[The activation patterns during atrial fibrillation in an experimental model]

A Review on Atrial Fibrillation (Computer Simulation and Clinical Perspectives)

Atrial fibrillation (AF), a heart condition, has been a well-researched topic for the past few decades. This multidisciplinary field of study deals with signal processing, finite element analysis, mathematical modeling, optimization, and clinical procedure. This article is focused on a comprehensive review of journal articles published in the field of AF. Topics from the age-old fundamental concepts to specialized modern techniques involved in today’s AF research are discussed. It was found that a lot of research articles have already been published in modeling and simulation of AF. In comparison to that, the diagnosis and post-operative procedures for AF patients have not yet been totally understood or explored by the researchers. The simulation and modeling of AF have been investigated by many researchers in this field. Cellular model, tissue model, and geometric model among others have been used to simulate AF. Due to a very complex nature, the causes of AF have not been fully perceived to date, but the simulated results are validated with real-life patient data. Many algorithms have been proposed to detect the source of AF in human atria. There are many ablation strategies for AF patients, but the search for more efficient ablation strategies is still going on. AF management for patients with different stages of AF has been discussed in the literature as well but is somehow limited mostly to the patients with persistent AF. The authors hope that this study helps to find existing research gaps in the analysis and the diagnosis of AF.

Significance of the morphological patterns of electrograms recorded during ventricular fibrillation: an experimental study

Mapping techniques are used to study the significance of the morphological patterns of the electrograms (EGMs) obtained during VF in an experimental model. In 24 isolated rabbit heart preparations recordings were made of activation during VF using a multiple electrode (121 unipolar electrodes) positioned on the lateral wall of the left ventricle. Three types of activation maps were selected: (A) with functional block of an activation front; (B) with epicardial breakthrough; and (C) with a single broad wavefront without block lines. The EGMs were classified as negative (Q), positive-negative with a predominance of the negative (rS) or positive wave (Rs), and positive (R). In 60 type A maps the morphology in the zone limiting the block line corresponded to an R wave in 55 (92%) cases and to Rs in 5 (8%) cases. In 67 type B maps, the EGM in the earliest activation zone most often showed Q wave morphology (48 [72%] cases), followed by rS (18 [27%] cases), and Rs morphology (1 [1%] case); in no case was R wave morphology seen. Finally, in 78 type C maps the morphology corresponded to a Q wave in 15 (19%) cases, rS in 38 (49%), Rs in 24 (31%), and R in a 1 (1%) case. The differences between the three types of maps were significant (P < 0.0001). Q wave EGM sensitivity for indicating the existence of an epicardial breakthrough pattern was 72%, with a specificity of 89%, and positive and negative predictive values of 76% and 87%, respectively. R wave EGM sensitivity for indicating the existence of conduction block was 92%, with a specificity of 99%, and positive and negative predictive values of 98% and 97%, respectively. R wave morphology is highly sensitive and specific for indicating conduction block. EGM recordings with initial positivity predominance are infrequent in the earliest activation zones of epicardial breakthrough during VF. The recording of the EGM with Q wave morphology indicates centrifugal activation from the explored zone.

[Pitfalls in left atrial reduction. Superior neo-cava with pericardium]

Left atrial reduction is a surgical technique that has been proposed for eliminating chronic atrial fibrillation associated with mitral valve disease. A potential complication of the technique is the unnoticed rotation of the heart while the left atrium is anastomosed. Such an event makes it impossible to reconstruct the superior vena cava, leading to superior vena cava syndrome due to the rotation of that vessel. We report our experience with a case of left atrial reduction and rotation of the superior vena cava while it was being anastomosed. The complication was successfully resolved by placing an autologous pericardial tube between the two ends of the superior vena cava.

[Characteristics of atrial electrograms recorded in radiofrequency induced block lines in an experimental model]

AIM

To analyze and quantify atrial electrogram modifications following the induction of linear lesions in the atrial wall using radiofrequency ablation procedures.

METHODS

An epicardial multiple electrode (221 unipolar electrodes) was used in 12 Langendorff perfused rabbit hearts to analyze atrial activation before and after radiofrequency induction of a linear lesion in the left atrial wall. After confirming the existence of conduction blockade in the lesion zone by epicardial mapping and propagation vector analysis, six electrodes each were selected in the lesioned and non-lesioned zones in all experiments, comparing the amplitude, maximum negative slope and morphology of the electrograms in both zones, before (control) and after radiofrequency delivery.

