|

TYPE Original Research

PUBLISHED 25 July 2024
DOI 10.3389/fcvm.2024.1417029

Body surface mapping of

P-waves in sinus rhythm to
EDITED BY
Kyungmoo Paul Ryu,
Abbott, United States
predict recurrence following
REVIEWED BY
Vassilios Vassilikos,
cardioversion for atrial fibrillation
Aristotle University of Thessaloniki, Greece
Christopher Aldo Rinaldi, Ibrahim Antoun1*, Xin Li2, Ahmed Kotb3, Joseph Barker3,
St Thomas’ Hospital, United Kingdom
Akash Mavilakandy3, Ivelin Koev1, Zakariyya Vali1, Riyaz Somani1,3
*CORRESPONDENCE
Ibrahim Antoun
and G. André Ng1,3,4
[email protected] 1
Department of Cardiology, University Hospitals of Leicester NHS Trust, Glenfield Hospital, Leicester,
RECEIVED 13 April 2024 United Kingdom, 2Department of Engineering, University of Leicester, Glenfield Hospital, Leicester,
United Kingdom, 3Department of Cardiovascular Sciences, Clinical Science Wing, University of
ACCEPTED 15 July 2024
Leicester, Glenfield Hospital, Leicester, United Kingdom, 4Department of Research, National Institute
PUBLISHED 25 July 2024
for Health Research Leicester Research Biomedical Centre, Leicester, United Kingdom
CITATION
Antoun I, Li X, Kotb A, Barker J, Mavilakandy A,
Koev I, Vali Z, Somani R and Ng GA (2024) Background: Direct current cardioversion (DCCV) is used as elective and
Body surface mapping of P-waves in sinus
emergency rhythm control for atrial fibrillation (AF). We aimed to explore the
rhythm to predict recurrence following
cardioversion for atrial fibrillation. role of P-wave parameters measured during sinus rhythm using body surface
Front. Cardiovasc. Med. 11:1417029. mapping (BSM) in predicting successful DCCV for persistent atrial fibrillation
doi: 10.3389/fcvm.2024.1417029 (persAF) at 12 months.
COPYRIGHT Methods: This case–control study included 56 males >18 years old who
© 2024 Antoun, Li, Kotb, Barker, Mavilakandy, underwent DCCV for persAF. P-wave parameter collection after DCCV for AF
Koev, Vali, Somani and Ng. This is an open-
access article distributed under the terms of
was done using 128 unipolar leads. A band-pass filter of 1–50 Hz was utilised.
the Creative Commons Attribution License Corrected P-wave duration (PWDc), P-wave amplitude, and P-wave dispersion
(CC BY). The use, distribution or reproduction were measured to predict 12-month outcomes and time of recurrence.
in other forums is permitted, provided the
original author(s) and the copyright owner(s)
Results: The mean age was 64 ± 4 years, and 23 patients (44%) were on
are credited and that the original publication in amiodarone. The 12-month success rate was 44% (n = 23), while the rest
this journal is cited, in accordance with reverted to AF after 2.6 ± 0.4 months. The parameters were comparable between
accepted academic practice. No use,
distribution or reproduction is permitted
successful and failed DCCV in the entire cohort and patients not on amiodarone.
which does not comply with these terms. In patients on amiodarone, patients with failed arms had higher PWDc than
those with successful arms (188 vs. 150 ms, P = 0.04). Receiver operator
characteristic curve analysis for PWDc in the amiodarone cohort showed an area
under the curve (AUC) of 0.75 and P = 0.049. A recurrence cut-off >161 ms had
a sensitivity of 69% and a specificity of 100%, with a hazard ratio of 10.7,
P = 0.004. The parameters were not predictive of the time of recurrence.
Conclusion: In patients on amiodarone, increased PWDc measured using BSM was
associated with higher AF recurrence at 12 months following DCCV for persAF.

KEYWORDS

atrial fibrillation, body surface mapping, p-waves, cardioversion, electrical cardioversion

Introduction
Atrial fibrillation (AF) is the most common sustained arrhythmia worldwide,
increasing the risk of stroke and mortality (1, 2). AF could be caused by triggers or
substrate driven by electrical or structural remodelling. The recent European Society of
Cardiology guidelines suggested a rate or rhythm control approach to manage AF (3).
Rhythm control options include anti-arrhythmic drugs, catheter ablation, and direct
current cardioversion (DCCV). DCCV is particularly useful when the patient is

