1
|
Tao Y, Zhang D, Tan C, Wang Y, Shi L, Chi H, Geng S, Ma Z, Hong S, Liu XP. An artificial intelligence-enabled electrocardiogram algorithm for the prediction of left atrial low-voltage areas in persistent atrial fibrillation. J Cardiovasc Electrophysiol 2024; 35:1849-1858. [PMID: 39054663 DOI: 10.1111/jce.16373] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/12/2024] [Revised: 06/19/2024] [Accepted: 07/07/2024] [Indexed: 07/27/2024]
Abstract
OBJECTIVES We aimed to construct an artificial intelligence-enabled electrocardiogram (ECG) algorithm that can accurately predict the presence of left atrial low-voltage areas (LVAs) in patients with persistent atrial fibrillation. METHODS The study included 587 patients with persistent atrial fibrillation who underwent catheter ablation procedures between March 2012 and December 2023 and 942 scanned images of 12-lead ECGs obtained before the ablation procedures were performed. Artificial intelligence-based algorithms were used to construct models for predicting the presence of LVAs. The DR-FLASH and APPLE clinical scores for LVA prediction were calculated. We used a receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis to evaluate model performance. RESULTS The data obtained from the participants were split into training (n = 469), validation (n = 58), and test sets (n = 60). LVAs were detected in 53.7% of all participants. Using ECG alone, the deep learning algorithm achieved an area under the ROC curve (AUROC) of 0.752, outperforming both the DR-FLASH score (AUROC = 0.610) and the APPLE score (AUROC = 0.510). The random forest classification model, which integrated a probabilistic deep learning model and clinical features, showed a maximum AUROC of 0.759. Moreover, the ECG-based deep learning algorithm for predicting extensive LVAs achieved an AUROC of 0.775, with a sensitivity of 0.816 and a specificity of 0.896. The random forest classification model for predicting extensive LVAs achieved an AUROC of 0.897, with a sensitivity of 0.862, and a specificity of 0.935. CONCLUSION The deep learning model based exclusively on ECG data and the machine learning model that combined a probabilistic deep learning model and clinical features both predicted the presence of LVAs with a higher degree of accuracy than the DR-FLASH and the APPLE risk scores.
Collapse
Affiliation(s)
- Yirao Tao
- Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
- Heart Center and Beijing Key Laboratory of Hypertension, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
| | - Deyun Zhang
- HeartVoice Medical Technology, Hefei, China
- HeartRhythm-HeartVoice Joint Laboratory, Beijing, China
| | - Chen Tan
- Department of Cardiology, Hebei Yanda Hospital, Hebei, Hebei Province, China
| | - Yanjiang Wang
- Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
- Heart Center and Beijing Key Laboratory of Hypertension, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
| | - Liang Shi
- Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
- Heart Center and Beijing Key Laboratory of Hypertension, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
| | - Hongjie Chi
- Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
- Heart Center and Beijing Key Laboratory of Hypertension, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
| | - Shijia Geng
- HeartVoice Medical Technology, Hefei, China
- HeartRhythm-HeartVoice Joint Laboratory, Beijing, China
| | - Zhimin Ma
- Department of Cardiology, Heart Rhythm Cardiovascular Hospital, Shandong, China
| | - Shenda Hong
- National Institute of Health Data Science, Peking University, Beijing, China
- Health Science Center of Peking University, Institute of Medical Technology, Beijing, China
| | - Xing Peng Liu
- Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
- Heart Center and Beijing Key Laboratory of Hypertension, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China
| |
Collapse
|
2
|
Menger V, Frick M, Sharif‐Yakan A, Emrani M, Zink MD, Napp A, Marx N, Gramlich M. Procedural performance between two cryoballoon systems for ablation of atrial fibrillation depends on pulmonary vein anatomy. J Arrhythm 2023; 39:341-351. [PMID: 37324756 PMCID: PMC10264750 DOI: 10.1002/joa3.12842] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/27/2022] [Revised: 02/13/2023] [Accepted: 03/05/2023] [Indexed: 03/19/2023] Open