RESULTS

Analysis of the reproducibility of the measurements in two consecutive cycles showed a variation of 1 +/- 5% for amplitude (NS) and 1 +/- 9% for maximum negative slope (NS). In the non-damaged zone, amplitude (105 +/- 22%) and slope (92 +/- 16%) (values normalized with respect to those recorded before radiofrequency) did not vary significantly following radiofrequency, and simple electrograms were the most frequent recordings (82 vs 83% in control; NS). Amplitude (19 +/- 7%, p < 0.001) and slope (24 +/- 11%; p < 0.001) decreased significantly in the lesion zone, as did the percentage of simple electrograms (6 vs 86% in control; p < 0,001). In this same zone the morphology could not be determined in 12% of the recordings, while multiple electrograms were obtained in 15% (vs 2% in control; p < 0.01), and the most frequent type corresponded to double electrograms (67 vs 12% in control, p < 0.001), with both components coinciding in time with atrial activation in the zones proximal and distal to the lesion line.

CONCLUSIONS

Electrograms recorded directly in radiofrequency induce block lines show a significant decrease in amplitude and maximum negative slope. Double electrograms predominate in these recordings, both components of which represent activation on either side of the lesion. In a small proportion of cases simple and multiple electrograms can also be recorded in the block line.

[Epicardial mapping of reentrant activation during ventricular fibrillation. An experimental study]

INTRODUCTION AND OBJECTIVES

High-resolution epicardial mapping was used in an experimental model to analyze reentrant activation during ventricular fibrillation.

METHODS

In 30 isolated Langendorff-perfused rabbit hearts, recordings were made of ventricular fibrillation activity using an epicardial multiple electrode. In the activation maps with reentrant activation patterns, determinations were made of the number of consecutive rotations, the maximum length of the central core, the area encompassed by the core and two electrodes surrounding it, and the cycle defined by reentrant activation.

RESULTS

Most of the activation maps analyzed showed complex patterns with two or more wave fronts that either collided or remained separated by functional block lines (514 maps, 86%). In 112 maps (19%) activation patterns compatible with epicardial breakthrough of the depolarization process were observed. Reentrant activity was recorded in 42 maps (7%) - the maximum number of consecutive rotations being 3 (mean = 1.3 +/- 0.5). The maximum length of the central core ranged from 3 to 7 mm (mean = 5 +/- 1 mm), while the area encompassed by the central core plus two electrodes surrounding it ranged from 35 to 55 mm2 (mean = 45 +/- 6 mm2). The reentrant cycle length (mean = 47 +/- 8 ms) showed a linear relation to the maximum length of the central core reentry (cycle = 4.52 x length + 24.6; r = 0.7; p < 0.0001).

CONCLUSIONS

a) Epicardial mapping allowed the identification of reentrant activation patterns during ventricular fibrillation in the experimental model used; b) the reentrant activity detected is infrequent and unstable, and c) a linear relation exists between the duration of the cycles defined by reentrant activity and the maximum length of central core reentry.

Temporal organization of atrial activity and irregular ventricular rhythm during spontaneous atrial fibrillation: an in vivo study in the horse

INTRODUCTION

Atrial fibrillation (AF) is common in healthy horses. We studied the temporal organization of AF to test the hypothesis that the arrhythmia is governed by a high degree of periodicity and therefore is not random in the horse. Further, we surmised that concealed conduction of AF impulses in the AV node results in an inverse relationship between AF frequency and ventricular frequency.

METHODS AND RESULTS

Fast Fourier transform (FFT) analysis of atrial activity was done on signal-averaged ECGs (n = 11) and atrial electrograms (n = 3) of horses with AF at control (C), after quinidine sulfate (22 mg/kg by mouth every 2 hours) at 50% time to conversion (T50), and immediately before conversion (T90) to sinus rhythm. FFT always revealed a single dominant frequency peak. The mean dominant frequency decreased until conversion (C = 6.84 +/- 0.85 Hz, T50 = 4.87 +/- 1.5 Hz, T90 = 3.41 +/- 1.18 Hz; P < 0.001). Mean AA intervals (n = 500) gradually increased after quinidine. Mean RR intervals (n = 500), standard deviation of the mean (SDM), Poincaré plots, and serial autocorrelograms (SACs) of 500 RR intervals were measured at C and T90 to determine the ventricular response to AF and quinidine-induced changes in the variability of the ventricular response. Mean RR interval and SDM were reduced after quinidine (C = 1431 +/- 266 msec and 695 +/- 23 msec; T90 = 974 +/- 116 msec and 273 +/- 158 msec, respectively; P < 0.01). Poincaré plots and SAC at C and at T90 revealed a significant correlation of consecutive RR intervals typical of a system with a deterministic behavior. At T90, the variability of RR intervals was reduced and the overall periodicity of RR intervals was increased after quinidine administration.

CONCLUSION

In the horse, AF is a complex arrhythmia characterized by a high degree of underlying periodicity. The inverse AA-to-RR interval relationship and reduced variability of RR intervals after quinidine suggest that the ventricular response during AF results from rate-dependent concealment of AF wavelets bombarding the AV node, which nevertheless results in a significant degree of short-term predictability of beat-to-beat changes in RR intervals.