Frontiers in Cardiovascular Medicine 01 frontiersin.org

Antoun et al. 10.3389/fcvm.2024.1417029

hemodynamically unstable or there is an urgency to resume sinus acute success was defined as the immediate return to SR. Long-
rhythm (SR). The mechanism of DCCV was proposed to prevent term DCCV success was determined by the lack of AF or atrial
the maintenance of re-entrant tachycardia by the remaining flutter up to 1 year following DCCV. P-wave parameters
myocardial tissue after depolarisation of a critical mass (4). The measured using BSM were used to predict DCCV outcome at
1-year success rate of external DCCV for AF varied between 15% 12 months and recurrence time. Clinical details of DCCV,
and 47% (5–8). Predicting DCCV outcome after reversion to SR follow-ups, and medication were obtained by retrospective
has been explored in the literature. Symptom monitoring was analysis of clinical paper records. This study was a part of the
unreliable in predicting DCCV outcomes due to asymptomatic USURP-AF II study approved by the East Midlands–Leicester
AF (9). Predictors of DCCV outcomes in the literature include Central Research Ethics Committee. REC reference: 19/EM/003
demographics, the use of anti-arrhythmic medication (10), The selection criteria included the following:
cardiac imaging (11), and P-wave analysis (12).
P-wave represents the spatiotemporal convolution of the 1. Patients with persistent atrial fibrillation (persAF) who
depolarisation wavefront in the atria as collected in the torso. P-wave underwent external DCCV in Glenfield Hospital, Leicester,
parameters provide insights into atrial electrophysiology and essential UK, between January 2013 and May 2015
information regarding atrial remodelling, which triggers and 2. Patients >18 years old and who consented to the study
maintains AF (13). These parameters utilised 12-lead ECG and signal- 3. Planting anterior leads around the breast tissue in females
averaged ECG (SAEG) following initial successful cardioversion to could be challenging. Therefore, only male patients were
predict freedom of arrhythmia rate (Table 1). However, these studies selected.
have yet to utilise body surface mapping (BSM) to assess the same 4. Patients who had ECG-documented AF directly before DCCV
hypothesis. This study aimed at determining the value of 128-lead 5. Patients who were adequately anti-coagulated for 4 weeks before
BSM of P-waves in SR in predicting 12-month freedom from DCCV [therapeutic international normalised ratio (INR)].
arrhythmia rate after initially successful DCCV for AF.
P-wave parameter collection after DCCV for AF was done using 128
unipolar leads with 64 leads on the front and back of the torso
(Figure 1). The vest utilised electrodes from Biosemi (Amsterdam,
Methods The Netherlands). The electrodes record digital ECGs before,
during, and after DCCV. If DCCV was acutely successful, SR
This case–control study included 56 patients who underwent resumption would allow recording digital P-waves following
DCCV for AF between October 2013 and May 2015. DCCV DCCV. A MATLAB code was then applied to the digital BSM
data using a band-pass filter of 1–50 Hz. It anointed the isoelectric
TABLE 1 Demographics, clinical outcomes, and medication details of
study cohort. line and P-wave peak to measure the P-wave duration (PWD) and
amplitude. Also, it allowed the measurement of the P-wave
Outcome at 12 months Success Failure p-value manually by adjusting its start and end. The first four identifiable
(n = 23) (n = 29)
P-waves were measured and averaged in each lead. The
Demographics parameters measured include the following:
Age (years) 64 ± 5.6 63.6 ± 3.5 0.89
The following P-wave parameters were produced:
Acute success (%) 22 (100%) 28 (97%) 0.4
Immediate reversion to AF (%) 0 (0%) 6 (20%) 0.026
1. PWD: Distance from P-wave onset to offset. It demonstrates a
AF duration (months) 11.6 ± 3.2 13.5 ± 2.8 0.32
marker of atrial conduction.
Required more than 1 shock (%) 2 (9%) 1 (3%) 0.39
Conscious sedation (%) 21 (95%) 29 (97%) 0.83 2. P-wave amplitude (PWA): The area under the P-wave was
Previous DCCV (%) 5 (22%) 5 (17%) 0.53 estimated using the trapezoidal method by integrating the
DCCV within 12 months (%) 0 (0%) 4 (12%) 0.08 total area into a little trapezoid. It demonstrates atrial voltage.
Ablation within 12 months (%) 0 (0%) 4 (12%) 0.08 Regarding biphasic P-waves, the highest PWA absolute value
Diabetes mellitus (%) 3 (14%) 2 (7%) 0.41 between the positive and negative phases was accepted.
Heart failure (%) 4 (18%) 11 (37%) 0.15
Cerebrovascular event (%) 2 (9%) 2 (7%) 0.75
Ischemic heart disease (%) 1 (5%) 3 (10%) 0.48
Hypertension (%) 12 (46%) 17 (57%) 0.88
Indexed left atrial volume (ml/m2) 63.6 ± 1.9 73.7 ± 2.1 0.25
Body mass index (kg/m2) 33.6 ± 1.5 32.3 ± 1.4 0.6

