Three-dimensional-guided and ICE-guided transseptal puncture for cardiac ablations: A propensity score match study.

INTRODUCTION: Transseptal puncture (TSP) is routinely performed for left atrial ablation procedures. The use of a three-dimensional (3D) mapping system or intracardiac echocardiography (ICE) is useful in localizing the fossa ovalis and reducing fluoroscopy use. We aimed to compare the safety and efficacy between 3D mapping system-guided TSP and ICE-guided TSP techniques. METHODS: We conducted a prospective observational study of patients undergoing TSP for left atrial catheter ablation procedures (mostly atrial fibrillation ablation). Propensity scoring was used to match patients undergoing 3D-guided TSP with patients undergoing ICE-guided TSP. Logistic regression was used to compare the clinical data, procedural data, fluoroscopy time, success rate, and complications between the groups. RESULTS: Sixty-five patients underwent 3D-guided TSP, and 151 propensity score-matched patients underwent ICE-guided TSP. The TSP success rate was 100% in both the 3D-guided and ICE-guided groups. Median needle time was 4.00 min (interquartile range [IQR]: 2.57-5.08) in patients with 3D-guided TSP compared to 4.02 min (IQR: 2.83-6.95) in those with ICE-guided TSP (p = .22). Mean fluoroscopy time was 0.2 min (IQR: 0.1-0.4) in patients with 3D-guided TSP compared to 1.2 min (IQR: 0.7-2.2) in those with ICE-guided TSP (p < .001). There were no complications related to TSP in both group. CONCLUSIONS: Three-dimensional mapping-guided TSP is as safe and effective as ICE-guided TSP without additional cost.

Fourier space approach for convolutional neural network (CNN) electrocardiogram (ECG) classification: A proof-of-concept study.

There has been a proliferation of machine learning (ML) electrocardiogram (ECG) classification algorithms reaching > 85% accuracy for various cardiac pathologies. Although the accuracy within institutions might be high, models trained at one institution might not be generalizable enough for accurate detection when deployed in other institutions due to differences in type of signal acquisition, sampling frequency, time of acquisition, device noise characteristics and number of leads. In this proof-of-concept study, we leverage the publicly available PTB-XL dataset to investigate the use of time-domain (TD) and frequency-domain (FD) convolutional neural networks (CNN) to detect myocardial infarction (MI), ST/T-wave changes (STTC), atrial fibrillation (AFIB) and sinus arrhythmia (SARRH). To simulate interinstitutional deployment, the TD and FD implementations were also compared on adapted test sets using different sampling frequencies 50 Hz, 100 Hz and 250 Hz, and acquisition times of 5 s and 10s at 100 Hz sampling frequency from the training dataset. When tested on the original sampling frequency and duration, the FD approach showed comparable results to TD for MI (0.92 FD - 0.93 TD AUROC) and STTC (0.94 FD - 0.95 TD AUROC), and better performance for AFIB (0.99 FD - 0.86 TD AUROC) and SARRH (0.91 FD - 0.65 TD AUROC). Although both methods were robust to changes in sampling frequency, changes in acquisition time were detrimental to the TD MI and STTC AUROCs, at 0.72 and 0.58 respectively. Alternatively, the FD approach was able to maintain the same level of performance, and, therefore, showed better potential for interinstitutional deployment.

Multi-center atrial fibrillation electrocardiogram (ECG) classification using Fourier space convolutional neural networks (FD-CNN) and transfer learning.

There has been a proliferation of machine learning (ML) electrocardiogram (ECG) classification algorithms reaching >85% accuracy for various cardiac pathologies. Despite the high accuracy at individual institutions, challenges remain when it comes to multi-center deployment. Transfer learning (TL) is a technique in which a model trained for a specific task is repurposed for another related task, in this case ECG ML model trained at one institution is fine-tuned to be utilized to classify ECGs at another institution. Models trained at one institution, however, might not be generalizable for accurate classification when deployed broadly due to differences in type, time, and sampling rate of traditional ECG acquisition. In this study, we evaluate the performance of time domain (TD) and frequency domain (FD) convolutional neural network (CNN) classification models in an inter-institutional scenario leveraging three different publicly available datasets. The larger PTB-XL ECG dataset was used to initially train TD and FD CNN models for atrial fibrillation (AFIB) classification. The models were then tested on two different data sets, Lobachevsky University Electrocardiography Database (LUDB) and Korea University Medical Center database (KURIAS). The FD model was able to retain most of its performance (>0.81 F1-score), whereas TD was highly affected (<0.53 F1-score) by the dataset variations, even with TL applied. The FD CNN showed superior robustness to cross-institutional variability and has potential for widespread application with no compromise to ECG classification performance.

