Abstract Funding Acknowledgements Type of funding sources: None. Introduction Implantable cardiac monitors (ICMs) are increasingly used to monitor atrial fibrillation (AF) burden following catheter ablation. AF burden recorded by the Confirm Rx™ ICM cannot be modified even after adjudication of false-positive (FP) episodes. We sought to investigate accuracy of the AF burden detection in a UK cohort. Methods This multicentre retrospective study included patients with Confirm Rx™ and at least one episode >6 minutes across 9 UK hospitals. Each episode had a corresponding 120-second EGM (electrogram) and heart rate scatterplot which was considered representative of the whole episode. One cardiologist adjudicated all EGMs as ‘True AF’ or ‘False positive’ and a random sample of 10% was reviewed to account for intra and interobserver variability. AF burden was computed as the duration of all episodes classified as AF by the Confirm-Rx divided by the total duration of follow-up. ‘True-AF’ burden was calculated by dividing the duration of episodes adjudicated as ‘True-AF’ by the total duration of follow-up. We also investigated the accuracy of AF burden according to implantation indication and episode duration. Results A total of 16,230 individual AF episodes were included from 232 consecutive patients. Overall, 26,137 hours of AF were recorded and a total follow-up 315 patient-years which equates to an AF burden of 0.95%. However, only 24,404.7 (93.3%) hours represented time in ‘True-AF’ and a ‘True-AF’ burden for the whole cohort of 0.89% (Table 1). Patients with a Confirm-Rx™ for palpitations and suspected AF had the lowest proportion of ‘True-AF’ burden and had a modest contribution to the overall AF burden (Figure 1). Conversely, patients with known AF had the highest proportion of ‘True-AF’ burden recorded. Most AF (84.5%) episodes lasted less than 1 hour with approximately a quarter adjudicated as false-positive detections, but their contribution towards overall AF burden was very small (Figure 2A-2B). In contrast, AF >3 hours accounted for 76.4% of time in AF and the proportion of ‘True-AF’ burden was 98.5%. Conclusion The accuracy of the estimated AF burden for the whole cohort was excellent (93.3%), driven by the high proportion of ‘True-AF’ burden in AF>3 hours.

Abstract Funding Acknowledgements Type of funding sources: None. Background Radiofrequency (RF) current is a classic ablation technology used in the EP lab. By now it is assumed that RF lesion size is mainly determined by RF power, duration and contact force. This ex vivo animal model study shows that a new parameter - "electrode tissue coupling" (ETC) - plays an additional major role in RF lesion creation. The ETC level describes the amount of electric contact between the RF electrode and cardiac tissue: In minor ETC levels, only the distal electrode connects to tissue. In full ETC levels, the whole RF electrode is embedded in cardiac tissue. Methods RF-lesions were created using an ex vivo porcine cardiac model. The experimental setup consisted of a saline-filled container, a dispersive electrode, a heated thermostat and a circulation pump to imitate in vivo conditions. Global impedance was kept at 120 Ohm as well as the temperature at 37°C. RF-lesions were created using identic values of RF duration and contact force. A RF power of 20W, 30W, 40W, and 50W was used. The ETC levels were systematically varied between minor and full coupling. All parameters (power, temperature, global and local impedance, contact force, ETC, lesion size) were measured constantly during application of RF-current, enabling real-time correlation of RF parameters and lesion size. Results In total, 1923 measurements during application of RF-current were analyzed. In ETC III (full tissue coupling), lesions became significantly wider and deeper. In 20W ablations, lesion diameters were significantly (1.68 fold) larger when applying ETC III instead of ETC I. This relation was found in variations of RF parameters (Table 1). Interestingly, baseline local impedance and local impedance drop showed a high correlation with selected ETC levels. The average baseline local impedance in ETC I was 207.2 Ohm, compared to 267.3 Ohm in ETC III (p < 0.01). Discussion In addition to by now known parameters (power, duration, contact force), electrode tissue coupling is a main determinant of lesion size. Higher ETC levels result in higher amounts of RF current going into adjacent tissue instead of current dissipation into the blood pool. In clinical practice, the ETC level can be predicted by baseline local and global impedance. Observation of these parameters should become clinical practice during RF ablation.

