CHADS2 and CHA2DS2-VASc Scores Predict the Risk of Ischemic Stroke Outcome in Patients with Interatrial Block without Atrial Fibrillation

NaV1.5 autoantibodies in Brugada syndrome: pathogenetic implications

BACKGROUND AND AIMS

Patients suffering from Brugada syndrome (BrS) are predisposed to life-threatening cardiac arrhythmias. Diagnosis is challenging due to the elusive electrocardiographic (ECG) signature that often requires unconventional ECG lead placement and drug challenges to be detected. Although NaV1.5 sodium channel dysfunction is a recognized pathophysiological mechanism in BrS, only 25% of patients have detectable SCN5A variants. Given the emerging role of autoimmunity in cardiac ion channel function, this study explores the presence and potential impact of anti-NaV1.5 autoantibodies in BrS patients.

METHODS

Using engineered HEK293A cells expressing recombinant NaV1.5 protein, plasma from 50 BrS patients and 50 controls was screened for anti-NaV1.5 autoantibodies via western blot, with specificity confirmed by immunoprecipitation and immunofluorescence. The impact of these autoantibodies on sodium current density and their pathophysiological effects were assessed in cellular models and through plasma injection in wild-type mice.

RESULTS

Anti-NaV1.5 autoantibodies were detected in 90% of BrS patients vs. 6% of controls, yielding a diagnostic area under the curve of .92, with 94% specificity and 90% sensitivity. These findings were consistent across varying patient demographics and independent of SCN5A mutation status. Electrophysiological studies demonstrated a significant reduction specifically in sodium current density. Notably, mice injected with BrS plasma showed Brugada-like ECG abnormalities, supporting the pathogenic role of these autoantibodies.

CONCLUSIONS

The study demonstrates the presence of anti-NaV1.5 autoantibodies in the majority of BrS patients, suggesting an immunopathogenic component of the syndrome beyond genetic predispositions. These autoantibodies, which could serve as additional diagnostic markers, also prompt reconsideration of the underlying mechanisms of BrS, as evidenced by their role in inducing the ECG signature of the syndrome in wild-type mice. These findings encourage a more comprehensive diagnostic approach and point to new avenues for therapeutic research.

Edge AI Model Deployed for Real-Time Detection of Atrial Fibrillation Risk during Sinus Rhythm

Objectives: The study aimed to develop a deep learning-based edge AI model deployed on electrocardiograph (ECG) devices for the real-time detection of atrial fibrillation (AF) risk during sinus rhythm (SR) using standard 10 s, 12-lead electrocardiograms (ECGs). Methods: A novel approach was used to convert standard 12-lead ECGs into binary images for model input, and a lightweight convolutional neural network (CNN)-based model was trained using data collected by the Japan Agency for Medical and Research Development (AMED) between 2019 and 2022. Patients over 40 years old with digital, SR ECGs were retrospectively enrolled and divided into AF and non-AF groups. The data labeling was supervised by cardiologists. The dataset was randomly allocated into training, validation, and internal testing datasets. External testing was conducted on data collected from other hospitals. Results: The best-trained model achieved an AUC of 0.82 and 0.80, sensitivity of 79.5% and 72.3%, specificity of 77.8% and 77.7%, precision of 78.2% and 76.4%, and overall accuracy of 78.6% and 75.0% in the internal and external testing datasets, respectively. The deployed model and app package utilized 2.5 MB and 40 MB of the available ROM and RAM capacity on the edge ECG device, correspondingly. The processing time for AF risk detection was approximately 2 s. Conclusions: The model maintains comparable performance and improves its suitability for deployment on resource-constrained ECG devices, thereby expanding its potential impact to a wide range of healthcare settings. Its successful deployment enables real-time AF risk detection during SR, allowing for timely intervention to prevent AF-related serious consequences like stroke and premature death.

Deep learning unmasks the ECG signature of Brugada syndrome

One in 10 cases of sudden cardiac death strikes without warning as the result of an inherited arrhythmic cardiomyopathy, such as Brugada Syndrome (BrS). Normal physiological variations often obscure visible signs of this and related life-threatening channelopathies in conventional electrocardiograms (ECGs). Sodium channel blockers can reveal previously hidden diagnostic ECG features, however, their use carries the risk of life-threatening proarrhythmic side effects. The absence of a nonintrusive test places a grossly underestimated fraction of the population at risk of SCD. Here, we present a machine-learning algorithm that extracts, aligns, and classifies ECG waveforms for the presence of BrS. This protocol, which succeeds without the use of a sodium channel blocker (88.4% accuracy, 0.934 AUC in validation), can aid clinicians in identifying the presence of this potentially life-threatening heart disease.

