Background: The optimal technique for recording of the electrogram to detect complete cavo-tricuspid isthmus (CTI) block during radiofrequency ablation (RFA) of the typical atrial flutter (AFL) using non-contact mapping system remains unclear. The aim of this study was to investigate the characteristics of the local virtual unipolar electrogram recorded at the ablation line during coronary sinus pacing after RFA of the CTI. Methods: Non-contact mapping was performed in 8 patients with AFL. Non-contact virtual unipolar electrograms were obtained before and after CTI linear ablation. All unipolar electrograms were acquired with both wide-band (0.5-300 Hz and 0.5-IC 150 Hz) and narrow-band filtering (32-300 Hz). The unipolar electrogram measurements included both the peak-to-peak voltage and peak-negative voltage for both filter settings, and the morphological characteristics of the second component of double potentials. Results: Comparison of the electrogram voltage along the ablation line before and after ablation demonstrated: 1) peak-negative voltage reduction at the mid-portion of the ablation line by 0.5-IC 150 Hz filtering, and 2) the second component of unipolar electrogram morphologies of R, Rs or rSR were the most sensitive and speci-c recording technique to predict conduction block. Conclusions: Non-contact unipolar electrograms using the peak-negative voltage with 0.5-IC 150 Hz and a predominant R-wave pattern in the second component at the ablation line indicate complete CTI block, even in the presence of transcristal conduction.