Abstract Funding Acknowledgements Type of funding sources: None. Background Oesophageal protection using a dedicated device to provide controlled active thermal protection of the oesophagus during atrial fibrillation ablation has been shown to be effective at reducing endoscopically detected oesophageal lesions. Real world registry data of this device and established evidence on the effect of cooling on localized tissue after thermal injury have been under review. Purpose To determine the safety of an oesophageal temperature control device by a review of real-world registry data on its clinical use during catheter ablation procedures and to clarify basic mechanisms of its therapeutic action by a literature review of scientific studies on cooling in the context of thermal injury. Methods The United States Food and Drug Administration (FDA) Manufacturer and User Facility Device Experience (MAUDE), FDA Medical and Radiation Emitting Device Recalls, the Medicines and Healthcare products Regulatory Agency (MHRA) Medical Device Alerts and SwissMedic records of Field Safety Corrective Actions (FSCA) databases were reviewed for any device-related adverse events. A systematic literature review was conducted to clarify the findings from studies investigating the physiological processes behind the therapeutic effect of cooling after tissue thermal injury. Results Of over 20,000 oesophageal temperature control devices clinically used, 7976 were recorded as having been used for the purpose of oesophageal protection during left atrial catheter ablations. No adverse events occurred related to its use during left atrial catheter ablations. No case of clinically significant oesophageal injury was reported in a patient who had been protected by the oesophageal temperature control device. 208 research articles retrieved from PubMed and MEDLINE that met the search criteria were reviewed. The common finding in all the studies was that cooling had an anti-inflammatory and restorative effect via modulation of several immune-mediated pathways, local cellular function and genetic expression. Conclusions There have been no adverse events reported to date in real world clinical use of an oesophageal temperature control device during left atrial catheter ablations, for the purpose of active thermal protection. Literature review data suggests that there are complex biophysical and cellular effects from cooling that leads to its therapeutic effect but further work is required to define the mechanisms of action of thermal protection in this specific context.

Abstract Funding Acknowledgements Type of funding sources: Other. Main funding source(s): This work was supported by Aarhus University, the Danish Heart Foundation (grant number R140-A9482-B2407); Health Research Foundation of Central Denmark Region (grant number R64-A3194-B1667); and Gangstedfonden. Background Magnetic resonance imaging (MRI) is increasingly becoming the imaging modality of choice for many clinical disorders. According to guidelines, MRI can safely be performed 6 weeks after implantation of a MRI-conditional permanent pacemaker. However, there is no recommendations and only very limited evidence regarding MRI in patients treated with temporary external pacemakers. Such patients thus may be denied the opportunity for well-indicated MRI. Purpose To describe safety and feasibility of MRI in patients with temporary external pacemakers. Method We reviewed patient records of all patients treated with temporary external pacing between 2016 and 2020 at a tertiary centre. Temporary pacing was established using a transvenous standard active fixation pacing lead inserted percutaneously and connected to a MRI-conditional pacemaker taped to the skin. Information on baseline characteristics, indication, days of hospital and pacemaker function before and after MRI was gathered. Before MRI, devices were programmed according to guidelines for permanent pacemakers and the device was fixated to avoid movement due to the magnetic field of the MRI system. Patients were monitored during MRI. Values are reported as N (%), and continuous variables are presented as median (interquartile range [IQR]) or mean ± SD. Results A total of 827 patients received a temporary pacemaker. MRI was performed in 44 (5%) patients. Mean age was 71 ­± 13 years and 13 (30%) were females. Indication for temporary device were atrioventricular block (20 [45%]), aborted cardiac