A comparison of clinical outcomes and cost of radiofrequency catheter ablation for atrial fibrillation with monitored anesthesia care versus general anesthesia.

INTRODUCTION: Monitored anesthesia care (MAC) or general anesthesia (GA) can be used during catheter ablation (CA) of atrial fibrillation (AF). However, each approach may have advantages and disadvantages with variability in operator preferences. The optimal approach has not been well established. The purpose of this study was to compare procedural efficacy, safety, clinical outcomes, and cost of CA for AF performed with MAC versus GA. METHODS: The study population consisted of 810 consecutive patients (mean age: 63 ± 10 years, paroxysmal AF: 48%) who underwent a first CA for AF. All patients completed a preprocedural evaluation by the anesthesiologists. Among the 810 patients, MAC was used in 534 (66%) and GA in 276 (34%). Ten patients (1.5%) had to convert to GA during the CA. RESULTS: Although the total anesthesia care was longer with GA particularly in patients with persistent AF, CA was shorter by 5 min with GA than MAC (p < 0.01). Prevalence of perioperative complications was similar between the two groups (4% vs. 4%, p = 0.89). There was no atrioesophageal fistula with either approach. GA was associated with a small, ~7% increase in total charges due to longer anesthesia care. During 43 ± 17 months of follow-up after a single ablation procedure, 271/534 patients (51%) in the MAC and 129/276 (47%) patients in the GA groups were in sinus rhythm without concomitant antiarrhythmic drug therapy (p = 0.28). CONCLUSION: With the participation of an anesthesiologist, and proper preoperative assessment, CA of AF using GA or MAC has similar efficacy and safety.

Association between symptoms, affect and heart rhythm in patients with persistent or paroxysmal atrial fibrillation: an ambulatory pilot study.

Symptoms in atrial fibrillation are generally assumed to correspond to heart rhythm; however, patient affect - the experience of feelings, emotion or mood - is known to frequently modulate how patients report symptoms but this has not been studied in atrial fibrillation. In this study, we investigated the relationship between affect, symptoms and heart rhythm in patients with paroxysmal or persistent atrial fibrillation. We found that presence of negative affect portended reporting of more severe symptoms to the same or greater extent than heart rhythm.

Efficacy and tolerability of quinidine as salvage therapy for monomorphic ventricular tachycardia in patients with structural heart disease.

INTRODUCTION: Quinidine is an effective therapy for a subset of polymorphic ventricular tachycardia and ventricular fibrillation (VF) syndromes; however, the efficacy of quinidine in scar-related monomorphic ventricular tachycardia (MMVT) is unclear. METHODS AND RESULTS: Between 2009 and 2020 a single VT referral center, a total of 23 patients with MMVT and structural heart disease (age 66.7 ± 10.9, 20 males, 15 with ischemic cardiomyopathy, mean LVEF 22.2 ± 12.3%, 9 with left ventricular assist device [LVAD]) were treated with quinidine (14 quinidine gluconate; 996 ± 321 mg, 8 quinidine sulfate; 1062 ± 588 mg). Quinidine was used in combination with other antiarrhythmics (AAD) in 19 (13 also on amiodarone). All patients previously failed >1 AAD (amiodarone 100%, mexiletine 73%, sotalol 32%, other 32%) and eight had prior ablations (median of 1.5). Quinidine was initiated in the setting of VT storm despite AADs (6), inability to tolerate other AADs (4), or recurrent VT(12). Ventricular arrhythmias recurred despite quinidine in 13 (59%) patients at a median of 26 (4-240) days after quinidine initiation. In patients with recurrent MMVT, VT cycle length increased from 359 to 434 ms (p = .02). Six (27.3%) patients remained on quinidine at 1 year with recurrence of ventricular arrhythmias in all. The following adverse effects were seen: gastrointestinal side effects (6), QT prolongation (2), rash (1), thrombocytopenia (1), neurologic side effects (1). One patient discontinued due to cost. CONCLUSION: Quinidine therapy has limited tolerability and long-term efficacy when used in the management of amiodarone-refractory scar-related MMVT.

Lead damage after cardiac implantable device replacement procedure: Comparison between electrical plasma tool and electrocautery.

