Search

The objective of this study was to investigate the prevalence, electrophysiologic properties, and clinical implications of dissociated pulmonary vein (PV) activity after PV antrum isolation (PVAI) in patients with paroxysmal atrial fibrillation (AF). One hundred seventy-three consecutive patients (61 ±10 years old, 141 men) with drug-refractory paroxysmal AF who underwent AF ablation were analyzed. After identification of arrhythmogenic foci, PVAI was performed in all patients. Of the total 346 isolated ipsilateral PVs, 97 (28.0%) were silent, 35 (10.1%) demonstrated isolated ectopic beats, 209 (60.4%) demonstrated a regular ectopic rhythm, and 5 (1.4%) demonstrated fibrillatory activity. The culprit thoracic vein was identified in 77 patients (44.5%). After isolation of ipsilateral PVs, venous activity was observed in 68 (79.1%) and 178 (68.5%) PVs among the 86 PVs with AF triggers and 260 PVs without AF triggers, respectively (p = 0.06). There was no significant difference in the incidence of acute PV reconnections exposed by adenosine triphosphate between the 97 silent ipsilateral PVs and 209 ipsilateral PVs with dissociated PV activity after the PVAI (20.6% vs 19.1%, p = 0.78). After a mean follow-up of 48.7 ± 7.9 months there was no significant difference in rates of freedom from atrial tachyarrhythmias after a single procedure between patients with and those without dissociated activity (62.1% vs 63.3%, p = 0.74, log-rank test). In conclusion, although dissociated PV activity appearing after PV isolation is an important electrophysiologic finding to prove bidirectional conduction block between the left atrium and the PV during the procedure, the clinical implications might be limited.

This preliminary study demonstrates that the conduction velocity in the venae cavae and the stimulation threshold and refractory period of the venae cavae, azygos and pulmonary veins in dogs were very similar to those found in atrial muscle. It is quite likely that atrial muscle extends for a distance into these veins and that the muscle receives vagal innervation. Superior vena caval pacing was performed in 1 patient.

Atrial and thoracic vein mapping in patients showed that the earliest site of atrial activity ranged from a low to high position along the sulcus terminalis and that cardiac muscle extended for 2 to 4 cm into all of the thoracic veins except the inferior vena cava. To understand the epicardial wave forms, experiments were carried out in open and closed chest dogs. Atrial muscle extended superiorly into the superior vena cava of the dog to a far greater degree than it extended into this vessel in man. In some dogs, cardiac muscle in the azygous vein extended to the spine, and at this site the heart could be paced; no cardiac electrical activity occurred in the azygous vein of man. The shape of most wave forms (a rapid deflection superimposed on a prolonged negative curve) was due to a superposition of effect of a large excitation wave in the atrium and a smaller excitation wave in the thoracic veins. Ultrastructural studies revealed atrial muscle in the thoracic veins of the dog. The extension of cardiac muscle into the thoracic veins of human embryos was far greater than that found in children and adults. In a patient with a persistent left superior vena cava, cardiac excitation progressed far above the heart to the junction with the innominate vein, a position from which the heart could be paced. The results suggest that the Lewis theory of atrial flutter—that is, that the venae cavae present an obstacle around which excitation waves progress—is untenable since wave fronts propagate in the venous muscle in a manner that quickly renders the tissue refractory by the merging and collision of waves.