arrest (9 [21%]), infective endocarditis (9 [21%]), and other (6 [14%]). MRI was performed to rule out or confirm structural heart disease (16 [36%]), cerebral disease (9 [20%]), spondylodiscitis (7 [16%]), or other (12 [27%]), which was not able to do with other imaging modalities. Cardiac MRI was performed in 22 (50%) patients, while MRI of cerebrum, spine, abdomen, and lower extremities were performed in the remaining patients. Median (IQR) time from implantation of the temporary device to MRI was 6 (3–11) days. During MRI, we observed no device-related malfunction or arrhythmia. Nor did we detect any change in lead sensing, impedance, or pacing threshold. A final diagnosis was confirmed in 27 (61%) patients after MRI. However, MRI provided information to guide the clinical management in all cases without artifacts compromising the diagnostic image quality. Instead of a dual chamber pacemaker, 7 (22%) patients had implanted a cardiac resynchronization therapy and/or implantable cardioverter defibrillator due to the MRI diagnosis. A permanent pacemaker was implanted in 32 (73%) patients within median (IQR) 12 (7–23) days after temporary pacing was established. Conclusion MRI is feasible and safe in patients with temporary external pacing established with a regular MRI-conditional device and a standard active fixation lead. The MRI was helpful in determining type of permanent device.

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Funding: It was supported by NKFIH grants (K-19992, K- 135464, GINOP-2.3.2-15-2016-00047). Background The positive impact of regular exercise on a healthy and fulfilling lifestyle. Therefore, active athletes are considered the healthiest members of our society. However, there is an increasing amount of evidence that long-term high-intensity sporting activity could also have adverse effects on the heart, such as impaired electrophysiological properties. Heavy long-term training can lead to structural and functional remodelling of the heart, which in turn, can evoke malignant cardiac arrhythmias. Purpose To develop animal models with a significant translational value of the human athlete’s heart and to investigate underlying malignant drivers of cardiac arrhythmias due to the long-term endurance training in in vitro studies. Methods 24 dogs from both sexes and 26 male guinea pigs were randomly assigned to sedentary (’Sed’) and trained (’Tr’) groups (n = 12-12; n=13-13). The latter group underwent a long-term endurance interval-training program on the treadmill 5 days a week for 4 months. ECG recordings and echocardiography validated the characteristic of athlete’s heart. After heart removal, the degree of interstitial fibrosis was quantified and ventricular myocytes were enzymatically dissociated via retrograde perfusion. The transmembrane ionic currents were recorded using the whole-cell configuration of the patch-clamp technique. The action potentials were measured by the perforated patch-clamp technique. Immunocytochemistry measurements were performed to determine the density of transmembrane ion channels. Results Based on the ECG and ECHO results, the vigorous training program resulted in significant cardiac adaptation in both species. In addition, it caused mild ventricular fibrosis. The repolarization is reflected as the 90 percent of action potential duration (APD90). It was significantly lengthened in the left ventricular myocytes of ‘Tr’ dogs. (‘Tr’ vs. ‘Sed’ 472.8±29.6 ms; =29 vs. 369.3±31.4 ms; n=24, p=0.023) and there was no difference in the case of guinea pigs. The amplitude of the transient outward current (Ito), which is not expressed in the guinea pig heart, was significantly smaller in the ‘Tr’ dogs (‘Tr’ vs. ‘Sed’ 7.6±0.6 pA/pF, n=54 vs. 10.2±1.0 pA/pF, n=42, p<0.05). Under the currently used protocols, no differences were detected in the magnitude of other ionic currents. The HCN4 gene expression was significantly higher in isolated myocytes in ’Tr’ dogs. Conclusion Increased ectopic activity is not rare among top athletes. Our results suggest an association between increased arrhythmia susceptibility and impaired repolarisation reserve related to down-regulation of Ito and prolonged APD90 and enhanced fibrotic changes. The overexpression of HCN4 gene in hypertrophic hearts, similar to heart failure, may evoke malignant ventricular arrhythmias. Further studies are warranted to clarify this hypothesis in more detail.