BACKGROUND: Lead damage is a complication caused by lead manipulation or heating damage from conventional electrocautery (EC) after cardiovascular implantable electronic device (CIED) replacement. Application of electrical plasma (PEAK PlasmaBlade) is a new technology that reportedly reduces this risk. OBJECTIVES: This study was designed to compare the effect of EC versus PEAK PlasmaBlade on lead parameters and complications after generator replacement procedures. METHODS: We retrospectively studied 410 consecutive patients (840 leads) who underwent CIED replacement using EC (EC group) and 410 consecutive patients (824 leads) using PEAK PlasmaBlade (PlamaBlade group). Pacing lead impedance, incidence of lead damage, and complications were compared between both groups. RESULTS: Lead impedance increased in 393 leads (46.8%) in the EC group versus 282 leads (34.2%) in the PlasmaBlade group (p < .01) with average percent changes of 6.7% and 4.0% (p < .01), respectively. Lead impedance decreased in 438 leads (52.1%) in the EC group versus 507 leads (61.5%) in the PlasmaBlade group (p < .01) with average percent changes of -5.7% and -7.1% (p < .01), respectively. Lead damage requiring lead revision occurred in five leads (0.6%) or after five procedures (1.2%) in the EC group compared to three leads (0.4%, p = .50) or after three procedures (0.7%, p = .48) in the PlasmaBlade group. There were no significant differences in the procedural-related complications between the EC group (nine patients, 2.2%) and the PlasmaBlade group (five patients, 1.2%, p = .28). CONCLUSION: Conventional electrocautery can potentially damage lead insulations. However, this study shows that when used carefully electrocautery is as safe as the PEAK PlasmaBlade™.

Effect of metformin on outcomes of catheter ablation for atrial fibrillation.

BACKGROUND: Diabetes mellitus (DM) is a risk factor for atrial fibrillation (AF). The effect of antidiabetic medications on AF or the outcomes of catheter ablation (CA) has not been well described. We sought to determine whether metformin treatment is associated with a lower risk of atrial arrhythmias after CA in patients with DM and AF. METHODS AND RESULTS: A first CA was performed in 271 consecutive patients with DM and AF (age: 65 ± 9 years, women: 34%; and paroxysmal AF: 51%). At a median of 13 months after CA (interquartile range: 6-30), 100/182 patients (55%) treated with metformin remained in sinus rhythm without antiarrhythmic drug therapy, compared with 36/89 patients (40%) not receiving metformin (p = .03). There was a significant association between metformin therapy and freedom from recurrent atrial arrhythmias after CA in multivariable Cox hazards models (hazard ratio [HR]: 0.66; ±95% confidence interval [CI]: 0.44-0.98; p = .04) that adjusted for age, sex, body mass index, AF type (paroxysmal vs. nonparoxysmal), antiarrhythmic medication, obstructive sleep apnea, chronic kidney disease, coronary artery disease, left ventricular ejection fraction, and left atrial diameter. A Cox model that also incorporated other antidiabetic agents and fasting blood glucose demonstrated a similar reduction in the risk of recurrent atrial arrhythmias with metformin treatment (HR: 0.63; ±95% CI: 0.42-0.96; p = .03). CONCLUSIONS: In patients with DM, treatment with metformin appears to be independently associated with a significant reduction in the risk of recurrent atrial arrhythmias after CA for AF. Whether this effect is due to glycemic control or pleiotropic effects on electroanatomical mechanisms of AF remains to be determined.

Predicting atrial fibrillation episodes with rapid ventricular rates associated with low levels of activity.

BACKGROUND: Rapid and irregular ventricular rates (RVR) are an important consequence of atrial fibrillation (AF). Raw accelerometry data in combination with electrocardiogram (ECG) data have the potential to distinguish inappropriate from appropriate tachycardia in AF. This can allow for the development of a just-in-time intervention for clinical treatments of AF events. The objective of this study is to develop a machine learning algorithm that can distinguish episodes of AF with RVR that are associated with low levels of activity. METHODS: This study involves 45 patients with persistent or paroxysmal AF. The ECG and accelerometer data were recorded continuously for up to 3 weeks. The prediction of AF episodes with RVR and low activity was achieved using a deterministic probabilistic finite-state automata (DPFA)-based approach. Rapid and irregular ventricular rate (RVR) is defined as having heart rates (HR) greater than 110 beats per minute (BPM) and high activity is defined as greater than 0.75 quantile of the activity level. The AF events were annotated using the FDA-cleared BeatLogic algorithm. Various time intervals prior to the events were used to determine the longest prediction intervals for predicting AF with RVR episodes associated with low levels of activity. RESULTS: Among the 961 annotated AF events, 292 met the criterion for RVR episode. There were 176 and 116 episodes with low and high activity levels respectively. Out of the 961 AF episodes, 770 (80.1%) were used in the training data set and the remaining 191 intervals were held out for testing. The model was able to predict AF with RVR and low activity up to 4.5 min before the events. The mean prediction performance gradually decreased as the time to events increased. The overall Area under the ROC Curve (AUC) for the model lies within the range of 0.67-0.78. CONCLUSION: The DPFA algorithm can predict AF with RVR associated with low levels of activity up to 4.5 min before the onset of the event. This would enable the development of just-in-time interventions that could reduce the morbidity and mortality associated with AF and other similar arrhythmias.

