Figure legends
Figure 1. Flowchart of patient enrollment and disposition in this study. AF = atrial fibrillation, AT = atrial tachycardia, EIVOM = ethanol infusion in the vein of Marshall, MI = mitral isthmus, PSVT = paroxysmal supraventricular tachycardia, RF = radiofrequency, VOM = vein of Marshall
Figure 2. (A) A thermometer was in the esophagus, guiding catheter, and balloon catheter placed in the great cardiac vein. The cineangiography while inflating a balloon catheter revealed the existence of the Marshall vein (black arrows). CS = coronary sinus, GCV = great cardiac vein, RAO = right anterior oblique, VOM = vein of Marshall (B) A guidewire is inserted into the VOM through an over-the-wire balloon. An ablation catheter and Advisor HD gridTM mapping catheter are in the left atrium. (C) An over-the-wire balloon is placed in the most distal site of the VOM. Cineangiography is performed to confirm that the over-the-wire balloon is correctly inserted in the VOM. The balloon inflation pressure is set at 4 atm to avoid leakage of the contrast medium. Following angiography, anhydrous ethanol of 2 mL was infused in the VOM over 2 min. (D) Immediately after the first infusion, the over-the-wire balloon was pulled back to the most proximal site of the VOM, 3 mL more ethanol was infused in the VOM over 2 min while inflating the balloon at 4 atm. (E) RF applications (30 W, 40 s) were first delivered in the endocardium of the MI line using a steerable introducer under LAA pacing. RF = radiofrequency, LAA = left atrial appendage, MI = mitral isthmus (F) If the MI line block was not fully achieved, additional RF applications (25 W, 30 s) were delivered in the CS. (G) If those extra applications failed to make a complete MI line, we approached the CS from the jugular vein to gain enough contact force.
Figure 3. (A) The schema shows reconnection of both endocardium (red arrow) and epicardium (blue arrow). The electrogram reveals double and parallel potentials, which consist of endocardium (dull) and epicardium (spiky) potential, which were conducteda from the distal to the proximal site of the CS under LAA pacing. (B) The schema represents an endocardium-only reconnection. The dull potentials conducted from the distal to the proximal site were observed. Meanwhile, the spiky potentials were recognized in the proximal-to-distal sequence. Note that these two potentials were not discreated enough. (C) The schema represents an epicardium-only reconnection. The spiky potentials are recognized as the proximal-to-distal sequence. The dull potentials were recognized as the proximal-to-distal sequence, which are blocked at the site of CS 3-4.
Figure 4. (A) A representative case (Patient 16 in the Supplementary data 1) of low-voltage area caused by EIVOM. Low-voltage and scarred area are set at <0.5 and <0.05 mV, respectively. EIVOM = ethanol infusion in the vein of Marshall, LIPV = left inferior pulmonary vein, LSPV = left superior pulmonary vein, MA = mitral annulus, RIPV = right inferior pulmonary vein, RSPV = right superior pulmonary vein (B) Point and box-whisker plots represent individual low-voltage areas of initial, just after EIVOM, and one hour after EIVOM in the 19 patients of the EIVOM/RF group. These were measured by two independent observers, excluding the authors. The height of the boxes corresponds to the interquartile range, and the horizontal lines on boxes represent median value.
Figure 5. Cumulative AF- or AT-free survivals of the first and re-do ablation cases are shown. Dotted line represents a three-month blanking period after the procedure. AF = atrial fibrillation, AT = atrial tachycardia, EIVOM = ethanol infusion in the vein of Marshall, HR = hazard ratio, RF = radiofrequency