Abstract Funding Acknowledgements Type of funding sources: None. Background The cryoballoon technology has been widely proven to be effective in achieving acute pulmonary vein (PV) isolation and favorable clinical outcome. To date, little information is still available about occurrence of late PV reconnection after cryoballoon ablation. Purpose To assess the rate of durable PV isolation following cryoballoon ablation of atrial fibrillation in the setting of repeat procedures. Methods and Results A total of 154 consecutive patients (90 male, 58.4%; mean age 60.6±12.6 years) underwent a repeat procedure, after a mean 21.5±12.4 months (median 14 months), after index ablation using the 28-mm second-generation cryoballoon. All repeat ablations were performed using a 3-dimensional electro-anatomical mapping system. Among all 590 PVs, including 26 left common ostiums (LCOs), 99 (16.8%) showed a PV reconnection in 86 patients (1.15 per patient). Persistent PV isolation could be documented in 491 of 590 PVs (83.2%). In 68 of 154 patients (44%), all PVs were electrically isolated. In the multivariable analysis, balloon nadir temperature (p<0.01) and time to PV isolation (P=0.02) independently predicted late PV reconnection. Conclusions The rate of late PV reconnection after second-generation cryoballoon ablation was low (1.15 PVs/patient). Globally, the rate of durable PV isolation assessed during repeat procedures was 83.2%. The balloon nadir temperature and faster time to isolation were independently associated with durable PV isolation.

Abstract Funding Acknowledgements Type of funding sources: None. Background Atrial fibrillation (AF) is the most common clinical arrhythmia. Pulmonary vein isolation (PVI) by catheter ablation has become a cornerstone in the treatment of AF. Serious complications to PVI have been reported to be at an acceptable level and risk of death after AF ablation is low. Purpose In a contemporary nationwide cohort of patients undergoing first-time PVI by catheter ablation, we wanted to investigate the 30-day mortality after ablation, and to examine risk and potential risk factors of PVI-related complications. Methods Population-based cohort study in patients who underwent first-time PVI by catheter ablation between 2011-2018 identified from the National Danish Ablation Registry. Primary outcome was early post-procedural mortality, defined as death of any cause within 30 days of index PVI procedure, or in connection to a hospitalization started within 30 days. Secondary outcomes were all-cause rehospitalization and complication, including postoperative infection, cardiac, vascular, neurological, vascular, and pulmonary complications within 30 days. Data on mortality and complications were collected from national health and administrative registries. Binary regression was used to estimate risk ratio (RR) with 95% confidence intervals (CI) for association between selected predictors and any complication, and adjusted gender, age, BMI, prior ablation, calendar period (ablation from 2011-2013, 2014-2016, and >2016). Results We included 8560 patients. Median age was 62, 66% were men, 12% had a history of heart failure, and median CHA2DS2VASc score was 1 (Interquartile range [IQR]; 1-2). Charlton Comorbidity index (CCI) was none in 66%, moderate in 29% and severe in 5%. A total of 10 (0.12%) patients died within 30 days of ablation, of which 4 patients died during initial hospitalization. Median time to death was 20 (IQR, 12 to 29) days. Patients who died were more likely to have experienced a procedure-related complication (40% vs. 4%, P<0.001). Procedure-related complications occurred in 298 (3.5%), and the risk was 4.4%, 3.0% and 3.3% in the time periods between 2011-2013, 2014-2016 and >2016, respectively. Most common complications were postoperative infection (26%), cardiac complication (26%), and vascular complications (18%). Complication risk was increased in patients with higher age (aRR, 65-74 year; 1.67 [1.32-2.11] and >74 years; 2.48 [1.60-3.84]), moderate CCI (aRR 1.45 [1.14-1.83]), cardiovascular disease (aRR 1.52 [1.09-2.11]) and antithrombotic treatment (aRR 1.41 [1.05-1.89]). After first-time PVI, 1.963 (23%) patients were re-hospitalized within 30 days, and most common primary discharge diagnoses were AF (87%) and direct cardioversion was performed in 765 (39%) patients. Conclusion In a nationwide cohort of patients who underwent first-time PVI, number of deaths within 30 days of ablation was low (0.12%). Risk of complication was low and 23% of the patients were re-hospitalized within 30 days.

