized that these conduction differences would continue with the onset of AF, which would affect AF spatiotemporal organization, resulting in model-specific characteristics of AF. With frequency domain analysis of electrograms acquired from high-density optical mapping, AF from the right (RA) and left (LA) atrium in animals with RAP and MR were compared with control animals. At follow-up, the hearts were excised and perfused, and optical action potentials were recorded from a 2 × 2-cm area each of the RA and LA free wall with a 16 × 16 photodiode array. AF was induced with extra stimuli, several 2.4-s AF episodes were recorded in each dog, and a fast Fourier transform was calculated. The dominant frequency (DF) was determined, and the organization (organization index, OI) was calculated as the ratio of the area under the dominant peak and its harmonics to the total area of the spectrum. All possible pairs of electrograms for each episode were cross-correlated. LA AF in the chronic MR model showed an increase in the highest DF, the number of DF domains, and in frequency gradient compared with AF in control or RAP models. In addition, there was a decrease in OI and in the correlation coefficients in the LA of the MR model. These results suggest that the AF substrate in the MR model may be different from that of control or RAP models.

Basic mechanisms of atrial fibrillation

The mechanism of AF remains controversial as support still exists for multiple wavelets, "mother rotor," and focal sources. These mechanisms need not be mutually exclusive. For example, the mother rotor hypothesis may not be distinct form the focal AF, if the rotor is of small size (ie, microre-entry). With the development of several animal models to study AF, along with improving technologies and mapping techniques, a further understanding of the pathophysiology of AF is being gained. Each animal model has unique electrophysiological and structural abnormalities, and one may not be able to generalize from one model to the next. It is likely that there is not one mechanism for all AF, but that there are substrate-specific mechanisms, and that AF may be comprised of several different mechanisms.

Sequential Characterization of Atrial Tachyarrhythmias Based on ECG Time-Frequency Analysis

A new method for characterization of atrial arrhythmias is presented which is based on the time-frequency distribution of an atrial electrocardiographic signal. A set of parameters are derived which describe fundamental frequency, amplitude, shape, and signal-to-noise ratio. The method uses frequency-shifting of an adaptively updated spectral profile, representing the shape of the atrial waveforms, in order to match each new spectrum of the distribution. The method tracks how well the spectral profile fits each spectrum as well as if a valid atrial signal is present. The results are based on the analysis of a learning database with signals from 40 subjects, of which 24 have atrial arrhythmias, and an evaluation database with 211 patients diagnosed with atrial fibrillation. It is shown that the method robustly estimates fibrillation frequency and amplitude and produces spectral profiles with narrower peaks and more discernible harmonics when compared to the conventional power spectrum. The results suggest that a rather strong correlation exist between atrial fibrillation frequency and f wave shape. The developed set of parameters may be used as a basis for automated classification of different atrial rhythms.

Quantifying intracardiac organization of atrial arrhythmias using temporospatial phase of the electrocardiogram

INTRODUCTION

Separating nonisthmus-dependent atrial flutter (AFL) from "organized" atrial fibrillation (AF), or isthmus-dependent AFL, may be difficult using ECG characteristics alone. We hypothesized that temporal and spatial phase analysis of ECG atrial waveforms could effectively separate these rhythms by quantifying subtle variations in ECG atrial activation during supraventricular tachycardias (SVT).

METHODS AND RESULTS

We studied 52 patients at electrophysiologic study (EPS) who demonstrated isthmus-dependent (n = 15) and nonisthmus-dependent (n = 9) AFL, atrial tachycardia (n = 6), AV nodal reentry (n = 9), orthodromic reciprocating tachycardia (n = 6), and AF (n = 7). Atrial activity was represented as a series of correlations of an atrial template to successive time samples of the arrhythmia ECG. Spatial phase was analyzed as a reproducible relationship of this atrial activity between leads over time; temporal regularity was measured from power spectra. Spatial phase was maintained (coherent) in lead planes V5/aVF (XY), V5/V1 (XZ), and aVF/V1 (YZ) in 15 of 15 cases of isthmus-dependent AFL, but in only 1 of 9 cases of nonisthmus-dependent AFL (P < 0.01; chi2). Temporally, all cases of AFL showed one dominant peak on correlation spectra (magnitude >6 dB), suggesting one activation wavefront, although this was smeared in nonisthmus-dependent cases. In contrast, AF showed inconsistent spatial phase in all planes and broad band spectra, consistent with multiple and/or variable activation paths. All other SVTs showed spatial coherence and one dominant spectral peak.

CONCLUSION

Coherence of temporal and spatial phase is a powerful approach to measure the spatial organization of intracardiac activation from the ECG that reveals a spectrum from SVT to isthmus-dependent and nonisthmus-dependent AFL, to AF.

A method for quantifying atrial fibrillation organization based on wave-morphology similarity

A new method for quantifying the organization of single bipolar electrograms recorded in the human atria during atrial fibrillation (AF) is presented. The algorithm relies on the comparison between pairs of local activation waves (LAWs) to estimate their morphological similarity, and returns a regularity index (rho) which measures the extent of repetitiveness over time of the detected activations. The database consisted of endocardial data from a multipolar basket catheter during AF and intraatrial recordings during atrial flutter. The index showed maximum regularity (rho = 1) for all atrial flutter episodes and decreased significantly when increasing AF complexity as defined by Wells (type I: rho = 0.75 +/- 0.23; type II: rho = 0.35 +/- 0.11; type III: rho = 0.15 +/- 0.08; P < 0.01). The ability to distinguish different AF episodes was assessed by designing a classification scheme based on a minimum distance analysis, obtaining an accuracy of 85.5%. The algorithm was able to discriminate among AF types even in presence of few depolarizations as no significant rho changes were observed by reducing the signal length down to include five LAWs. Finally, the capability to detect transient instances of AF complexity and to map the local regularity over the atrial surface was addressed by the dynamic and multisite evaluation of rho, suggesting that our algorithm could improve the understanding of AF mechanisms and become useful for its clinical treatment.

Use of global atrial fibrillation organization to optimize the success of burst pace termination

OBJECTIVES

The purpose of this study was to determine if burst atrial pacing would have an effect on terminating atrial fibrillation.

BACKGROUND

We hypothesized that frequency domain analysis of a filtered wide bipolar atrial electrogram describes the global organization of atrial fibrillation (AF) and should vary over time. Timing burst pacing to periods of high organization of AF should promote regional atrial conduction block and terminate AF.

METHODS

Nine dogs were conditioned with rapid atrial pacing for 48 h. Electrogram recordings were made from a wide right atrium (RA) to left atrium (LA) bipole and digitally filtered. A fast-Fourier transform was performed every 0.5 s on a sliding 2-s window, and the organization index (OI) was calculated as a ratio of the area of the first four harmonic peaks to the total power of the spectrum. Organization indexes >0.5 indicated more organized AF activity. Right atrium and LA burst pacing (burst) (cycle length 50 ms, 9.9 ms, 9.9 mA, 1 to 4 s) was performed through decapolar catheters. Burst was either random or synchronized to OI >0.5.

RESULTS

Burst termination was attempted 1,814 times (889 OI sync, 925 random) and succeeded in seven of nine dogs. Burst had an overall success rate of 11.1% versus 6.3% for random (p < 0.0003). Biatrial pacing had the highest efficacy for terminating AF, with a success rate of 16.5% for OI sync versus 8.2% for random (p < 0.0001).

CONCLUSIONS

Timing the delivery of the burst pace when the OI is >0.5 increases the efficacy of burst pace termination of AF. Biatrial pacing is more effective than either RA or LA pacing alone.