CHA2DS2-VASc score, P-wave indexes, and echocardiographic parameters in sinus rhythm patients without valvular heart disease

OBJECTIVE

The aim of this study was to evaluate the correlation between P-wave indexes, echocardiographic parameters, and CHA2DS2-VASc score in patients without atrial fibrillation and valvular disease.

METHODS

This retrospective cross-sectional study included patients of a tertiary hospital with no history of atrial fibrillation, atrial flutter, or valve disease and collected data from June 2021 to May 2022. The exclusion criteria were as follows: unavailable medical records, pacemaker carriers, absence of echocardiogram report, or uninterpretable ECG. Clinical, electrocardiographic [i.e., P-wave duration, amplitude, dispersion, variability, maximum, minimum, and P-wave voltage in lead I, Morris index, PR interval, P/PR ratio, and P-wave peak time], and echocardiographic data [i.e., left atrium and left ventricle size, left ventricle ejection fraction, left ventricle mass, and left ventricle indexed mass] from 272 patients were analyzed.

RESULTS

PR interval (RHO=0.13, p=0.032), left atrium (RHO=0.301, p<0.001) and left ventricle diameter (RHO=0.197, p=0.001), left ventricle mass (RHO=0.261, p<0.001), and left ventricle indexed mass (RHO=0.340, p<0.001) were positively associated with CHA2DS2-VASc score, whereas P-wave amplitude (RHO=-0.141, p=0.02), P-wave voltage in lead I (RHO=-0.191, p=0.002), and left ventricle ejection fraction (RHO=-0.344, p<0.001) were negatively associated with the same score. The presence of the Morris index was associated with high CHA2DS2-VASc (p=0.022).

CONCLUSION

Prolonged PR interval, Morris index, increased left atrium diameter, left ventricle diameter, left ventricle mass, and left ventricle indexed mass values as well as lower P-wave amplitude, P-wave voltage in lead I, and left ventricle ejection fraction values were correlated with higher CHA2DS2-VASc scores.

P-wave duration and interatrial block as predictors of new-onset atrial fibrillation: A systematic review and meta-analysis

BACKGROUND

Early detection of atrial fibrillation (AF) could improve patient outcomes. P-wave duration (PWD) and interatrial block (IAB) are known predictors of new-onset AF and could improve selection for AF screening. This meta-analysis reviews the published evidence and offers practical implications.

METHODS

Publication databases were systematically searched, and studies reporting PWD and/or morphology at baseline and new-onset AF during follow-up were included. IAB was defined as partial (pIAB) if PWD≥120 ms or advanced (aIAB) if the P-wave was biphasic in the inferior leads. After quality assessment and data extraction, random-effects analysis calculated odds ratio (OR) and confidence intervals (CI). Subgroup analysis was performed for those with implantable devices (continuous monitoring).

RESULTS

Among 16,830 patients (13 studies, mean 66 years old), 2,521 (15%) had new-onset AF over a median of 44 months. New-onset AF was associated with a longer PWD (mean pooled difference: 11.5 ms, 13 studies, p < 0.001). The OR for new-onset AF was 2.05 (95% CI: 1.3-3.2) for pIAB (5 studies, p = 0.002) and 3.9 (95% CI: 2.6-5.8) for aIAB (7 studies, p < 0.001). Patients with pIAB and devices had higher AF-detection risk (OR: 2.33, p < 0.001) than those without devices (OR: 1.36, p = 0.56). Patients with aIAB had similarly high risk regardless of device presence. There was significant heterogeneity but no publication bias.

CONCLUSION

Interatrial block is an independent predictor of new-onset AF. The association is stronger for patients with implantable devices (close monitoring). Thus, PWD and IAB could be used as selection criteria for intensive screening, follow-up or interventions.

An artificial intelligence-enabled ECG algorithm for identifying ventricular premature contraction during sinus rhythm

Background

Ventricular premature complex (VPC) is a common arrhythmia in clinical practice. VPC could trigger ventricular tachycardia/fibrillation or VPC-induced cardiomyopathy in susceptible patients. Existing screening methods require prolonged monitoring and are limited by cost and low yield when the frequency of VPC is low. Twelve-lead electrocardiogram (ECG) is low cost and widely used. We aimed to identify patients with VPC during normal sinus rhythm (NSR) using artificial intelligence (AI) and machine learning-based ECG reading.