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): Bayer Background Atrial fibrillation (AF) is a predominant risk factor of ischemic stroke and prophylactic treatment with oral anticoagulants (OACs) is recommended in older patients. Advancing age increases the risk of stroke, however, ageing simultaneously increases the prevalence of bleeding risk factors such as multimorbidity, polypharmacy, frail vessels as well as the tendency to fall. As such, treatment with OACs in older frail patients is often associated with complex risk-benefit considerations. Purpose To examine treatment patterns with OACs in older patients with AF. Methods Using nationwide registries and databases, Danish patients > 75 years of age with a primary diagnosis of AF between 2010 and 2018 were identified. The primary event of interest was claimed prescriptions for OACs in a period of 180 days after hospital contact due to AF. Proportions of patients treated with OACs were estimated and depicted graphically. Clinical factors associated with the probability of receiving OAC treatment were identified using adjusted logistic regression models with estimates presented as Odds Ratios (OR) with 95% Confidence Intervals (95% CI). Results A total of 40,027 patients were included with a slight majority of women (54%). The median age was 81 years (IQR 78-86). We found that an overall 32,235 patients (81%) were prescribed an OAC after hospital contact due to AF. Increasing age was associated with a clear tendency toward a decreased probability of receiving OAC treatment. As such, among patients aged 75 years, 85 % received OAC treatment decreasing to 68 % in patients aged 90 years. (Figure 1) Collectively during the study period, a substantial increase in the proportion of patients treated with OACs was observed. Hence, in 2011, 63 % claimed OAC prescriptions whereas this proportion increased to 92 % in 2018. (Figure 2) Factors related to a low probability of OAC treatment were bleeding risk factors such as a history of bleeding (OR 0.58, 95% CI 0.53-0.82), hemorrhagic stroke (OR 0.21, 95% CI 0.16-0.27), treatment with dual antiplatelet therapy (OR 0.52, 95% CI 0.47-0.59) as well as markers of frailty such as osteoporosis (OR 0.78, 95% CI 0.71-0.85) and previous falling (OR 0.56, 95 % CI 0.51-0.62). Conclusions In this large nationwide study, we found that in older patients with AF, the overall rates of OAC prescription were relatively high (~ 80%) but decreased notably with age. Factors associated with not receiving guideline directed OAC treatment were generally related to bleeding risk factors or frailty. These data highlight a clinical conundrum as older patients potentially benefit the most from treatment with OACs. Figure legend Figure 1 Combined plot depicting the absolute number of patients by age (red bars) as well as the proportion of patients treated with oral anticoagulants by age (blue curve). Figure 2 Plot depicting the proportion of patients receiving treatment with oral anticoagulants by calendar year.