Abstract
Background Cryoballoon ablation is a first-line therapy for atrial fibrillation. We compared the efficacy and safety of two ablation systems and addressed the influence of pulmonary vein (PV) anatomy on performance and outcome. Methods We consecutively enrolled 122 patients who were planned for first-time cryoballoon ablation. Patients were assigned 1:1 for ablation with the POLARx or the Arctic Front Advance Pro (AFAP) system and followed-up for 12 months. Procedural parameters were recorded during the ablation. Before the procedure, a magnetic resonance angiography (MRA) of the PVs was generated and diameter, area, and shape of each PV ostium were assessed. We applied an evaluated PV anatomical scoring system on our MRA measurement data ranging from 0 (best anatomical combination) to 5. Results Procedures performed with POLARx were associated with shorter time to balloon temperature -30°C (p < .001), lower balloon nadir temperature (p < .001), and longer thawing time till 0°C (p < .001) in all PVs, however, time to isolation was similar. We observed a decreasing performance with each increase in the score for the AFAP, whereas the POLARx performed constant regardless of the score. At 1 year, AF recurred in 14 of 44 patients treated with AFAP (31.8%) and in 10 of 45 patients treated with POLARx (22.2%) (hazard ratio, 0.61; 95% CI 0.28 to 1.37; p = .225). There was no significant correlation between PV anatomy and clinical outcome. Conclusion We found significant differences in cooling kinetics, especially when anatomical conditions are difficult. However, both systems have a comparable outcome and safety profile.
Collapse
Affiliation(s)
- Vincent Menger
- Department of CardiologyUniversity Hospital RWTH AachenAachenGermany
| | - Michael Frick
- Department of CardiologyUniversity Hospital RWTH AachenAachenGermany
| | | | - Mahdi Emrani
- Department of CardiologyUniversity Hospital RWTH AachenAachenGermany
| | | | - Andreas Napp
- Department of CardiologyUniversity Hospital RWTH AachenAachenGermany
| | - Nikolaus Marx
- Department of CardiologyUniversity Hospital RWTH AachenAachenGermany
| | - Michael Gramlich
- Department of CardiologyUniversity Hospital RWTH AachenAachenGermany
| |
Collapse
|
3
|
Seewöster T, Dinov B, Nedios S, Hindricks G, Sommer P, Kornej J. Biatrial volume ratio predicts low voltage areas in atrial fibrillation. Clin Cardiol 2021; 44:1560-1566. [PMID: 34494677 PMCID: PMC8571553 DOI: 10.1002/clc.23720] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/26/2021] [Revised: 08/08/2021] [Accepted: 08/23/2021] [Indexed: 12/16/2022] Open
Abstract
BACKGROUND Left atrial volume (LAV) and low voltage areas (LVAs) are acknowledged markers for worse rhythm outcome after ablation of atrial fibrillation (AF). Some studies reported the importance of increased right atrial volume (RAV) as a predictor for arrhythmia recurrences in AF patients. OBJECTIVE To investigate association between the LAV/RAV ratio and LVAs presence. METHODS Patients undergoing first AF ablation were included. LVAs were assessed peri-procedurally using high-density 3D maps and defined as <0.5 mV. All patients underwent pre-procedural cardiovascular magnetic resonance imaging. LAV (biplane) and RAV (monoplane 4-chamber) were assessed prior to ablation, and the LAV/RAV ratio was calculated. RESULTS The study population included 189 patients (age mean 63 ± 10 years, 33% women, 57% persistent AF, 22% LVAs). There were 149 (79%) patients with LAV > RAV. In univariable analysis LAV > RAV was associated with LVAs (OR 6.803, 95%CI 1.395-26.514, p = .016). The association remained robust in multivariable model after adjustment for persistent AF, CHA2 DS2 -VASc score, and heart rate (OR 5.981, 95%CI 1.256-28.484, p = .025). Using receiver operator curve analysis, LAV > RAV (AUC 0.668, 95%CI 0.585-0.751, p = .001) was significant predictor for LVAs. In multivariable analysis, after adjustment for age, persistent AF, and renal function, RAV≥LAV was threefold higher in males (OR 3.040, 95%CI 1.050-8.802, p = .04). CONCLUSIONS LAV > RAV is useful for the prediction of electro-anatomical substrate in AF. LAV > RAV was associated with LVAs presence, while male sex remained associated with RAV≥LAV and less LVAs.
