The P-wave Time-domain Significant Features to Evaluate Substrate Modification After Catheter Ablation of Paroxysmal Atrial Fibrillation

Aikaterini Vraka1, Vicente Bertomeu-González2, Leif Sornmo3, Roberto Zangróniz4, Raul Alcaraz5, José J Rieta1, Universitat Politecnica Valencia, 2Cardiology Department, Saint John's University Hospital, 3Lund University, 4Research Group in Electronic, Biomedical and Telecommunication Engineering, University of Castilla-La Mancha, 5University of Castilla-La Mancha


Background. P-wave duration (PWD) is undoubtedly one of the most popular indices for atrial fibrillation (AF) substrate assessment, vastly used to predict catheter ablation (CA) outcome. Most studies, however, focus on the analysis of the PWD indivisibly, ignoring the possibility of different roles of right (RA) and left atrium (LA) to the atrial wavefront propagation, illustrated by the first and second P-wave parts, respectively. This work aims to study the effect of CA on the time-domain characteristics of each P-wave part and define if RA and LA are equally affected by CA of paroxysmal AF (pAF).

Methods. Five-minute lead II recordings before and after the first-ever CA of 40 pAF patients were analyzed. P-waves were delineated and the following characteristics were calculated: PWD(on-off) of the entire P-wave and each P-wave part [RA: PWD(on-peak) and LA: PWD(peak-off)] and the time from P-wave onset or offset to the R-peak [PWD(on-R) and PWD(off-R), respectively]. The effect of heart rate (HR) fluctuations was additionally mitigated by an HR-adjustment. Pre/post-CA comparison was performed with Mann-Whitney U-test and median values were calculated. Correlations between PWD and the rest of the features were specified with Pearson's correlation.

Results. The effect of CA with (Delta:-17.96%, p=0.0442) or without HR-adjustment (Delta:-9.84%, p=0.0085) was significant at the entire PWD(on-off) and at the PWD(peak-off) (HR-adjustment: Delta:-27.77%, p=0.0268, without HR-adjustment: Delta:-22.03%, p=0.0250). The rest of the features did not vary significantly (p>0.05). All correlations were statistically significant with p<0.0001. The full PWD(on-off) showed a stronger correlation with RA, being r= 0.745-0.805, than with LA, where r=0.419-0.541.

Conclusions. Time-domain features of P-waves corresponding to RA are more strongly related to the entire P-wave. Nevertheless, only the P-wave part associated to LA is significantly affected by CA. That being so, studies are encouraged to incorporate subinterval analysis of the P-wave when assessing substrate modification after CA.