Anti-arrhythmic drugs
Amiodarone 9 (39%) 14 (48%) 0.58
Sotalol 9 (41%) 8 (27%) 0.29
Flecainide 1 (5%) 0 (0%) 0.25
Bisoprolol 5 (23%) 12 (40%) 0.2
Diltiazem, verapamil 1 (5%) 1 (3%) 0.83 FIGURE 1
Anti-arrhythmic drug not stopped 18 (82%) 22 (73%) 0.42 Demonstration of real-time body surface mapping.
(long term)

Frontiers in Cardiovascular Medicine 02 frontiersin.org

Antoun et al. 10.3389/fcvm.2024.1417029

3. P-wave dispersion (PWDisp): The maximum difference frequency and percentage. The mean ± standard error of the
between P-wave durations. It demonstrates atrial mean was adopted to describe continuous parametric data.
depolarisation heterogeneity. Unpaired t-tests or Mann–Whitney U tests were utilised to
analyse unmatched data depending on the normality of the
Each measurement was averaged in all 128 leads to produce one distribution. Gubbs’ test was conducted to identify outlier
number representing a P-wave measurement in each patient measurements. The cut-off of P-wave parameters for the DCCV
(Figure 2). outcome was analysed using receiver operator curve (ROC)
In patients with immediate reversion to AF with <4 sinus beats, analysis. The detected cut-off was used in a Kaplan–Meier
P-waves were measured until AF reversion. Heart rate (HR) was also survival analysis to use the proposed cut-off in predicting DCCV
noted in every ECG. Amiodarone is known to adjust P-wave outcomes. P-value ≤0.05 represented statistical significance.
morphology (14). Therefore, P-wave analysis was done after
stratifying the patients based on amiodarone. It is well recognised
that changes in HR and cycle length may affect P-wave morphology
(15, 16). Previous studies have tried to tackle this by adjusting P-wave
Results
parameters to heart rate in regression analysis, a method that was not
The mean age was 64 ± 4 years. Three of the 56 patients involved
utilised in this study (17). As there are no current verified formulae
were in SR at their DCCV appointment and were excluded. One
to correct PWD for heart rate, we utilised a QT correcting formula to
patient was also excluded due to the lack of follow-up at 12 months
address this issue. Out of the formulae used in the literature, the
(Figure 3). The 12-month success rate was 44% (n = 23), while the
Hodges formula has been proposed to be the most reliable and was
rest reverted to AF after 2.6 ± 0.4 months. Patients’ clinical details
therefore utilised in this study to correct PWD for heart rate,
and demographics are provided in Table 2. Intra-observer
producing corrected P-wave duration (PWDc) (18). The intra-
variability was highest in PWA at 10% (0.05 mV), followed by
observer variability test anonymously analysed 20 BSMs on 2 days.
PWDisp at 8% (9.5 ms) and PWD at 5% (8.4 ms). Gubbs’ test did
not yield outlier values.
P-wave parameters were compared between successful and
Statistical analysis failed arms (PWDc: 163 vs. 178 ms, P = 0.29. PWA: 0.33 vs.
0.38 mV, P = 0.1. PWDisp: 24 vs. 27 ms, P = 0.35).
Statistical analysis was conducted using GraphPad Prism V9.3 After stratifying for amiodarone, the success rates in the amiodarone
(San Diego, CA, USA). Categorical variables were expressed as and non-amiodarone cohorts were 39% vs. 48%, P = 0.58, respectively.

FIGURE 2
Demonstration of P-wave parameters anointing using body surface mapping at the front 64 leads. The resulting parameters are P-wave duration,
amplitude, and dispersion. The red dot represents the beginning of the P-wave, while the green dot represents the end of the P-wave.

Frontiers in Cardiovascular Medicine 03 frontiersin.org

Antoun et al. 10.3389/fcvm.2024.1417029

FIGURE 3
Clinical outcomes of the cohort.

TABLE 2 Studies that correlated P-wave parameter with direct current cardioversion outcome.

Author and year AF n Follow-up ECG Parameter Recurrence change Recurrence cut-off p
Opolski et al., 1997 (19) PersAF 35 6 months SAECG PWD ↑ >137 ms <0.0001
Stafford et al., 1998 (20) PersAF (77%) 31 1 week SAECG P-wave energy ↑ >25% drop 0.03
Aytemir et al., 1999 (21) PersAF 73 6 months SAECG Filtered PWD ↑ >128 ms 0.001
Raitt et al., 2000 (22) PersAF 20 1 year SAECG PWD ↑ >130–140 ms 0.005
Guo et al., 2003 (23) PersAF 60 6 months SAECG Filtered PWD ↑ Nil <0.0001
Ehrlich et al., 2003 (24) No mention 111 1 week SAECG PWD ↑ >145 ms <0.001
Dixen et al., 2004 (25) PersAF 131 1 month SAECG PWD ↑ >160 ms 0.03
Dogan et al., 2004 (26) PersAF (45%) 64 6 months SAECG PWDisp ↑ >46 ms <0.001
Perzanowski et al., 2005 (27) PersAF 45 6 months SAECG PWDisp ↑ >80 ms 0.05
Budeus et al., 2005 (28) PersAF 141 1 year SAECG PWD ↑ >126 ms 0.0001
Başar et al., 2011 (29) PersAF 26 1 year 12 leads PWDisp ↑ 0.001
Gonna et al., 2014 (30) PersAF 77 1 month 12 leads PWD ↑ >125 ms 0.025
Blanche et al., 2014 (31) PersAF 133 9 months SAECG Nil Nil Nil Non-significant
Fujimoto et al., 2018 (32) PersAF 141 1 month 12 leads PWDisp ↑ Nil 0.001
Choi et al., 2021 (12) PersAF 272 2 months 12 leads PWD ↑ >134 ms 0.012
PTFV1 >50 mm.ms 0.011

