Session S32.2

Non-Invasive Estimates of Left Atrial Activation in a Patient with Dissociated Left Atrial Tachycardia following Ablation of Atrial Fibrillation

PG Platonov*, I Nault, M Stridh, F Holmqvist, M Haissaguerre

Lund University
Lund, Sweden

Introduction: Interatrial frequency gradient is used to guide catheter ablation of atrial fibrillation (AF) but reliable tools for its non-invasive estimation are lacking. Atrial fibrillatory cycle length (AFCL) in lead V1 has shown good agreement with right atrial AFCL in earlier studies, however robust, non-invasive estimates of left atrial fibrillatory activity are lacking. Previous attempts to non-invasively assess the left atrial AFCL have been hampered by lack of major differences in the AFCL between the right and left atrium. The objective of this case report was to identify which surface ECG leads that most closely reflect left atrial fibrillatory activity in a patient with dissociated right and left atrial rhythms.
Methods: 12-lead ECG was recorded simultaneously with electrograms from the right and left atrial appendages (RAA/LAA) in a 78 year old male patient who underwent catheter ablation of AF. As a result of ablation, AF was converted to sinus rhythm (47 bpm) but atrial tachycardia persisted in the LAA that was electrically isolated from the rest of the atria thus allowing identification of the LAA AFCL in the surface ECG leads. Atrial fibrillatory frequency spectra have been calculated from all 12 leads using spatiotemporal QRST cancellation and Welch periodogram. The dominating AFCL in the surface ECG leads was subsequently compared with the LAA AFCL.
Results: 1). RAA-CL=1276 ms (47 bpm), LAA-AFCL=252 ms (238 bpm). 2) Frequency spectra in leads II, aVF, aVR, V1 and V6 demonstrated the presence of dominating AFCL at 238 ms, 14 ms shorter that LAA-AFCL. AFCL-V2 was 228 ms while in the remaining chest and extremity leads the difference in regard to LAA-AFCL exceeded 80 ms. 3) Among the leads showing good agreement with LAA-AFCL, the dominating AFCL peaks were most distinct in the leads V1 and aVR.
Conclusions: Lead V1 reflects not only right but also left atrial fibrillatory activity and can therefore not be used for assessment of interatrial gradient but rather as a crude estimate of ‘global’ atrial fibrillatory rate. Lead aVR demonstrates easily definable dominating frequency peak in the frequency spectrum corresponding to the LAA activity but has not been tested for AFCL measurements using power frequency spectrum earlier. Informative value of frequency content in leads V6, II and aVF demands further investigations.

(Abstract Control Number: 39)