Multiparametric assessment of left atrial remodeling using 18F-FDG PET/CT cardiac imaging: A pilot study.

BACKGROUND: Left atrial (LA) remodeling is associated with structural, electric, and metabolic LA changes. Integrated evaluation of these features in vivo is lacking. METHODS: Patients undergoing 18F-fluorodeoxyglucose (FDG) PET-CT during a hyperinsulinemic-euglycemic clamp were classified into sinus rhythm (SR), paroxysmal AF (PAF), and persistent AF (PerAF). The LA was semiautomatically segmented, and global FDG uptake was quantified using standardized uptake values (SUVmax and SUVmean) in gated, attenuation-corrected images and normalized to LA blood pool activity. Regression was used to relate FDG data to AF burden and critical patient factors. Continuous variables were compared using t-tests or Mann-Whitney tests. RESULTS: 117 patients were included (76% men, age 66.4 ± 11.0, ejection fraction (EF) 25[22-35]%) including those with SR (n = 48), PAF (n = 55), and PerAF (n = 14). Patients with any AF had increased SUVmean (2.3[1.5-2.4] vs 2.0[1.5-2.5], P = 0.006), SUVmax (4.4[2.8-6.7] vs 3.2[2.3-4.3], P < 0.001), uptake coefficient of variation (CoV) 0.28[0.22-0.40] vs 0.25[0.2-0.33], P < 0.001), and hypometabolic scar (32%[14%-53%] vs 16.5%[0%-38.5%], P = 0.01). AF burden correlated with increased SUVmean, SUVmax, CoV, and scar independent of age, gender, EF, or LA size (P < 0.03 for all). CONCLUSIONS: LA structure and metabolism can be assessed using FDG PET/CT. Greater AF burden correlates with the increased LA metabolism and scar.

Baseline and decline in device-derived activity level predict risk of death and heart failure in patients with an ICD for primary prevention.

BACKGROUND: Implanted defibrillators are capable of recording activity data based on company-specific proprietary algorithms. This study aimed to determine the prognostic significance of baseline and decline in device-derived activity level across different device companies in the real world. METHODS: We performed a retrospective cohort study of patients (n = 280) who underwent a defibrillator implantation (Boston, Medtronic, St. Jude, and Biotronik) for primary prevention at the University of Michigan from 2014 to 2016. Graphical data obtained from device interrogations were retrospectively converted to numerical data. The activity level averaged over a month from a week postimplantation was used as baseline. Subsequent weekly average activity levels (SALs) were standardized to this baseline. SAL below 59.4% was used as a threshold to group patients. All-cause mortality and death/heart failure were the primary end-points of this study. RESULTS: Fifty-six patients died in this study. On average, they experienced a 50% decline in SAL prior to death. Patients (n = 129) who dropped their SAL below threshold were more likely to be older, male, diabetic, and have more symptomatic heart failure. They also had a significantly increased risk of heart failure/death (hazard ratio [HR] 3.6, 95% confidence interval [95% CI] 2.3-5.8, P < .0001) or death (HR 4.2, 95% CI 2.2-7.7, P < .0001) compared to those who had sustained activity levels. Lower baseline activity level was also associated with significantly increased risk of heart failure/death and death. CONCLUSION: Significant decline in device-derived activity level and low baseline activity level are associated with increased mortality and heart failure in patients with an ICD for primary prevention.

Ablation of paroxysmal atrial fibrillation using a second-generation cryoballoon catheter or contact-force sensing radiofrequency ablation catheter: A comparison of costs and long-term clinical outcomes.