PTFV1, P-wave terminal force in V1.

P-wave parameters were comparable in the non-amiodarone cohort DCCV outcomes (Figure 4). There was no correlation between P-
(172 vs. 167 ms, P = 0.82; PWA: 0.34 vs. 0.41 mV, P = 0.2; PWDisp: wave parameters and AF recurrence time in the entire cohort (PWDc:
26 vs. 24 ms, P = 0.58). In patients on amiodarone, those who had r = −0.23, P = 0.24; PWA: r = 0.33, P = 0.09; PWDisp: r = −0.13, P =
failed arms had higher PWDc than those with successful arms 0.52), the amiodarone cohort (PWDc: r = −0.02, P = 0.95; PWA: r =
(188 vs. 150 ms, P = 0.04), while PWA and PWDisp were comparable 0.5, P = 0.07; PWDisp: r = –0.23, P = 0.45), and the non-amiodarone
between both arms (0.3 vs. 0.4 mV, P = 0.09, and 21 vs. 20 ms, cohort (PWDc: r = −0.36, P = 0.24; PWA: r = −0.1, P = 0.74; PWDisp:
P = 0.94, respectively). ROC curve analysis for PWDc in the r = −0.12, P = 0.71).
amiodarone cohort showed an area under the curve (AUC) of 0.75
and P = 0.049. A recurrence cut-off >161 ms had a sensitivity of 69%
and a specificity of 100%. The Kaplan–Meier survival analysis using a Discussion
PWDc cut-off of 161 ms in the amiodarone cohort indicated a hazard
ratio of 10.7, P = 0.004. By contrast, the non-amiodarone and full Although other studies correlated P-wave parameters after
cohorts did not show a difference between cut-off and 12-month DCCV with outcome using signal-averaged ECG or 12-lead ECG

Frontiers in Cardiovascular Medicine 04 frontiersin.org

Antoun et al. 10.3389/fcvm.2024.1417029

delay in patients prone to AF recurrence (38). Furthermore, a

previous study demonstrated shorter PWD in patients with AF
successfully treated with amiodarone (14). These results were not
evident in patients who were not on amiodarone. The reason behind
this finding is unclear. One reason could be that patients did not
have refractory AF significant enough to be on amiodarone, leading
to the lack of LA conduction delay. This may be possible to detect
with imaging modalities and is advised in future studies. Another
reason could be the potential small size to detect a significant effect.
As for PWA, one would expect a correlation between low PWA
(caused by fibrosis) and DCCV 1-year failure. This corresponded to
the PVI studies that proposed this (39, 40). Although only one study
correlated reduced PWA with AF recurrence after internal
cardioversion (17), DCCV (external) was used in this study, and all
studies in Table 2 did not show similar findings. It was unclear why
PWA was predictive of outcome in internal DCCV but not external
DCCV. However, one theory is that external cardioversion could
have caused lower atrial stunning, not affecting PWA, unlike internal
cardioversion (41). Further randomised trials would be beneficial in
establishing a mechanism.
PWDisp reflects inhomogeneous atrial refractoriness and AF
vulnerability (42). It was not correlated to the DCCV 12-month
outcome in this cohort. According to previous studies, amiodarone
decreases PWDisp because of increased atrial repolarisation (14, 43).
Following the hypothesis that the amiodarone cohort had more LA
remodelling, it is possible that the increased PWDisp from
remodelling was decreased to normal by the amiodarone effect.
Patients not on amiodarone did not have enough remodelling to
cause a notable PWDisp effect predictive of DCCV outcomes per
previous studies (Table 2). These results warrant further investigation
into predicting DCCV outcomes in patients on and off amiodarone.
Further studies should directly compare BSM, signal-averaged
FIGURE 4
Kaplan–Meier survival analysis for the corrected P-wave duration electrocardiogram (SAECG), and 12-lead ECG in utilising P-wave
cut-off in all cohorts. parameters to predict 12-month DCCV outcome for AF.

