Catheter ablation of atrial fibrillation : technological advancements and long-term effects on morbidity and mortality

Background Atrial fibrillation (AF) is associated with reduced quality of life and an increased risk of stroke, heart failure (HF), and death. In Europe, one in three individuals above the age of 45 will develop AF at some point in their lives; therefore, AF represents a highly relevant public health concern. Consequently, a substantial body of research has recently focused on improving strategies for AF treatment, including the development of novel technologies for mapping and ablation of the arrhythmia. In parallel, several studies have investigated the impact of AF ablation on hard clinical endpoints such as HF, stroke, and death. This thesis aims to evaluate two such novel techniques (intracardiac echocardiography [ICE] integration for automatic anatomical reconstruction of the left atrium [LA] and ethanol infusion of the vein of Marshall) and to study the long-term effects of AF ablation on the risk of stroke, mortality, and dementia in a large national register- based real-world cohort. Methods and results Study 1 evaluated a novel ICE software machine learning algorithm in 16 patients who underwent mapping of the LA as part of an ablation procedure for AF or other indications. The algorithm was validated in two steps: 1) identification of the main LA anatomical structures; and 2) analysis of the location of automated lesion tags in relation to the PV antrum of the ICE-generated automated LA map. In all patients, all anatomical structures were correctly identified by the algorithm and the pulmonary vein (PV) antrum accuracy relative to the lesion tags was acceptable, with 69% of them classified as satisfactory. Studies 2 and 3 evaluated the association between AF ablation and the risk of stroke or death (Study 2), or a dementia diagnosis (Study 3). For both studies, consecutive patients who underwent first-time AF ablation between 2008 and 2018 at three major EP centers in Sweden were included. A control cohort of medically managed AF patients was identified from the Swedish National Patient Register. Propensity score matching based on several clinical and socioeconomic variables resulted in two study cohorts with similar characteristics. Mean follow-up was 4.5+2.8 and 4.9+2.8 years for Studies 2 and 3, respectively. In both studies, AF ablation was associated with a reduction in the primary endpoints: all-cause mortality or stroke (hazard ratio [HR] 0.58, 95% confidence interval [CI] 0.48 - 0.69) and dementia diagnosis (HR 0.44, 95% CI 0.22-0.86). Study 4 prospectively included consecutive patients with recurrent persistent AF who underwent a repeat ablation procedure showing durable PV isolation (PVI) and who then received a Marshall-Plan lesion set. As historical controls, patients with a comparable clinical profile who had undergone repeat ablation for persistent AF at our center prior to the introduction of the Marshall-Plan were retrospectively included (n=38). The primary outcome was freedom from AF or atrial tachycardia (AT) at 12 months. All patients in the Marshall-Plan group (n=38) underwent a complete lesion set with block across all anatomical lines. Freedom from AF or AT at 12 months was achieved in 26 (68.4%) patients in the Marshall-Plan group vs. 17 (44.7%) in the control group (HR 0.45, 95% CI 0.22- 0.93). Conclusions A novel ICE machine learning algorithm for automatic anatomical reconstruction of the LA was deemed feasible and correctly identified the anatomical structures of the LA in all patients studied, with an acceptable alignment of the ablation lesion tags to most parts of the PV antra. Although future validation is necessary, the results suggest that ICE-integration algorithms may improve future procedural efficiency. In a population-based registry study, patients undergoing first-time AF ablation at three centers were compared with medically treated patients identified from a national register. AF ablation was associated with a reduced risk of the composite endpoint of mortality or stroke as well as a dementia diagnosis. Finally, in patients with persistent AF who presented with durable PVI at a repeat ablation procedure, the Marshall-Plan ablation strategy was associated with a higher freedom from AF/AT and an acceptable complication rate compared with conventional ablation alone. List of scientific papers I. FINN ÅKERSTRÖM, Nikola Drca, Mats Jensen-Urstad, Frieder Braunschweig. Feasibility of a novel algorithm for automated reconstruction of the left atrial anatomy based on intracardiac echocardiography. Pacing and Clinical Electrophysiology, 2022(45);1288-1294. https://doi.org/10.1111/pace.14599 II. FINN ÅKERSTRÖM, Julie Hutter, Emmanouil Charitakis, Fariborz Tabrizi, Fahd Asaad, Hamid Bastani, Tara Bourke, Frieder Braunschweig, Nikola Drca, Anders Englund, Leif Friberg, Per Insulander, Anders Hassel Jönsson, Göran Kennebäck, Astrid Paul- Nordin, Bita Sadigh, Ott Saluveer, Serkan Saygi, Jonas Schwieler, Emma Svennberg, Jari Tapanainen, Yusuf Türkmen, Mats Jensen-Urstad. Association between catheter ablation of atrial fibrillation and mortality or stroke. Heart 2024;110:163-169. https://doi.org/10.1136/heartjnl-2023-322883 III. FINN ÅKERSTRÖM, Emmanouil Charitakis, Astrid Paul-Nordin, Frieder Braunschweig, Leif Friberg, Fariborz Tabrizi, Mats Jensen-Urstad, Nikola Drca. Reduced dementia risk in patients with optimized anticoagulation therapy undergoing atrial fibrillation ablation. Heart Rhythm 2024;21:1487-1492. https://doi.org/10.1016/j.hrthm.2024.04.038 IV. FINN ÅKERSTRÖM, Emmanouil Charitakis, Astrid Paul-Nordin, Serkan Saygi, Emma Svennberg MD, PhD, Frieder Braunschweig, Mats Jensen-Urstad, MD, Nikola Drca, MD, PhD. Marshall-Plan ablation strategy for recurrent persistent atrial fibrillation in patients with durably isolated pulmonary veins. [Submitted]