Collapse
Affiliation(s)
- Timm Seewöster
- Department of Electrophysiology, Heart Center Leipzig - University Hospital of Cardiology, Leipzig, Germany
| | - Borislav Dinov
- Department of Electrophysiology, Heart Center Leipzig - University Hospital of Cardiology, Leipzig, Germany
| | - Sotirios Nedios
- Department of Electrophysiology, Heart Center Leipzig - University Hospital of Cardiology, Leipzig, Germany
| | - Gerhard Hindricks
- Department of Electrophysiology, Heart Center Leipzig - University Hospital of Cardiology, Leipzig, Germany
| | - Philipp Sommer
- Department of Electrophysiology, Herz- und Diabeteszentrum NRW, Bad Oeynhausen, Germany
| | - Jelena Kornej
- School of Medicine - Cardiovascular Medicine, Boston University, Boston, Massachusetts, USA
| |
Collapse
|
4
|
Effectiveness of Radiofrequency Catheter Ablation Using Ablation Index Versus Second Generation Cryoballoon in the Treatment of Persistent Atrial Fibrillation: A Matching-Adjusted Indirect Comparison. Adv Ther 2021; 38:4388-4402. [PMID: 34250584 PMCID: PMC8342373 DOI: 10.1007/s12325-021-01846-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/19/2021] [Accepted: 06/24/2021] [Indexed: 11/29/2022]
Abstract
Introduction Both radiofrequency (RF) and cryoballoon (CB) ablation are treatment options for persistent atrial fibrillation (PsAF). An important recent innovation in RF ablation is Ablation Index (AI), known also as the VISITAG SURPOINT™ Module, a composite lesion quality marker whose use has been shown to significantly reduce the incidence of acute and late pulmonary vein (PV) reconnection and the recurrence of atrial arrhythmias in PsAF. Due to a lack of direct comparative evidence between the latest generations of technologies, there is uncertainty regarding the best treatment option in PsAF. The objective of the present study was to conduct a matching-adjusted indirect treatment comparison (MAIC) using individual patient-level data (IPD) to assess the comparative effectiveness of the THERMOCOOL SMARTTOUCH™ Catheter or the THERMOCOOL SMARTTOUCH™ SF Catheter with AI/VISITAG SURPOINT™ Module (STAI) versus the second-generation CB catheter (Arctic Front Advance™; herein referred to as CB) with respect to 12-month atrial arrhythmia recurrence, fluoroscopy time, and procedural efficiency. Methods IPD for STAI were obtained from four investigator-initiated studies and were pooled. Comparable CB studies identified from a systematic literature review were also pooled. In the absence of a common treatment arm between STAI and CB studies, an unanchored MAIC was conducted. The primary analysis compared the pooled STAI IPD to the pooled CB cohort, with corrections for differences across trials, including eligibility criteria and patient baseline characteristics. Scenario and sensitivity analyses were conducted to assess the robustness of the primary analysis. Results In the primary analysis, which was adjusted for left atrial diameter (LAD), age, diabetes, and sex, STAI was associated with a statistically significant 65% relative reduction in the rate of arrhythmia recurrence compared to CB at 12-month follow-up (HR 0.35; 95% CI 0.23, 0.52). STAI was associated with shorter total fluoroscopy time than CB but longer procedure time. Results were consistent across scenario and sensitivity analyses. Conclusion Radiofrequency ablation with AI significantly reduced atrial arrhythmia recurrence at 12-month follow-up and fluoroscopy time compared to CB, with longer procedure times. Supplementary Information The online version contains supplementary material available at 10.1007/s12325-021-01846-z.