(summarised in Table 2), this is the first study to perform this using Conclusion
BSM of P-waves. The 1-year success rate in our cohort falls into the
higher range of previous studies demonstrating relatively good Predicting DCCV outcome using P-wave parameters measured in
outcomes in our centre. However, the inclusion of males might SR using BSM was only feasible in patients on amiodarone. In this
have positively affected the one-year success rate (33). Unlike cohort, increased PWDc in SR directly following DCCV was
former studies, demographics in this cohort were not predictive associated with failed DCCV at 12 months. PWDc >161 ms was 100%
of the outcome (34, 35). One would expect a positive correlation specific for AF recurrence by 12 months after initial successful DCCV.
of AF risk factors with AF recurrence after DCCV. It was noted This could serve as a marker for considering an early rate control strategy.
that this cohort only involved male patients. This could cause
selection bias and affect the rest of the demographics. Therefore,
further studies with a random selection of subjects would be Limitations
advised to investigate predictive factors (36, 37).
Regarding P-wave analysis, to the best of our knowledge, no This is a single-centre retrospective study with AF recurrence
previous study has correlated P-wave BSM with 12 months of detected using 12-lead ECG or Holter monitoring. Long-term
DCCV outcome and assessed recurrence time. Stratification by being monitoring was not done, and the AF burden was not evaluated.
on amiodarone was conducted to address its effects on P-wave This could have missed sub-clinical and micro-AF episodes. The
parameters. Increased PWDc after DCCV in the amiodarone cohort relatively low sample size with post-hoc power analysis of 67% and
was associated with 12-month failure with high specificity. This was 71% in the full cohort and amiodarone cohort to detect a significant
in keeping with previous studies in Table 2. This can be explained difference in PWDc may lead to a type-2 error. Therefore, future
by left atrium (LA) remodelling, fibrosis, and intra-atrial conduction studies using BSM with a higher number and pre-study sample size

Frontiers in Cardiovascular Medicine 05 frontiersin.org

Antoun et al. 10.3389/fcvm.2024.1417029

calculation are advised. Direct comparison between BSM, 12-lead JB: Writing – review & editing. AM: Writing – review & editing. IK:
ECG, and SAECG was not conducted and is suggested in future Writing – review & editing. ZV: Writing – review & editing. RS:
studies. This study only included male patients, which can cause Writing – review & editing. GA: Supervision, Writing – review & editing.
selection bias. Although not impossible, including female patients
would require the adjustment of lead positions, making comparisons
challenging. Flecainide and sotalol used in our cohort affected PWD Funding
(44, 45). Furthermore, patients stopping their anti-arrhythmic drugs
were included in the analysis. Future studies matching patients by The authors declare that no financial support was
anti-arrhythmic drugs and their cessation are needed to limit received for the research, authorship, and/or publication of
confounding. The Hodges formula is currently not verified as a this article.
methodology in the literature to correct PWD for HR. A future
dedicated study would be useful to confirm the utility of this formula
for the benefit of future studies utilising PWD. One of the main
limitations of this study is the age of the data utilised. The data were
Acknowledgements
collected between 2013 and 2015, which may affect the applicability
This study is part of the research portfolio supported by the
of the findings to current clinical practice. In future studies,
Leicester National Institute for Health and Care Biomedical
incorporating recent data with longer follow-ups is advised. This Research Centre’s cardiovascular theme.
study did not measure potential relevant pre-DCCV parameters,
including AF cycle length and AF coarseness. The study did not
utilise magnetic resonance imaging to evaluate LA fibrosis, which is
recommended for future studies. Conflict of interest
AK is supported with a clinical research fellowship from Abbott.
Data availability statement GAN is supported by a British Heart Foundation Programme Grant
(RG/17/3/32,774) and the Medical Research Council Biomedical
The raw data supporting the conclusions of this article will be Catalyst Developmental Pathway Funding Scheme (MR/S037306/1).
made available by the authors without undue reservation. GAN also discloses consultancy fees, speaker honorarium fellow
support from Biosense Webster, research fellow funding from
Abbott, and consultancy fees from Catheter Precision.
Ethics statement The remaining authors declare that the research was conducted
in the absence of any commercial or financial relationships that
The studies involving humans were approved by the Leicester could be construed as a potential conflict of interest.
Central Research Ethics Committee. The studies were conducted
in accordance with the local legislation and institutional
requirements. The participants provided their written informed Publisher’s note
consent to participate in this study.
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their affiliated
Author contributions organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
IA: Writing – original draft, Writing – review & editing. XL: claim that may be made by its manufacturer, is not guaranteed
Software, Writing – review & editing. AK: Writing – review & editing. or endorsed by the publisher.