Methods

We developed AI-enabled ECG algorithm using a convolutional neural network (CNN) to detect the ECG signature of VPC presented during NSR using standard 12-lead ECGs. A total of 2515 ECG records from 398 patients with VPC were collected. Among them, only ECG records of NSR without VPC (1617 ECG records) were parsed.

Results

A total of 753 normal ECG records from 387 patients under NSR were used for comparison. Both image and time-series datasets were parsed for the training process by the CNN models. The computer architectures were optimized to select the best model for the training process. Both the single-input image model (InceptionV3, accuracy: 0.895, 95% confidence interval [CI] 0.683–0.937) and multi-input time-series model (ResNet50V2, accuracy: 0.880, 95% CI 0.646–0.943) yielded satisfactory results for VPC prediction, both of which were better than the single-input time-series model (ResNet50V2, accuracy: 0.840, 95% CI 0.629–0.952).

Conclusions

AI-enabled ECG acquired during NSR permits rapid identification at point of care of individuals with VPC and has the potential to predict VPC episodes automatically rather than traditional long-time monitoring.

Children's Hip Predictive (CHiP) Score: A Triage Tool for Hip Dislocation in Children Referred With Suspected Hip Dysplasia

BACKGROUND

A fundamental tenent of treating developmental dysplasia of the hip is to identify patients with dislocated hips early so as to avoid the long-term sequelae of late diagnosis. The aim of this study was to develop a readily useable triage tool for patients with suspected hip dislocation, based on the clinical history and examination findings of the referring practitioner.

METHODS

All primary care referrals (n=934) over a 3-year period for suspected developmental dysplasia of the hip to a tertiary pediatric center were evaluated. Defined parameters with respect to history and clinical examination were evaluated. Multivariable logistic regression was used to establish predictors of hip dislocation, and from this a predictive model was derived which incorporated significant predictors of dislocation. An illustrative nomogram translated this predictive model into a usable numerical scoring system called the Children's Hip Prediction score, which estimates probability of hip dislocation.

RESULTS

There were 97 dislocated hips in 85 patients. The final predictive model included age, sex, family history, breech, gait concerns, decreased abduction, leg length discrepancy, and medical/neurological syndrome. The area under receiver operating curve for the model is 0.761. A Children's Hip Prediction score of≥5 corresponds to a sensitivity of 76.3% and a score of≥15 has a specificity of 97.8%, corresponding to an odds ratio of 27.3 for increased risk of dislocation.

CONCLUSION

We found that a novel clinical prediction score, based on readily available history and examination parameters strongly predicted risk of dislocations in hip dysplasia referral. It is hoped that this tool could be utilized to optimize resource allocation and may be of particular benefit in less well-resourced health care systems.

LEVEL OF EVIDENCE

Level II.

Is there a Concordance between CHA2DS2 VASc and HAS-BLED Scores in Middle Eastern Patients with Nonvalvular AF? Analysis of the Jordan Atrial Fibrillation (JoFib) Study

Background:

Nonvalvular Atrial Fibrillation (NVAF) is a common arrhythmia that potentially contributes to stroke and systemic embolization. Anticoagulants may contribute to the risk of bleeding in these patients. Reports of NVAF on Middle Eastern populations are scarce and outdated. This study investigated the concordance between congestive heart failure, hypertension, ≥75 years of age, diabetes mellitus, stroke or transient ischemic attack, vascular disease, 65 to 74 years of age group, sex category (CHA2DS2 VASc), and Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol Concomitantly (HAS-BLED) scores, and the risks of stroke and bleeding in Jordanian NVAF patients.

Methods:

Results were extracted from the Jordan Atrial Fibrillation (JoFib) multicenter registry (NCT03917992). To assess the risk of stroke, a CHA2DS2 VASc score was used. The HAS-BLED score was used to assess the risk of bleeding. Subgroup analysis for males and females was carried out.

