P436
The impact of filter settings on morphology of unipolar fibrillation potentials

N.M.S. de Groot, MD, PhD was supported by CVON-AFFIP (914728), NWO-Vidi (91717339) and Biosense Webster USA (ICD 783454).

Using unipolar atrial electrogram morphology as guidance for ablative therapy is regaining interest. Although standardly used in clinical practice during ablative therapy, the impact of filter settings on morphology of unipolar AF potentials is unknown.

This study aims to elucidate consequences of high-pass, low-pass and notch filtering on unipolar atrial fibrillation (AF) potentials.

Thirty different filter settings were applied to high-resolution epicardial AF potentials recorded from ten patients. . Local activation times were determined by marking deflections with slope ≤-0.05mV/ms and amplitude ≥0.3mV. Fibrillation potentials were analyzed for peak-to-peak amplitude (mV), number of deflections (f) and fractionation delay time (FDT, interval between first and last deflection), and classified as single potential (SP, f = 1), double potential (DP, f = 2) or complex fractionated potential (CP, f ≥ 3).

In total, 3000 seconds of AF recordings were analyzed, containing 255,7045 fibrillation potentials. Changing high-pass filtering from 0.5 up to 100 Hz decreased the number of detected fibrillation potentials (with 25-65%), median deflection amplitude (min-max 0.59-0.96 to 0.44-0.57 mV), percentage of DP (25.8-32.9 to 22.5-31.6%) and CP (15.9-36.0 to 7.3-37.8%) as well as median FDT (16-25 to 11-15 ms) (all p < 0.01). Gradually changing low-pass filtering from 400 Hz to 20 Hz induced an exponential decrease in fibrillation potentials (to 0%), percentage of DP (25.8-32.9 to 0%) and CP (15.9-36.0 to 0%), whereas deflection amplitude, percentage of SP and median FDT exponentially increased (0.59-0.96 to 1.82-2.39 mV, 36.1-57.6 to 100%, 16-25 to 33-38 ms, respectively) (all p < 0.01). Notch filtering at 50 Hz decreased the number of detected fibrillation potentials (with ∼1%), median deflection amplitude (0.59-0.96 to 0.59-0.95 mV) (both p < 0.01), whereas the percentage of CP increased (15.9-36.0 to 16.2-37.0%, p = 0.016).

Filtering significantly impacted  of unipolar fibrillation potentials and decreased the number of detected potentials, becoming a potential source of error in identification of LATs, low-voltage areas, fractionated potentials and thus ablative targets during mapping.

Abstract Figure. Impact of filtering on AF morphology