Abstract Background There is increased interest in exploring substrate of atrial fibrillation (AF) and recent studies have shown that atrial conduction velocity (CV) reflects the electrophysiological phenomena and plays key role in perpetuating AF. The novel omnipolar technology (OT) use three unipolar and two orthogonal bipolar signals allow us to analyze the voltage, activation direction, and CV independent of catheter position. Objective The purpose of the present study was to evaluate if the left atrium (LA) CV which assessed by OT predict the any atrial tachy-arrhythmia (AT) recurrence after initial AF ablation. Methods We enrolled 58 consecutive patients who underwent initial pulmonary vein isolation (PVI) for drug refractory AF. (mean age: 68.2±10.2 years old, 32 male(55.1%), 32 persistent AF(55.1%), CHADS2 score:1.6±0.6). After electrical PVI was achieved and remained in sinus rhythm, the LA was mapped with high density mapping catheter (Advisor HD Grid, Abott USA) utilizing with the novel OT combined with Ensite X system (Abott, USA). The conduction distance and conduction times were measured from the earliest activation site to the latest activation site in LA (Figure 1). Each conduction velocity in anterior, posterior, and septal area were also calculated as conduction distance divided by conduction time. The LA voltage was also measured. Results During a mean follow up period of 316±143 days, the recurrence of AT after initial PVI was observed in 9 (15.5%) patients. There were no significant differences in baseline demographics between AT recurrence and non-recurrence group (mean age:66±13.3 vs 68.5±10.5 years old P=0.66, CHADS2 score: 1.5±0.57 vs 1.68±0.62 P=0.59, proportion of persistent AF: 15.4% vs 15.5% P=0.82, LA volume: 102±35.9 vs 93.6±37.3ml P=0.63). The earliest activation site in LA were identified in 3 areas including anterior site of right pulmonary vein (51.7%), Bachmann bundle area (41.4%), and antero-septal area (6.89%) (Figure 2). In patients with AT recurrence, the CV in LA was significantly slower than those without AF recurrence(0.61±0.21 vs 0.81±0.16 m/s, P=0.036) whereas the whole LA voltage was not (1.52±0.75 vs 2.22±1.06 mv, P=0.28). The CV in each LA portion are similar between recurrence and non-recurrence AT group. (anterior:0.57±0.24 vs 0.81±0.23 m/s, P=0.18, posterior: 0.87±0.18 vs 1.2±0.49 m/s P=0.36, septal: 0.89±0.76 vs 1.15±0.39 m/s, P=0.41). Conclusion The earliest activation site in LA was varied in each case. The whole CV in LA is a predictor of AF recurrence after PVI whereas the CV of each LA portion and LA voltage was not.