Results:

We included 1823 NVAF patients, almost equally distributed among males and females. The overall mean age was 68.2 years. Most patients had a body mass index (BMI) ≥ 25 (71.5%). Almost half of the patients were smokers, 45.5% had diabetes, and 76.3% had hypertension. Paroxysmal AF was the most common subtype (38%), followed by the permanent (28.6%), long-standing (17.3%), and persistent types (16.1%). Females had a significantly higher CHA2DS2 VASc score than males (p=0.012). For the risk of bleeding, 48.9% of patients had a low risk, 31.5% had intermediate risk, and 19.6% had a high risk. A correlation was found between the two scoring systems. Out of the 357 patients who had a high HAS-BLED score, 354 (99.2%) patients also had a high CHA2DS2 VASc score.

Conclusion:

In almost all NVAF patients with a high risk of bleeding (high HAS-BLED score), the use of anticoagulants is highly recommended and strongly indicated according to the CHA2DS2 VASc score stratification. Therefore, strict strategies of well-controlled administration of anticoagulants should always be considered and followed in Middle Eastern patients with NVAF and a high risk of bleeding.

Electrocardiographic predictors of atrial fibrillation in patients with cryptogenic stroke

BACKGROUND

Empiric anticoagulation is not routinely indicated in patients with cryptogenic stroke without documentation of atrial fibrillation (AF). Therefore, identification of patients at increased risk of AF from this vulnerable group is vital.

OBJECTIVES

To identify electrocardiographic (ECG) predictors of AF in patients with cryptogenic stroke or transient ischemic attack (TIA) undergoing insertion of an implantable cardiac monitor (ICM).

METHODS

In this single-center study, 48 patients with cryptogenic stroke or TIA had an ICM implanted for detection of AF between January 2013 and September 2019. Patients with and without AF were compared in terms of p-wave duration and a novel index (MVP score).

RESULTS

During a mean follow-up of 16±14 months, AF was detected in 7 patients (15%). Diagnosis of AF was made after a mean of 10 ± 14 months, with time to first AF detection ranging between 1 and 40 months. Patients with AF had a longer p-wave duration (136±9 ms vs 116±10 ms; p = 0.0001) and a higher MVP score (4.5±1.2 vs 2.0±0.9, p = 0.0001) than those without AF. Advanced interatrial block was observed in 43% of patients with ICM evidence of AF and 0% of those without AF (p = 0.002). Age, LA size or LVEF were not predictors of AF.

CONCLUSION

An increased p-wave duration, advanced interatrial block and high MVP score are associated with AF occurrence in patients with cryptogenic stroke. Identifying patients with these markers may be helpful as they may benefit from more exhaustive and prolonged monitoring. This article is protected by copyright. All rights reserved.

Machine learning aids clinical decision-making in patients presenting with angina and non-obstructive coronary artery disease

Aims

The current gold standard comprehensive assessment of coronary microvascular dysfunction (CMD) is through a limited-access invasive catheterization lab procedure. We aimed to develop a point-of-care tool to assist clinical guidance in patients presenting with chest pain and/or an abnormal cardiac functional stress test and with non-obstructive coronary artery disease (NOCAD).

Methods and results

This study included 1893 NOCAD patients (<50% angiographic stenosis) who underwent CMD evaluation as well as an electrocardiogram (ECG) up to 1-year prior. Endothelial-independent CMD was defined by coronary flow reserve (CFR) ≤2.5 in response to intracoronary adenosine. Endothelial-dependent CMD was defined by a maximal percent increase in coronary blood flow (%ΔCBF) ≤50% in response to intracoronary acetylcholine infusion. We trained algorithms to distinguish between the following outcomes: CFR ≤2.5, %ΔCBF ≤50, and the combination of both. Two classes of algorithms were trained, one depending on ECG waveforms as input, and another using tabular clinical data. Mean age was 51 ± 12 years and 66% were females (n = 1257). Area under the curve values ranged from 0.49 to 0.67 for all the outcomes. The best performance in our analysis was for the outcome CFR ≤2.5 with clinical variables. Area under the curve and accuracy were 0.67% and 60%. When decreasing the threshold of positivity, sensitivity and negative predictive value increased to 92% and 90%, respectively, while specificity and positive predictive value decreased to 25% and 29%, respectively.

Conclusion

An artificial intelligence-enabled algorithm may be able to assist clinical guidance by ruling out CMD in patients presenting with chest pain and/or an abnormal functional stress test. This algorithm needs to be prospectively validated in different cohorts.

