false
Catalog
Mapping Atrial Fibrillation in 2020: Key Updates, ...
Body Surface Mapping
Body Surface Mapping
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Good morning, my name is Felipe Atienza and I'm clinical electrophysiologist at the Gregorio Marañón Hospital in Madrid, Spain. I'm going to present the results of our group in the field of body surface mapping of atrial fibrillation. The following are my disclosures. This is the outline of my presentation. Electrical activities in the electrogram recordings during atrial fibrillation is the manifestation of sources that exhibit high frequency of activation and reentrant activity that could be effectively terminated by ablation. In these cases, AF driving sources detection was performed using different multipolar catheters and signal processing methods to determine dominant frequency and sequence of activation. However, current diagnostic tools do not always permit unequivocal determination of the mechanisms responsible for atrial fibrillation. Electrogram-based contact mapping is limited by the recording's configuration, number and composition of the electrodes and additional annotation tools. Therefore, the identification of atrial fibrillation sources is difficult by the sequential nature of mapping hindering global atrial activation that reduced the sensitivity and specificity of mapping. Aiming to obtain a global assessment of the atrial behavior, we perform non-invasive evaluation of the electrical status of the heart using body surface potential mapping. This method is an extension of the standard ECG using a large number of surface unipolar electrodes, recording the cardiac potential on the body surface, not on the heart, by solving the forward problem of the electrocardiography. Here we present non-invasive location of maximal frequency sites and rotational sites in atrial fibrillation patients. In panel A, we show the body surface recording with black arrow indicating the highest dominant frequency site in the middle of the posterior wall. In panel B, simultaneous intracardiac recording shows the highest dominant frequency site located at the less superior pulmonary vein, close to the torso site shown above. In panel C, unfiltered and filtered surface face maps of the unipolar voltage time series showing stable and longer lasting rotor episodes only after bandpass filtering. Here we show the correlation between surface and intracardiac recordings. Panels A and B show patients with left-to-right and right-to-left dominant frequency gradient in the intracardiac recordings. Panels C and D show summary maps of the torso projections of the matching portions in blue with a mean frequency difference between surface dominant frequencies and intracardiac dominant frequencies below 0.5 Hz. Panels E and F depict the correlation plot in patients with left-to-right and right-to-left dominant frequency gradient. Such a strong, site-specific correspondence of frequencies enables reliable determination of atrial frequency gradients noninvasively. Next, we aim to determine the location of atrial fibrillation sources using surface recording. Electrocardiographic imaging computes potentials on the epicardial surface of the heart from recorded body surface potentials, solving the inverse problem of electrocardiography. This constitutes an inverse solution to Laplace equation, which describes the electric potential field in the volume between the heart surface and the body surface. Electrocardiographic imaging provides maps of cardiac electrical excitation in relation to the anatomy of the heart. ECGI computes potentials on the epicardial surface of the heart from recorded body surface potentials. Then, the recorded torso potential and the geometrical information of the heart obtained from the MRI imaging provide the input data for the algorithm. Site-specific 3D atrial geometry and exact locations of all body surface electrodes on the patient's torso are required. Then, ECGI constructs electrograms on the epicardium to obtain the epicardial activation sequence and repolarization pattern. The presented slide shows the clinical validation of the inverse problem solution method to non-invasively estimate epicardial dominant frequency regions distribution. In panel A, shows bilateral dominant frequency map obtained using multipolar basket catheters in the left and the right atria, with highest dominant frequency site located at the right atrial appendage. In panel B, shows surface dominant frequency maps with maximal dominant frequency at the torso site closest to the right atrial appendage. In panel C, inverse computed dominant frequency maps calculated from the surface recordings showed excellent correlation with endocardial recordings in terms of maximal dominant frequency location and values. We used computational models to investigate the accuracy of ECGI during different electrical patterns. In panel A, we compared correlation coefficients of atrial electrical activity during sinus rhythm and complex atrial fibrillation with inverse computed signals of the torso. In this setting, we found a better reconstruction of sinus rhythm potentials as compared to complex atrial fibrillation patterns. As shown in panel B, the decrease in correlation coefficients was due to a simplification in the electric potentials at a distance from the hard surface, because of mutual cancellation of the electric field of waveforms with opposed directions, that cannot be retrieved when solving the inverse problem. However, identification of the sites maintaining atrial fibrillation by using noninvasive imaging is overall quite satisfactory, in spite of the poor reconstruction of the electrogram shapes. As shown in panels A and B, the concordance in the highest dominant frequency region of the electrograms and their inverse computed counterpart was 85%, and the distance between the centers of rotation of rotors on the electrograms versus inverse computed phase maps was 0.7 centimeters. Inverse computed phase maps were notably simpler than electrogram phase maps, but this didn't prevent the identification of the atria harboring rotors by using ECGI, which was satisfactory in more than 90% of the cases. In panel C, markers of atrial fibrillation driver location were found to be robust against uncertainties inherent to inverse problem reconstruction, such as the presence of electrical noise or an inaccurate estimation of the atrial geometry. For extreme signal-to-noise cases, the combination of information of rotor location and the highest dominant frequency site was able to successfully locate the rotor driving site. We performed mathematical simulations of