Correlation of P-wave properties with the size of left atrial low voltage areas in patients with atrial fibrillation.
12-lead surface ECG
Atrial fibrillation
Catheter ablation
Left atrial scarring
Journal
Journal of electrocardiology
ISSN: 1532-8430
Titre abrégé: J Electrocardiol
Pays: United States
ID NLM: 0153605
Informations de publication
Date de publication:
Historique:
received:
29
03
2019
revised:
30
05
2019
accepted:
12
06
2019
pubmed:
1
7
2019
medline:
22
6
2021
entrez:
1
7
2019
Statut:
ppublish
Résumé
Left atrial low voltage areas (LVA) are associated with increased recurrence rates of atrial fibrillation (AF) after catheter ablation and can be a potential ablation target during the procedure. Therefore, noninvasive prediction of the presence and the distribution of LVA may help physicians to predict ablation outcomes and to guide antiarrhythmic management. Seventy-three consecutive patients with atrial fibrillation undergoing first time left atrial ablation for paroxysmal or persistent AF were enrolled. P-wave properties (amplitude and duration) were measured in all limb and precordial leads in pre-interventional sinus rhythm surface ECGs and correlated with total LVA size. LVA were detected via high density low voltage maps of the left atrium in sinus rhythm. LVA were then manually encircled, their total size was calculated and given as a percentage of the total LA surface area. A significant, inverse correlation with LVA size was shown for P-wave amplitude for leads I, II, aVR, aVF, V1, V4, V5 and V6. Additionally, a significant positive correlation between LVA size and P-wave duration was shown for leads V1, V2 and V3. As the strongest correlation was shown for the amplitude in lead I (R = -0.578), this lead was used to find a potential cutoff for LVA prediction. The best cut-off for a P-wave amplitude in lead I to predict severe scarring (defined as LVA size >35%, according to UTAH stadium IV) was 0.062 mV with an area-under-the receiver-operating-characteristic curve of 0.935, a sensitivity of 85% and a specificity of 88%. P-wave duration and amplitude show significant correlations with LVA size and may be used as a noninvasive tool to predict severe scarring. Amplitudes in lead I smaller than 0.062 mV were found to be predictive of LVA >35%.
Sections du résumé
BACKGROUND
Left atrial low voltage areas (LVA) are associated with increased recurrence rates of atrial fibrillation (AF) after catheter ablation and can be a potential ablation target during the procedure. Therefore, noninvasive prediction of the presence and the distribution of LVA may help physicians to predict ablation outcomes and to guide antiarrhythmic management.
MATERIAL AND METHODS
Seventy-three consecutive patients with atrial fibrillation undergoing first time left atrial ablation for paroxysmal or persistent AF were enrolled. P-wave properties (amplitude and duration) were measured in all limb and precordial leads in pre-interventional sinus rhythm surface ECGs and correlated with total LVA size. LVA were detected via high density low voltage maps of the left atrium in sinus rhythm. LVA were then manually encircled, their total size was calculated and given as a percentage of the total LA surface area.
RESULTS
A significant, inverse correlation with LVA size was shown for P-wave amplitude for leads I, II, aVR, aVF, V1, V4, V5 and V6. Additionally, a significant positive correlation between LVA size and P-wave duration was shown for leads V1, V2 and V3. As the strongest correlation was shown for the amplitude in lead I (R = -0.578), this lead was used to find a potential cutoff for LVA prediction. The best cut-off for a P-wave amplitude in lead I to predict severe scarring (defined as LVA size >35%, according to UTAH stadium IV) was 0.062 mV with an area-under-the receiver-operating-characteristic curve of 0.935, a sensitivity of 85% and a specificity of 88%.
CONCLUSIONS
P-wave duration and amplitude show significant correlations with LVA size and may be used as a noninvasive tool to predict severe scarring. Amplitudes in lead I smaller than 0.062 mV were found to be predictive of LVA >35%.
Identifiants
pubmed: 31255952
pii: S0022-0736(19)30248-1
doi: 10.1016/j.jelectrocard.2019.06.008
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
38-42Informations de copyright
Copyright © 2019 Elsevier Inc. All rights reserved.