Current ECG Aspects of Interatrial Block

Interatrial blocks like other types of block may be of first degree or partial second degree, also named transient atrial block or atrial aberrancy, and third degree or advanced. In first degree, partial interatrial block (P-IAB), the electrical impulse is conducted to the left atrium, through the Bachmann’s region, but with delay. The ECG shows a P-wave ≥ 120 ms. In third-degree, advanced interatrial block (A-IAB), the electrical impulse is blocked in the upper part of the interatrial septum (Bachmann region); the breakthrough to LA has to be performed retrogradely from the AV junction zone. This explains the p ± in leads II, III and aVF. In typical cases of A-IAB, the P-wave morphology is biphasic (±) in leads II, III and aVF, because the left atrium is activated retrogradely and, therefore, the last part of the atrial activation falls in the negative hemifield of leads II, III and aVF. Recently, some atypical cases of A-IAB have been described. The presence of A-IAB is a risk factor for atrial fibrillation, stroke, dementia, and premature death.

A Case of Cardiac Papillary Fibroelastoma - An Increasingly Described Cardiac Tumor with Fatal Consequences

Papillary fibroelastomas (PFE) are rare primary cardiac tumors characterized by non-malignant, pedunculated, endocardial lesions with a significant risk of embolic potential and death. With improvements in the imaging quality and availability of transthoracic echocardiograms (TTE), the diagnosis of PFE has become more common in the last 2 decades. PFE is changing from a rare "zebra" diagnosis to one that community providers will encounter in their practice and must appropriately treat to prevent morbidity and mortality. Data shows that there are significant survival and morbidity benefit associated with surgical excision over non-operative management, with the benefit of anticoagulation remaining unclear at this time. We report a case describing the diagnostic workup and management of a 58-year-old woman who presented with an unidentified endocardial mass determined to be a PFE. Based on current literature, we favor a strategy of early surgical excision of PFE for an optimal reduction in mortality and thromboembolic sequelae associated with this pathology.

Detection of Left Atrial Enlargement Using a Convolutional Neural Network-Enabled Electrocardiogram

Background: Left atrial enlargement (LAE) can independently predict the development of a variety of cardiovascular diseases.

Objectives: This study sought to develop an artificial intelligence approach for the detection of LAE based on 12-lead electrocardiography (ECG).

Methods: The study population came from an epidemiological survey of heart disease in Guangzhou. Elderly people (3,391) over 65 years old who had both 10-s 12 lead ECG and echocardiography were enrolled in this study. The left atrial (LA) anteroposterior diameter >40 mm on echocardiography was diagnosed as LAE, and the LA anteroposterior diameter was indexed by body surface area (BSA) to classify LAE into different degrees. A convolutional neural network (CNN) was trained and validated to detect LAE from normal ECGs. The performance of the model was evaluated by calculating the area under the curve (AUC), accuracy, sensitivity, specificity, and F1 score.

Results: In this study, gender, obesity, hypertension, and valvular heart disease seemed to be related to left atrial enlargement. The AI-enabled ECG identified LAE with an AUC of 0.949 (95% CI: 0.911–0.987). The sensitivity, specificity, accuracy, precision, and F1 score were 84.0%, 92.0%, 88.0%, 91.3%, and 0.875, respectively. Physicians identified LAE with sensitivity, specificity, accuracy, precision, and F1 scores of 38.0%, 84.0%, 61.0%, 70.4%, and 0.494, respectively. In classifying LAE in different degrees, the AUCs of identifying normal, mild LAE, and moderate-severe LAE ECGs were 0.942 (95% CI: 0.903–0.981), 0.951 (95% CI: 0.917–0.987), and 0.998 (95% CI: 0.996–1.00), respectively. The sensitivity, specificity, accuracy, positive predictive value, and F1 scores of diagnosing mild LAE were 82.0%, 92.0%, 88.7%, 89.1%, and 0.854, while the sensitivity, specificity, accuracy, positive predictive value, and F1 scores of diagnosing moderate-severe LAE were 98.0%, 84.0%, 88.7%, 96.1%, and 0.969, respectively.

Conclusions: An AI-enabled ECG acquired during sinus rhythm permits identification of individuals with a high likelihood of LAE. This model requires further refinement and external validation, but it may hold promise for LAE screening.