INTRODUCTION: Although noninferiority of cryoballoon ablation (CBA) and radiofrequency catheter ablation for antral pulmonary vein isolation (APVI) has been reported in patients with paroxysmal atrial fibrillation (PAF), it is not clear whether contact force sensing (CF-RFA) and CBA with the second-generation catheter have similar procedural costs and long-term outcomes. The objective of this study is to compare the long-term efficacy and cost implications of CBA and CF-RFA in patients with PAF. METHODS AND RESULTS: A first APVI was performed in 146 consecutive patients (age: 63 ± 10 years, men: 95 [65%], left atrial diameter: 42 ± 6 mm) with PAF using CBA (71) or CF-RFA (75). Clinical outcomes and procedural costs were compared. The mean procedure time was significantly shorter with CBA than with CF-RFA (98 ± 39 vs. 158 ± 47 minutes, P < 0.0001). Despite a higher equipment cost in the CBA than the CF-RFA group, the total procedure cost was similar between the two groups (P = 0.26), primarily driven by a shorter procedure duration that resulted in a lower anesthesia cost. At 25 ± 5 months after a single ablation procedure, 51 patients (72%) in the CBA, and 55 patients (73%) in the CF-RFA groups remained free from atrial arrhythmias without antiarrhythmic drug therapy (P = 0.84). CONCLUSIONS: The procedure duration was approximately 60 minutes shorter with CBA than CF-RFA. The procedural costs were similar with both approaches. At 2 years after a single procedure, CBA and CF-RFA have similar single-procedure efficacies of 72-73%.

Asphyxia-activated corticocardiac signaling accelerates onset of cardiac arrest.

The mechanism by which the healthy heart and brain die rapidly in the absence of oxygen is not well understood. We performed continuous electrocardiography and electroencephalography in rats undergoing experimental asphyxia and analyzed cortical release of core neurotransmitters, changes in brain and heart electrical activity, and brain-heart connectivity. Asphyxia stimulates a robust and sustained increase of functional and effective cortical connectivity, an immediate increase in cortical release of a large set of neurotransmitters, and a delayed activation of corticocardiac functional and effective connectivity that persists until the onset of ventricular fibrillation. Blocking the brain's autonomic outflow significantly delayed terminal ventricular fibrillation and lengthened the duration of detectable cortical activities despite the continued absence of oxygen. These results demonstrate that asphyxia activates a brainstorm, which accelerates premature death of the heart and the brain.

Game Theoretic Approach for Systematic Feature Selection; Application in False Alarm Detection in Intensive Care Units.

Intensive Care Units (ICUs) are equipped with many sophisticated sensors and monitoring devices to provide the highest quality of care for critically ill patients. However, these devices might generate false alarms that reduce standard of care and result in desensitization of caregivers to alarms. Therefore, reducing the number of false alarms is of great importance. Many approaches such as signal processing and machine learning, and designing more accurate sensors have been developed for this purpose. However, the significant intrinsic correlation among the extracted features from different sensors has been mostly overlooked. A majority of current data mining techniques fail to capture such correlation among the collected signals from different sensors that limits their alarm recognition capabilities. Here, we propose a novel information-theoretic predictive modeling technique based on the idea of coalition game theory to enhance the accuracy of false alarm detection in ICUs by accounting for the synergistic power of signal attributes in the feature selection stage. This approach brings together techniques from information theory and game theory to account for inter-features mutual information in determining the most correlated predictors with respect to false alarm by calculating Banzhaf power of each feature. The numerical results show that the proposed method can enhance classification accuracy and improve the area under the ROC (receiver operating characteristic) curve compared to other feature selection techniques, when integrated in classifiers such as Bayes-Net that consider inter-features dependencies.

Diagnostic Evaluation and Follow-Up of Patients with Atrial Fibrillation.

In this article, a review of the diagnostic evaluation and outpatient follow-up of patients with atrial fibrillation is presented. After exploring details of symptoms, past medical history, quality of life, and physical exam findings, diagnostic tools are then discussed. Furthermore, important considerations after the initial diagnosis and treatment of patients with atrial fibrillation are discussed.

Promises and Perils of Consumer Mobile Technologies in Cardiovascular Care: JACC Scientific Statement.

Direct-to-consumer (D2C) wearables are becoming increasingly popular in cardiovascular health management because of their affordability and capability to capture diverse health data. Wearables may enable continuous health care provider-patient partnerships and reduce the volume of episodic clinic-based care (thereby reducing health care costs). However, challenges arise from the unregulated use of these devices, including questionable data reliability, potential misinterpretation of information, unintended psychological impacts, and an influx of clinically nonactionable data that may overburden the health care system. Further, these technologies could exacerbate, rather than mitigate, health disparities. Experience with wearables in atrial fibrillation underscores these challenges. The prevalent use of D2C wearables necessitates a collaborative approach among stakeholders to ensure effective integration into cardiovascular care. Wearables are heralding innovative disease screening, diagnosis, and management paradigms, expanding therapeutic avenues, and anchoring personalized medicine.

