Mitral Isthmus Ablation: Is the Conduction Block Completed? The Importance of the Marshall Bundle Epicardial Connections.

A 66-year-old man underwent electrophysiological study and catheter ablation for persistent atrial fibrillation. At the beginning of the procedure, sinus rhythm was restored by internal cardioversion. Circumferential pulmonary vein isolation was performed, and the electric isolation of all 4 pulmonary veins was confirmed. Subsequently, linear ablation of a left atrial roof and the mitral isthmus (MI) was performed. The MI abla-tion was applied from the 4 o'clock direction of the mitral annulus (MA) to the left-side pulmonary vein bottoms, and further radiofrequency applications were delivered within the coronary sinus (CS) opposite of the endocardial MI line. We confirmed the complete conduction block of the roof line, and the activation sequence of the CS during the left atrial appendage (LAA) pacing was changed from distal to proxi-mal (Figure 1A) to proximal to distal (Figure 1B) while the MI ablation was being performed. Furthermore, the conduction time from the distal CS electrodes (CS 1–2) to the LAA during CS 1 to 2 pacing was longer than the conduction time from the proximal CS electrodes (CS 7–8) to the LAA during CS 7 to 8 pacing after the MI ablation (Figure 1C and 1D). Was the conduction block of the MI completed? Commentary If complete MI block is achieved, counterclockwise activation around the MA will be observed during LAA pacing, 1 and the conduction time from the distal CS electrode to the LAA during distal CS pacing will be longer than the conduction time from the proximal CS electrode to the LAA during proximal CS pacing. 2 The present case fulfilled the above conditions, which indicated apparent bidirectional MI block. However, the LAA activation sequence during CS 1 to 2 pacing was different from the LAA activation sequence during CS 7 to 8 pacing (Figure 1C and 1D), which suggested 2 or more pathways for propagation from the CS to the LAA during CS pacing. Because the complete conduction block of the roof line was confirmed, we suspected the presence of an epicardial conduction pathway via a Marshall bundle (MB) bypassing the endocardial MI line and inserted a 2F octapolar electrode catheter (EPstar, Japan Lifeline Corp., Tokyo, Japan) into the vein of Marshall (VOM; Figure 2A and 2B). The activation sequence in the VOM was distal to proximal during LAA pacing (Figure 3A), and we suspected the presence of electric conduction from the LAA to the VOM via the distal MB–LA connection. Radiofrequency …