[Bayés Syndrome, Stroke and Dementia]

Bayés's syndrome is a clinical entity based on the association between advanced interatrial block and the development of supraventricular tachyarrhythmia, being atrial fibrillation (AF) the most frequent. This association was discovered by Prof. Antoni Bayés de Luna in the '80s. Further studies by other groups found a strong relationship between Bayés's syndrome and thromboembolic phenomena, being stroke the most serious. Moreover, patients with this syndrome have an increased incidence of cognitive impairment and dementia. This observation triggered the question about whether the use of anticoagulation therapy prior to the documentation of AF could prevent A-IAB associated thromboembolic events. There are ongoing studies in different phases of development aiming to compare the efficacy of anticoagulation in patients with A-IAB with no prior documentation of AF. The outcomes of these studies will allow determining the efficacy of this early therapeutic intervention, and help deciding the role of anticoagulation in patients with A-IAB and no demonstrated AF.

Bayés' Syndrome-A Comprehensive Short Review

Prediction and early detection of atrial fibrillation (AF) remain a permanent challenge in everyday practice. Timely identification of an increased risk for AF episodes (which are frequently asymptomatic) is essential in the primary and secondary prevention of cardioembolic events. One of the noninvasive modalities of AF prediction is represented by the electrocardiographic P-wave analysis. This includes the study and diagnosis of interatrial conduction block (Bachmann’s bundle block). Bayés’ Syndrome (named after its first descriptor) denotes the association between interatrial conduction defect and supraventricular arrhythmias (mainly AF) predisposing to cardioembolic events. Our short review presents an update of the most important data concerning this syndrome: brief history, main ECG features, pathophysiological background and clinical implications.

Relation of Advanced Interatrial Block to Risk of Atrial Fibrillation and Stroke

Advanced interatrial block (A-IAB) has been associated to atrial fibrillation (AF) and ischemic stroke, raising the question as to whether such patients, even when still in sinus rhythm without documented AF, could benefit from oral anticoagulation. AF and A-IAB are both markers of stroke. The anatomical substrate in both is fibrotic atrial cardiomyopathy, resulting in atrial electromechanical dyssynchrony, dysfunction, and left atrial remodelling, that favour blood stasis and hypercoagulation. Under these conditions thrombogenic cascade may be triggered, resulting in systemic embolization. Before proposing oral anticoagulation in the management of selected patients with A-IAB, as is currently recommended in patients with AF and high CHA₂DS₂-Vasc score, a randomized clinical trial will have to demonstrate efficacy and safety of anticoagulation in this setting. In the meantime, an individualized approach may be considered based on the recognition of those patients at a higher risk of stroke. These may be elderly patients with A-IAB and several risk factors and, thus, with a high CHA₂DS₂-Vasc score and the presence of environmental arrhythmias.

Beyond CHA2DS2-VASc in atrial fibrillation: the atrium and the risk of stroke

Pathology affecting the atria have a significant impact on the occurrence of arrhythmias and the risk of stroke. The causal relationship between atrial fibrillation (AF) and ischaemic stroke has been challenged by the recent uncovering of the lack of temporal association between thrombo-embolic cerebral events and paroxysmal AF or tachycardia. General conditions, such as the one considered in the definition of the CHA₂DS₂-VASc score, or specific atrial pathology (also independently occurring), could predispose to cerebral embolism.

Ischemic Stroke Risk in Patients With Nonvalvular Atrial Fibrillation: JACC Review Topic of the Week

The last decade has witnessed remarkable advances in pharmacological and nonpharmacological strategies for stroke prevention in patients with atrial fibrillation. However, the currently available clinical stroke risk prediction models do not account for key nonclinical factors (arrhythmia burden, left atrial physiology and anatomy, chemical and electrocardiographic markers) and other competing clinical risks. Hence, their ability to identify patients who will derive the most benefit from pharmacological and mechanical risk prevention strategies remain limited. In this paper, the authors review the current and evolving ischemic stroke risk prediction schemes in patients with nonvalvular atrial fibrillation, highlight the strengths and weaknesses of the models, and discuss the unmet needs in this field.

An artificial intelligence-enabled ECG algorithm for the identification of patients with atrial fibrillation during sinus rhythm: a retrospective analysis of outcome prediction

BACKGROUND

Atrial fibrillation is frequently asymptomatic and thus underdetected but is associated with stroke, heart failure, and death. Existing screening methods require prolonged monitoring and are limited by cost and low yield. We aimed to develop a rapid, inexpensive, point-of-care means of identifying patients with atrial fibrillation using machine learning.