The potential of artificial intelligence to revolutionize health care delivery, research, and education in cardiac electrophysiology.

The field of electrophysiology (EP) has benefited from numerous seminal innovations and discoveries that have enabled clinicians to deliver therapies and interventions that save lives and promote quality of life. The rapid pace of innovation in EP may be hindered by several challenges including the aging population with increasing morbidity, the availability of multiple costly therapies that, in many instances, confer minor incremental benefit, the limitations of healthcare reimbursement, the lack of response to therapies by some patients, and the complications of the invasive procedures performed. To overcome these challenges and continue on a steadfast path of transformative innovation, the EP community must comprehensively explore how artificial intelligence (AI) can be applied to healthcare delivery, research, and education and consider all opportunities in which AI can catalyze innovation; create workflow, research, and education efficiencies; and improve patient outcomes at a lower cost. In this white paper, we define AI and discuss the potential of AI to revolutionize the EP field. We also address the requirements for implementing, maintaining, and enhancing quality when using AI and consider ethical, operational, and regulatory aspects of AI implementation. This manuscript will be followed by several perspective papers that will expand on some of these topics.

A Comprehensive Comparison of Six Publicly Available Algorithms for Localization of QRS Complex on Electrocardiograph.

The QRS complex is the most prominent feature of the electrocardiogram (ECG) that is used as a marker to identify the cardiac cycles. Identification of QRS complex locations enables arrhythmia detection and heart rate variability estimation. Therefore, accurate and consistent localization of the QRS complex is an important component of automated ECG analysis which is necessary for the early detection of cardiovascular diseases. This study evaluates the performance of six popular publicly available QRS complex detection methods on a large dataset of over half a million ECGs in a diverse population of patients. We found that a deep-learning method that won first place in the 2019 Chinese physiological challenge (CPSC-1) outperforms the remaining five QRS complex detection methods with an F1 score of 98.8% and an absolute sdRR error of 5.5 ms. We also examined the stratified performance of the studied methods on various cardiac conditions. All six methods had a lower performance in the detection of QRS complexes in ECG signals of patients with pacemakers, complete atrioventricular block, or indeterminate cardiac axis. We also concluded that, in the presence of different cardiac conditions, CPSC-1 is more robust than Pan-Tompkins which is the most popular model for QRS complex detection. We expect that this study can potentially serve as a guide for researchers on the appropriate QRS detection method for their target applications.Clinical Relevance-This study highlights the overall performance of publicly available QRS detection algorithms in a large dataset of diverse patients. We showed that there are specific cardiac conditions that are associated with the poor performance of QRS detection algorithms and may adversely influence the performance of algorithms that rely on accurate and reliable QRS detection.

Episodes of Atrial Fibrillation and Symptoms: A Temporal Analysis.

Background: Data on the relationship between symptoms and atrial fibrillation (AF) episodes are limited. Objective: The objective of this study was to determine the strength of temporal association between AF episodes and symptoms. Methods: This cross-sectional ambulatory assessment study was performed in a tertiary care center between June 2018 and December 2021. Patients with paroxysmal AF (1 episode of AF, burden not exceeding 95%) who used a mobile application and continuous wearable electrocardiogram monitor for 21 days were enrolled. The primary outcome was worse symptoms (symptoms above the mean score) over the study period. The association between worse symptoms and the presence of AF was evaluated for different time epochs. Multilevel mixed-effects models were used to quantify associations after accounting for confounders. Results: Worse symptoms were more likely to be associated with the presence of AF episodes 15 minutes prior to the reporting of palpitations (OR, 2.8 [95% CI, 1.6-5.0]; P < .001), shortness of breath (OR, 2.2 [95% CI, 1.3-3.7]; P = .003), dizziness/lightheadedness (OR, 2.0 [95% CI, 1.0-3.7]; P = .04), and fatigue (OR, 1.7 [95% CI, 1.0-2.9]; P = .03). The correlation between the severity of symptoms and AF lessened as the time interval from AF events to symptoms increased. Conclusion: There is a significant relationship between onset of AF episodes and reporting of symptoms. This association diminishes over time and varies across different symptoms. If confirmed in larger studies, these findings may inform AF interventions that target symptoms just in time prior to a clinical visit.