References
1. Wolf PA, Mitchell JB, Baker CS, Kannel WB, D’Agostino RB. Impact of atrial 4. Zipes DP, Fischer J, King RM, Nicoll A, Jolly WW. Termination of ventricular
fibrillation on mortality, stroke, and medical costs. Arch Intern Med. (1998) 158 fibrillation in dogs by depolarizing a critical amount of myocardium. Am J Cardiol.
(3):229–34. doi: 10.1001/archinte.158.3.229 (1975) 36(1):37–44. doi: 10.1016/0002-9149(75)90865-6
2. Wyndham CR. Atrial fibrillation: the most common arrhythmia. Tex Heart Inst J. 5. Van Gelder IC, Crijns HJ, Tieleman RG, Brügemann J, De Kam PJ, Gosselink
(2000) 27(3):257. AM, et al. Chronic atrial fibrillation: success of serial cardioversion therapy and
safety of oral anticoagulation. Arch Intern Med. (1996) 156(22):2585–92. doi: 10.
3. Hindricks G, Potpara T, Dagres N, Arbelo E, Bax JJ, Blomström-Lundqvist 1001/archinte.1996.00440210109011
C, et al. 2020 ESC guidelines for the diagnosis and management of atrial
fibrillation developed in collaboration with the European Association for 6. Tabery S, Bouwels L, Uijen G, Uppelschoten A, Verheugt F. Factors influencing
Cardio-Thoracic Surgery (EACTS): the task force for the diagnosis and immediate and long-term outcome of electrical cardioversion of persistent atrial
management of atrial fibrillation of the European Society of Cardiology fibrillation and flutter. Neth Heart J. (2001) 9(1):16.
(ESC) developed with the special contribution of the European Heart 7. Kuppahally SS, Foster E, Shoor S, Steimle AE. Short-term and long-term success
Rhythm Association (EHRA) of the ESC. Eur Heart J. (2021) 42(5):373–498. of electrical cardioversion in atrial fibrillation in managed care system. Int Arch Med.
doi: 10.1093/eurheartj/ehaa612 (2009) 2:1–9. doi: 10.1186/1755-7682-2-39

