Abstract
BACKGROUND: Peak-to-peak bipolar voltage varies with electrode orientation, fractionation, and collision events. Novel, omnipolar mapping is less dependent on electrode orientation but has limited data in humans. OBJECTIVES: This study sought to compare bipolar peak-to-peak voltage with omnipolar maximum voltage (V(max)) during sinus rhythm in the left atrium of patients with persistent (PerAF) or paroxysmal atrial fibrillation (PAF). METHODS: Baseline voltage maps were generated with bipolar and omnipolar mapping in 20 patients undergoing de novo catheter ablation for PerAF or PAF and 9 patients with known scar from prior cardiac surgery, to validate voltage-based scar approximations. Low voltage was defined as <0.5 mV and scar <0.1 mV. Mean voltage was compared with unpaired t testing. Percent low voltage and scar were compared with chi-square testing. A point-to-point comparison was performed with Bland-Altman analysis. RESULTS: The mean age was 62.2 ± 9.9 years, 34% were women, and 41% had heart failure. Omnipolar mapping identified significantly higher mean voltage than bipolar mapping and classified less points as low voltage (PerAF: 32.90% vs 43.40%; PAF: 19.20% vs 25.60%) and scar (PerAF: 7.72% vs 12.10%; PAF: 4.03% vs 6.07%) (all P < 0.0001). Omnipolar V(max) displayed significant disagreement with bipolar by Bland-Altman analysis. Scar and low-voltage approximations were validated in atria with known scar, in which bipolar mapping overestimated the extent of low voltage (P < 0.0001) and scar (P < 0.0001). CONCLUSIONS: Omnipolar mapping identifies higher voltage and has greater specificity for the detection of low voltage and scar than conventional bipolar mapping in patients with PerAF or PAF.