Collapse
|
5
|
Büttner P, Seewöster T, Obradovic D, Hindricks G, Thiele H, Kornej J. NT-proANP levels in peripheral and cardiac circulation. J Interv Card Electrophysiol 2021; 63:409-415. [PMID: 34173166 DOI: 10.1007/s10840-021-01020-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/31/2021] [Accepted: 06/07/2021] [Indexed: 10/21/2022]
Abstract
BACKGROUND Recent studies have reported an association between N-terminal atrial natriuretic peptide (NT-proANP) and the progression of atrial fibrillation (AF). However, NT-proANP levels in peripheral and cardiac circulation in AF patients and in non-AF individuals need to be defined. The aims of the current study are (1) to analyze NT-proANP levels in peripheral and cardiac circulation in AF patients and (2) to compare NT-proANP levels in individuals with and without AF. METHODS We recruited AF patients who were undergoing their first AF catheter ablation and non-AF individuals. Blood plasma samples taken from the femoral vein and the left atrium (LA) were collected before AF ablation in the AF patients and from the cubital vein in the non-AF controls. Low voltage areas (LVAs) were determined using high-density maps during catheter ablation and defined as < 0.5 mV. RESULTS The study included 189 AF patients (64 ± 10 years, 59% male, 61% persistent AF, 30% LVAs) and 26 non-AF individuals (58 ± 10 years, 50% male). Patients with AF were significantly older and had larger LA (p < 0.05). Compared to non-AF controls, peripheral and cardiac NT-proANP levels were significantly higher in AF patients without and with LVAs (median 5.4, 10.5, 14.8 ng/ml, respectively, p < 0.001). In multivariable analysis, NT-proANP (OR 1.238, 95% CI 1.007-1.521, p = 0.043) remained significantly different between non-AF individuals and AF patients. In AF, NT-proANP levels were significantly higher in the cardiac blood samples than in the peripheral blood (median 13.0 versus 11.4 ng/ml, p = 0.003). The ability to predict LVAs was modest when using cardiac NT-proANP (AUC 0.661) and peripheral NT-proANP (AUC 0.635), without statistical difference (p = 0.937). CONCLUSIONS NT-proANP levels are higher in individuals with AF than in controls and are more pronounced in progressed AF. Elevated cardiac and peripheral NT-proANP levels similarly predict LVAs.
Collapse
Affiliation(s)
- Petra Büttner
- Department of Cardiology, Heart Center Leipzig At University of Leipzig, Leipzig, Germany
| | - Timm Seewöster
- Department of Electrophysiology, Heart Center Leipzig At University of Leipzig, Leipzig, Germany
| | - Danilo Obradovic
- Department of Cardiology, Heart Center Leipzig At University of Leipzig, Leipzig, Germany
| | - Gerhard Hindricks
- Department of Electrophysiology, Heart Center Leipzig At University of Leipzig, Leipzig, Germany
| | - Holger Thiele
- Department of Cardiology, Heart Center Leipzig At University of Leipzig, Leipzig, Germany
| | - Jelena Kornej
- School of Medicine - Cardiovascular Medicine, Boston University, 72 E Concord St, Boston, MA, 02118, USA.
| |
Collapse
|
6
|
Volgman AS, Benjamin EJ, Curtis AB, Fang MC, Lindley KJ, Naccarelli GV, Pepine CJ, Quesada O, Vaseghi M, Waldo AL, Wenger NK, Russo AM. Women and atrial fibrillation. J Cardiovasc Electrophysiol 2020; 32:2793-2807. [PMID: 33332669 DOI: 10.1111/jce.14838] [Citation(s) in RCA: 56] [Impact Index Per Article: 11.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/30/2020] [Accepted: 12/07/2020] [Indexed: 02/06/2023]
Abstract
Atrial fibrillation (AF) remains a growing problem in the United States and worldwide, imposing a high individual and health system burden, including increased resource consumption due to repeated hospitalizations, stroke, dementia, heart failure, and death. This comprehensive review summarizes the most recent data on sex-related differences in risks associated with AF. Women with AF have increased risk of stroke and death compared to men, and possible reasons for this disparity are explored. Women also continue to have worse symptoms and quality of life, and poorer outcomes with stroke prevention, as well as with rate and rhythm control management strategies. Many current rhythm control treatment strategies for AF, including cardioversion and ablation, are used less frequently in women as compared to men, whereas women are more likely to be treated with rate control strategies or antiarrhythmic drugs. Sex differences should be considered in treating women with AF to improve outcomes and women and men should be offered the same interventions for AF. We need to improve the evidence base to understand if variation in utilization of rate and rhythm control management between men and women represents health inequities or appropriate clinical judgement.