atrial activity maintained by rotors, random activity, and atrial ectopies with SCAR, acting as electrical barriers that force wave breaks and turns to appear. Under this scenario, wave breaks during random activity may appear as phase singularities. However, imposing conditions on the linearity of the spatial phase transition improved the specificity of phase maps for the identification of rotational sites. An additional imposition of lifespan of phase singularities is also required to avoid wave turns to be detected as rotational sites, such those occurring during atrial tachycardia with SCAR. In that situation, by requiring a lifespan of at least 1.5 or 2 turns, most non-rotational singularity sites will be discarded. Despite the limitations of ECGI accuracy during atrial fibrillation, non-invasive imaging has shown to be a valuable tool for guiding ablation procedures, with a long-term success for persistent AFib of 80%. However, the adoption of ECGI as a reference technique by electrophysiologists is limited to very few hospitals in the world. Current methodological implementation of the system is preventing its wider adoption, such as the need of the patient to be in atrial fibrillation, being monitored a few hours in advance to the intervention, and the requirement to wear an electrode vest inside the CT scan. We used body surface potentials obtained from both mathematical simulations and patient data and solved the inverse problem with deviation in atrial position that differed from the original location. We found that the curvature of the L-curve was larger when the atria was located at the correct position. According to these results, errors in location of the atria inside the thorax below 1 cm would result in errors in the estimation of the rotational sites below 1 cm, which might be clinically relevant for guiding the identification of ablation sites in patients referred for ablation. We performed the first systematic evaluation of ECGI with simultaneous multipolar intracardiac recordings during atrial fibrillation. Body surface recordings along the torso were simultaneously obtained, together with multipolar high-density endocardial recordings from basket catheters from both atria, and applied phase mapping using the same identification method. Panel A shows the correlation between driver location in the right atrial appendage using intracardiac recordings and noninvasive imaging. Panel B shows the good correspondence in overall re-entrant measurements between intracardiac and body surface mapping techniques, with a positive, though modest, correlation on individual comparisons in terms of exact anatomical position. Next, we evaluated AFib complexity in relation to the acute efficacy of ablation. Panels A and B show an example of an AFib termination episode during driver-guided ablation, showing simultaneous intracardiac and body surface map with termination to sinus rhythm during ablation at the base of the left atrial appendage. Panel C shows the body surface re-entrant activity maps for representative patients in whom driver-guided ablation did and did not acutely terminate AFib. Panel D shows that patients with driver-guided ablation termination had less global noninvasive re-entrant regions than those in which AFib did not terminate. Therefore, this parameter is a reasonable predictor of guided ablation acute success, with an ROC curve of 0.88. We further developed a new score to evaluate the relationship between atrial fibrillation complexity and the long-term outcome of ablation. This score included ECGI parameters such as the highest dominant frequency distribution, singularity point density, and PUSY entropy, and performed a regression modeling to describe the six-month outcome after ablation. In panel A, patient 6 with a successful ablation had a CS score of 0.1, with highest dominant frequency and rotors located at the pulmonary vein area, and low entropy values. In contrast, patient 20 with an unsuccessful ablation had a CS score of 0.5, as shown by more complicated maps with similar presence of highest dominant frequency in both atrial cavities, multiple rotor sites, and high entropy values. As shown in panel B, CS score significantly improved the standard clinical classification to predict the efficacy of ablation. In summary, non-invasive cardiac imaging currently enables accurate evaluation of atrial fibrillation behavior of electrogram complexity. The ability to perform ECGI without requiring CT imaging will allow the identification of atrial fibrillation driving sources in a white sample of ambulatory patients. Accurate determination of atrial fibrillation complexity will establish the most indicated treatment, regardless of the clinical classification assignment. And finally, this will allow to personalize the ideal treatment strategy for each individual atrial fibrillation patient. Finally, I would like to acknowledge members of the following institutions, Vital Guerrero Marañón, Universidad Politécnica Valencia, Universidad Carlos III de Madrid, University of Michigan, and Stanford University, for their contribution to the works presented in this talk. Thank you very much for your attention.
Video Summary
In this presentation, Felipe Atienza discusses the use of body surface mapping in the study of atrial fibrillation (AFib). AFib is characterized by irregular electrical activity in the heart, and identifying the sources of this activity can help guide treatment. Traditional methods for mapping AFib have limitations, so Atienza and his team have turned to non-invasive body surface potential mapping. This technique involves using a large number of surface electrodes to record the electrical potentials on the body surface and then using computational models to calculate the electrical activity on the heart's surface. Atienza presents results showing the correlation between surface and intracardiac recordings, as well as the identification of atrial fibrillation sources using surface recording. He also discusses the use of body surface mapping in guiding ablation procedures and predicting treatment outcomes. Overall, body surface mapping shows promise as an accurate and personalized tool for studying AFib and guiding treatment decisions.
Keywords
body surface mapping
atrial fibrillation
electrical activity
non-invasive technique
treatment guidance
Heart Rhythm Society
1325 G Street NW, Suite 500
Washington, DC 20005
P: 202-464-3400 F: 202-464-3401
E: questions@heartrhythm365.org
© Heart Rhythm Society
Privacy Policy
|
Cookie Declaration
|
Linking Policy
|
Patient Education Disclaimer
|
State Nonprofit Disclosures
|
FAQ
×
Please select your language
1
English