Frontiers in Cardiovascular Medicine 06 frontiersin.org

Antoun et al. 10.3389/fcvm.2024.1417029

8. Castrichini M, Restivo L, Fabris E, Massa L, Di Meola R, Beltrame D, et al. 28. Budeus M, Hennersdorf M, Perings C, Wieneke H, Erbel R, Sack S. Prediction of
Prevalence and predictors of persistent sinus rhythm after elective electrical the recurrence of atrial fibrillation after successful cardioversion with P wave signal-
cardioversion for atrial fibrillation. J Cardiovasc Med. (2021) 22(8):626–30. doi: 10. averaged ECG. Ann Noninvasive Electrocardiol. (2005) 10(4):414–9. doi: 10.1111/j.
2459/JCM.0000000000001182 1542-474X.2005.00059.x
9. Fetsch T, Bauer P, Engberding R, Koch HP, Lukl J, Meinertz T, et al. Prevention 29. Başar N, Malçok Gürel O, Ozcan F, Ozlü MF, Biçer Yeşilay A, Cağlı K, et al.
of atrial fibrillation after cardioversion: results of the PAFAC trial. Eur Heart J. (2004) Diagnostic accuracy of P-wave dispersion in prediction of maintenance of sinus
25(16):1385–94. doi: 10.1016/j.ehj.2004.04.015 rhythm after external cardioversion of atrial fibrillation. Anadolu Kardiyol Derg.
(2011) 11(1):34–8. doi: 10.5152/akd.2011.006
10. Lafuente-Lafuente C, Valembois L, Bergmann JF, Belmin J. Antiarrhythmics for
maintaining sinus rhythm after cardioversion of atrial fibrillation. Cochrane Database 30. Gonna H, Gallagher MM, Guo XH, Yap YG, Hnatkova K, Camm AJ. P-wave
Syst Rev. (2019) 9:CD005049. doi: 10.1002/14651858.CD005049.pub5 abnormality predicts recurrence of atrial fibrillation after electrical cardioversion: a
prospective study. Ann Noninvasive Electrocardiol. (2014) 19(1):57–62. doi: 10.1111/
11. Luong C, Thompson DJ, Bennett M, Gin K, Jue J, Barnes ME, et al. Right atrial
anec.12087
volume is superior to left atrial volume for prediction of atrial fibrillation recurrence
after direct current cardioversion. Can J Cardiol. (2015) 31(1):29–35. doi: 10.1016/j. 31. Blanche C, Tran N, Carballo D, Rigamonti F, Burri H, Zimmermann M.
cjca.2014.10.009 Usefulness of P-wave signal averaging to predict atrial fibrillation recurrences after
electrical cardioversion. Ann Noninvasive Electrocardiol. (2014) 19(3):266–72.
12. Choi J-H, Kwon H-J, Kim HR, Park S-J, Kim JS, On YK, et al.
doi: 10.1111/anec.12131
Electrocardiographic predictors of early recurrence of atrial fibrillation. Ann
Noninvasive Electrocardiol. (2021) 26(6):e12884. doi: 10.1111/anec.12884 32. Fujimoto Y, Yodogawa K, Maru Y-j, Oka E, Hayashi H, Yamamoto T, et al.
Advanced interatrial block is an electrocardiographic marker for recurrence of atrial
13. Nielsen JB, Kühl JT, Pietersen A, Graff C, Lind B, Struijk JJ, et al. P-wave
fibrillation after electrical cardioversion. Int J Cardiol. (2018) 272:113–7. doi: 10.
duration and the risk of atrial fibrillation: results from the Copenhagen ECG study.
1016/j.ijcard.2018.07.135
Heart Rhythm. (2015) 12(9):1887–95. doi: 10.1016/j.hrthm.2015.04.026
33. Suttorp MJ, Kingma JH, Koomen EM, van’t Hof A, Tijssen JG, Lie KI.
14. Banasiak W, Telichowski A, Anker SD, Fuglewicz A, Kalka D, Molenda W, et al.
Recurrence of paroxysmal atrial fibrillation or flutter after successful cardioversion
Effects of amiodarone on the P-wave triggered signal-averaged electrocardiogram in
in patients with normal left ventricular function. Am J Cardiol. (1993) 71(8):710–3.
patients with paroxysmal atrial fibrillation and coronary artery disease. Am
doi: 10.1016/0002-9149(93)91015-A
J Cardiol. (1999) 83(1):112–4. doi: 10.1016/S0002-9149(98)00792-9
34. Alt E, Ammer R, Lehmann G, Pütter K, Ayers GM, Pasquantonio J, et al.
15. Boineau JP, Schuessler RB, Mooney CR, Wylds A, Miller C, Hudson R, et al.
Patient characteristics and underlying heart disease as predictors of recurrent
Multicentric origin of the atrial depolarization wave: the pacemaker complex.
atrial fibrillation after internal and external cardioversion in patients treated
Relation to dynamics of atrial conduction, P-wave changes and heart rate control.
with oral sotalol. Am Heart J. (1997) 134(3):419–25. doi: 10.1016/S0002-8703
Circulation. (1978) 58(6):1036–48. doi: 10.1161/01.CIR.58.6.1036
(97)70076-0
16. Forester J, Bo H, Sleigh J, Henderson J. Variability of RR, P wave-to-R wave, and
35. Soran H, Younis N, Currie P, Silas J, Jones I, Gill G. Influence of diabetes
R wave-to-T wave intervals. Am J Physiol Heart Circ Physiol. (1997) 273(6):H2857–60.
on the maintenance of sinus rhythm after a successful direct current
doi: 10.1152/ajpheart.1997.273.6.H2857
cardioversion in patients with atrial fibrillation. QJM. (2008) 101(3):181–7.
17. Magnani JW, Zhu L, Lopez F, Pencina MJ, Agarwal SK, Soliman EZ, et al. P- doi: 10.1093/qjmed/hcm123
wave indices and atrial fibrillation: cross-cohort assessments from the Framingham
36. Mercuro G, Deidda M, Piras A, Dessalvi CC, Maffei S, Rosano GM. Gender
Heart Study (FHS) and Atherosclerosis Risk in Communities (ARIC) study. Am
determinants of cardiovascular risk factors and diseases. J Cardiovasc Med. (2010)
Heart J. (2015) 169(1):53–61.e1. doi: 10.1016/j.ahj.2014.10.009