Collapse
Affiliation(s)
| | - Emelia J Benjamin
- Boston University School of Medicine and School of Public Health, Boston, Massachusetts, USA
| | - Anne B Curtis
- Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, Buffalo, New York, USA
| | - Margaret C Fang
- Division of Hospital Medicine, University of California, San Francisco, California, USA
| | | | | | - Carl J Pepine
- Division of Cardiovascular Medicine, College of Medicine, University of Florida, Gainesville, Florida, USA
| | - Odayme Quesada
- The Christ Hospital Women's Heart Center, Cincinnati, Ohio, USA
| | - Marmar Vaseghi
- UCLA Cardiac Arrhythmia Center, University of California, Los Angeles, California, USA
| | - Albert L Waldo
- Division of Cardiovascular Medicine, Case Western Reserve University Harrington Heart & Vascular Institute, University Hospitals Cleveland Medical Center, Cleveland, Ohio, USA
| | - Nanette K Wenger
- Department of Medicine, Section of Cardiology, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Andrea M Russo
- Cooper Medical School of Rowan University, Camden, New Jersey, USA
| | | |
Collapse
|
7
|
Maj R, Borio G, Osório TG, Iacopino S, Ströker E, Sieira J, Terasawa M, Rizzo A, Scala O, Galli A, Varnavas V, Paparella G, Capulzini L, Brugada P, De Asmundis C, Chierchia GB. Conversion of atrial fibrillation to sinus rhythm during cryoballoon ablation: A favorable and not unusual phenomenon during second-generation cryoballoon pulmonary vein isolation. J Arrhythm 2020; 36:319-327. [PMID: 32256881 PMCID: PMC7132212 DOI: 10.1002/joa3.12301] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/14/2019] [Revised: 11/05/2019] [Accepted: 12/26/2019] [Indexed: 11/07/2022] Open
Abstract
BACKGROUND The prevalence and the clinical impact of conversion of atrial fibrillation (AF) to sinus rhythm (SR) during cryoballoon ablation (CB-A) are unknown. OBJECTIVE The purpose of this study was to evaluate the prevalence of restoration of SR during CB-A and the clinical impact of this phenomenon. METHODS Between January 2012 and September 2018, all patients who experienced conversion of AF to SR during CB-A were included. This group was subsequently matched for gender, age, type of AF, diagnosis-to-ablation time, and left atrial size with patients who underwent CB-A and did not experienced conversion of AF to SR. After discharge, patients were scheduled for follow-up visits at 1, 3, 6, and 12 months and 24 hours Holter recordings were obtained at each follow-up visit. All documented AF episodes of >30 seconds were considered as recurrence. A 3 month post-procedural blanking period (BP) was applied. RESULTS A total of 1559 patients underwent pulmonary veins isolation by CB-A between January 2012 and September 2018; among them, 58 patients (3.7%) experienced restoration of SR during CB-A. In total, 53 patients (41 males [77.3%], mean age 61.4 ± 13.3 years) were included in the case group. During CB-A, restoration of SR occurred more frequently during right-side PVs applications (right inferior pulmonary vein 39.6%, right superior pulmonary vein 30.2%). If considering a BP, at 2 year follow-up, freedom from recurrences was 86.5% in the case group and 68.0% in the control group (P = .036). CONCLUSION Conversion of AF to SR is a favorable and relatively frequent phenomenon during cryoballoon pulmonary vein isolation ablation.
Collapse
Affiliation(s)
- Riccardo Maj
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | - Gianluca Borio
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | | | | | - Erwin Ströker
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | - Juan Sieira
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | - Muryo Terasawa
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | | | - Oriana Scala
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | - Alessio Galli
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | | | | | | | - Pedro Brugada
- Heart Rhythm Management CenterUZ Brussel‐VUBBrusselsBelgium
| | | | | |
Collapse
|