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): National Research Agency, 3IA Côte d’Azur (ANR-19-P3IA-0002), IHU Liryc (ANR- 10-IAHU-04) Background Infarct heterogeneity plays a critical role in the development of scar-related ventricular arrhythmias (VA). Wall thickness (WT) from CT was shown to correlate with arrhythmogenic sites in the context of ablation. Purpose To analyze the relationship between WT distribution and the presence VAs in patients with history of myocardial infarction (MI). Methods From 2010 to 2020, we retrospectively included consecutive patients who underwent CT more than 1 year after MI.Automated LV wall segmentation, reorientation, WT computation and flattening methods were applied to obtain 2D WT bullseye maps.The population was divided into a training set (3/4) and a testing set (1/4). On the training set, a conditional variational autoencoder (CVAE) model was trained to encode the WT map in its latent representation,which was then used by a classifier model to predict VA. For each prediction, a gradient back-projection method was used to generate attention maps highlighting the bullseye region most influential in the model’s decision. The ability of the trained CVAE to identify patients with VA was then assessed on the test population, and compared to that of other clinical variables (age, gender, LVEF, scar size). Results 641 patients were included (age 73±7 years, 83% males, LVEF 46±10%), including 166 (26%) with history of VA. From original CT images, automated processing methods allowed for the obtention of a WT bullseye, a VA prediction and an attention map in less than 2 min. On the testing population, univariable correlates of VA were LVEF (P<0.001), scar size defined as WT area (P<0.001), CVAE prediction (P<0.001), and male gender (P=0.007). Multivariable analysis identified CVAE prediction and male gender as independent VA correlates (P<0.001 and P=0.01, R2=0.364), while LVEF and scar size were not (P=0.052 and P=0.60). The CVAE model identified patients with VA with a sensitivity/specificity of 0.87/0.74. The analysis of attention maps confirmed that the CVAE actually used the MI region to predict VA, and showed that the CVAE was filtering out areas of physiological WT, thereby outperforming simple WT thresholding. Conclusions Fully automated analysis of WT from CT images is feasible in patients with MI, and allows for the extraction of robust markers of VA, outperforming conventional risk stratifiers

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): Daiichi Sankyo Background Patients with atrial fibrillation (AF) or venous thromboembolism (VTE) receiving long-term direct oral anticoagulant (DOAC) therapy undergo diagnostic or therapeutic procedures at a rate of approximately 10% annually. The prospective Global EMIT-AF/VTE program (Edoxaban Management in Diagnostic and Therapeutic Procedures; NCT02950168, NCT02951039) demonstrated that physician-guided periprocedural management of the DOAC edoxaban in these patients was associated with low bleeding and thromboembolic event rates. It is unclear whether the experience of a previous (index) procedure influences the periprocedural management of subsequent procedures. Purpose To analyze differences in periprocedural edoxaban management in patients on chronic anticoagulation therapy undergoing multiple diagnostic or therapeutic procedures. Methods Baseline characteristics were recorded in patients enrolled in the EMIT-AF/VTE program who underwent multiple procedures. Details of periprocedural edoxaban interruption were collected from patients who underwent two procedures of the same European Heart Rhythm Association (EHRA) bleeding risk level or procedural type. Only data from the index and second procedure of the same category were included in this analysis; procedures conducted less than 7 days apart were excluded. All analyses are exploratory and descriptive in nature. Results Among 227 patients who underwent multiple procedures, the most common types were vascular and gastrointestinal (GI) procedures. Patients had a mean ± standard deviation age of 72.1 ± 9.8 years, a CHA2DS2-VASc score of 3.2 ± 1.6, a HAS-BLED score of 1.9 ± 1.0, and were mostly male (67.0%). Patients who underwent low/minor risk procedures were less likely to undergo edoxaban interruption with their second procedure compared with their index procedure (Figure 1A), and the median interruption duration was shorter for the second procedure (Table 1). A second high risk procedure was associated with a higher rate of both pre- and postprocedural edoxaban interruption compared with a patient’s index procedure, but treatment resumed earlier (Figure 1B). Patients who underwent vascular procedures had a lower rate of pre- and postprocedural interruption and a shorter interruption time with their second procedure (Table 1). Conversely, patients who underwent GI procedures experienced pre- and postprocedural interruption more often for their second procedure. The median interruption duration was longer for GI procedures than for vascular procedures (Table 1). Conclusion Overall, periprocedural edoxaban interruption varied by procedural bleeding risk and type. Edoxaban interruption patterns differed between index and second procedures, indicating that periprocedural edoxaban management may be influenced by the experience of previous procedures.