11(3):207–20. doi: 10.2459/JCM.0b013e32833178ed
18. Chiladakis J, Kalogeropoulos A, Arvanitis P, Koutsogiannis N, Zagli F,
37. Andersen KK, Andersen ZJ, Olsen TS. Age-and gender-specific prevalence
Alexopoulos D. Preferred QT correction formula for the assessment of drug-
of cardiovascular risk factors in 40,102 patients with first-ever ischemic stroke:
induced QT interval prolongation. J Cardiovasc Electrophysiol. (2010) 21(8):905–13.
a nationwide Danish study. Stroke. (2010) 41(12):2768–74. doi: 10.1161/
doi: 10.1111/j.1540-8167.2010.01738.x
STROKEAHA.110.595785
19. Opolski G, Ścisło P, Stanisławska J, Górecki A, Steckiewicz R, Torbicki A.
38. Papageorgiou P, Monahan K, Boyle NG, Seifert MJ, Beswick P, Zebede J,
Detection of patients at risk for recurrence of atrial fibrillation after successful
et al. Site-dependent intra-atrial conduction delay: relationship to initiation
electrical cardioversion by signal-averaged P-wave ECG. Int J Cardiol. (1997) 60
of atrial fibrillation. Circulation. (1996) 94(3):384–9. doi: 10.1161/01.CIR.94.3.
(2):181–5. doi: 10.1016/S0167-5273(97)02982-3
384
20. Stafford PJ, Kamalvand K, Tan K, Vincent R, Sulke N. Prediction of
39. Vlachos K, Efremidis M, Letsas KP, Bazoukis G, Martin R, Kalafateli M, et al.
maintenance of sinus rhythm after cardioversion of atrial fibrillation by analysis of
Low-voltage areas detected by high-density electroanatomical mapping predict
serial signal-averaged P waves. Pacing Clin Electrophysiol. (1998) 21(7):1387–95.
recurrence after ablation for paroxysmal atrial fibrillation. J Cardiovasc
doi: 10.1111/j.1540-8159.1998.tb00209.x
Electrophysiol. (2017) 28(12):1393–402. doi: 10.1111/jce.13321
21. Aytemir K, Aksoyek S, Yildirir A, Ozer N, Oto A. Prediction of atrial fibrillation
40. Huang D, Li J-b, Zghaib T, Ipek EG, Balouch M, Spragg DD, et al. The extent of
recurrence after cardioversion by P wave signal-averaged electrocardiography. Int
left atrial low-voltage areas included in pulmonary vein isolation is associated with
J Cardiol. (1999) 70(1):15–21. doi: 10.1016/S0167-5273(99)00038-8
freedom from recurrent atrial arrhythmia. Can J Cardiol. (2018) 34(1):73–9. doi: 10.
22. Raitt MH, Ingram KD, Thurman SM. Signal-averaged P wave duration predicts 1016/j.cjca.2017.10.012
early recurrence of atrial fibrillation after cardioversion. Pacing Clin Electrophysiol.
41. Gorenek B, Birdane A, Kudaiberdieva G, Goktekin O, Cavusoglu Y, Unalir A,
(2000) 23(2):259–65. doi: 10.1111/j.1540-8159.2000.tb00808.x
et al. P wave amplitude and duration may predict immediate recurrence of atrial
23. Guo XH, Gallagher MM, Poloniecki J, Yi G, Camm AJ. Prognostic significance fibrillation after internal cardioversion. Ann Noninvasive Electrocardiol. (2003) 8
of serial P wave signal-averaged electrocardiograms following external electrical (3):215–8. doi: 10.1046/j.1542-474X.2003.08308.x
cardioversion for persistent atrial fibrillation: a prospective study. Pacing Clin
42. Li Z, Hertervig E, Carlson J, Johansson C, Olsson SB, Yuan S. Dispersion of
Electrophysiol. (2003) 26(1p2):299–304. doi: 10.1046/j.1460-9592.2003.00037.x
refractoriness in patients with paroxysmal atrial fibrillation: evaluation with
24. Ehrlich J, Schadow K, Steul K, Zhang G, Israel C, Hohnloser S. Prediction of simultaneous endocardial recordings from both atria. J Electrocardiol. (2002) 35
early recurrence of atrial fibrillation after external cardioversion by means of P (3):227–34. doi: 10.1054/jelc.2002.33973
wave signal-averaged electrocardiogram. Z Kardiol. (2003) 92(7):540–6. doi: 10.
43. Boriani G, Diemberger I, Biffi M, Camanini C, Valzania C, Corazza I, et al.
1007/s00392-003-0940-5
P wave dispersion and short-term vs. late atrial fibrillation recurrences
25. Dixen U, Joens C, Parner J, Rasmussen V, Pehrson S, Jensen G. Prolonged after cardioversion. Int J Cardiol. (2005) 101(3):355–61. doi: 10.1016/j.ijcard.
signal-averaged P wave duration after elective cardioversion increases the risk of 2004.03.039
recurrent atrial fibrillation. Scand Cardiovasc J. (2004) 38(3):147–51. doi: 10.1080/
44. Telichowski A, Banasiak W, Wiech K, Zeborowski J, Pierog M, Ponikowski
14017430410028645
P, et al. The effect of sotalol hydrochloride therapy on atrial signal-averaged ECG
26. Dogan A, Avsar A, Ozturk M. P-wave dispersion for predicting maintenance of in patients with paroxysmal atrial fibrillation. Pol Merkur Lekarski. (1996)
sinus rhythm after cardioversion of atrial fibrillation. Am J Cardiol. (2004) 93 1(5):303–9.
(3):368–71. doi: 10.1016/j.amjcard.2003.09.064
45. Redfearn DP, Lane J, Ward K, Stafford PJ. High-resolution analysis of the
27. Perzanowski C, Ho AT, Jacobson AK. Increased P-wave dispersion predicts surface P wave as a measure of atrial electrophysiological substrate. Ann
recurrent atrial fibrillation after cardioversion. J Electrocardiol. (2005) 38(1):43–6. Noninvasive Electrocardiol. (2006) 11(1):12–9. doi: 10.1111/j.1542-474X.2006.
doi: 10.1016/j.jelectrocard.2004.09.008 00058.x

Frontiers in Cardiovascular Medicine 07 frontiersin.org