METHODS

We developed an artificial intelligence (AI)-enabled electrocardiograph (ECG) using a convolutional neural network to detect the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm using standard 10-second, 12-lead ECGs. We included all patients aged 18 years or older with at least one digital, normal sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the Mayo Clinic ECG laboratory between Dec 31, 1993, and July 21, 2017, with rhythm labels validated by trained personnel under cardiologist supervision. We classified patients with at least one ECG with a rhythm of atrial fibrillation or atrial flutter as positive for atrial fibrillation. We allocated ECGs to the training, internal validation, and testing datasets in a 7:1:2 ratio. We calculated the area under the curve (AUC) of the receiver operatoring characteristic curve for the internal validation dataset to select a probability threshold, which we applied to the testing dataset. We evaluated model performance on the testing dataset by calculating the AUC and the accuracy, sensitivity, specificity, and F1 score with two-sided 95% CIs.

FINDINGS

We included 180 922 patients with 649 931 normal sinus rhythm ECGs for analysis: 454 789 ECGs recorded from 126 526 patients in the training dataset, 64 340 ECGs from 18 116 patients in the internal validation dataset, and 130 802 ECGs from 36 280 patients in the testing dataset. 3051 (8·4%) patients in the testing dataset had verified atrial fibrillation before the normal sinus rhythm ECG tested by the model. A single AI-enabled ECG identified atrial fibrillation with an AUC of 0·87 (95% CI 0·86-0·88), sensitivity of 79·0% (77·5-80·4), specificity of 79·5% (79·0-79·9), F1 score of 39·2% (38·1-40·3), and overall accuracy of 79·4% (79·0-79·9). Including all ECGs acquired during the first month of each patient's window of interest (ie, the study start date or 31 days before the first recorded atrial fibrillation ECG) increased the AUC to 0·90 (0·90-0·91), sensitivity to 82·3% (80·9-83·6), specificity to 83·4% (83·0-83·8), F1 score to 45·4% (44·2-46·5), and overall accuracy to 83·3% (83·0-83·7).

INTERPRETATION

An AI-enabled ECG acquired during normal sinus rhythm permits identification at point of care of individuals with atrial fibrillation.

FUNDING

None.

Independent associations of blood pressure and body mass index with interatrial block: a cross-sectional study in general Chinese population

OBJECTIVES

This current study was performed to characterise the independent associations of obesity and hypertension with interatrial block (IAB) after adjusting for cardiovascular risk factors, echocardiographic left atrial diameter (LAD) and left ventricular mass index (LVMI) in a large general Chinese population.

DESIGN

A cross-sectional study.

SETTING AND PARTICIPANTS

A total of 11 956 permanent residents (≥35 years of age) from Liaoning Province in China were included in this study. Following the completion of a questionnaire, the enrolled participants were subjected to physical examinations, laboratory analyses, ECG and echocardiogram. Linear and logistic regression analyses were performed to evaluate the associations of hypertension and obesity with IAB.

OUTCOME MEASURES

IAB was defined as a prolongation of the P wave duration ≥120 ms on a digital 12-lead ECG.

RESULTS

The prevalence of IAB in hypertensive individuals was higher than the normotensive in both men (9.5 vs 5.9%; p<0.001) and women (6.6 vs 3.6%; p<0.001). In addition, the prevalence of IAB exhibited a sharp increase with advancing body mass index (BMI) in both men (from 4.9% to 13.0%) and women (from 3.5% to 6.9%) (ps- for trend <0.001). Multiple relevant clinical covariates, echocardiographic LAD and LVMI were adjusted in the multivariate linear and logistic regression analyses. The results revealed that systolic blood pressure, diastolic blood pressure and BMI were all independently associated with P wave duration (β=0.02, 0.09 and 0.25, respectively; all ps <0.005). Furthermore, hypertension was found to be independently associated with IAB (OR=1.27; p=0.018), while both overweight and obesity exhibited higher odds of IAB (OR=1.42 and 1.67, respectively; ps <0.005), compared with BMI <24.0 kg/m².

CONCLUSIONS

The key findings of this study highlighted that hypertension and overweight/obesity were independently and significantly associated with IAB in general Chinese population.

Value of Interatrial Block for the Prediction of Silent Ischemic Brain Lesions

Introduction

Previous studies demonstrated that interatrial block (IAB) is associated with atrial fibrillation (AF) in different clinical scenarios. The aim of our study was to determine whether IAB could predict silent ischemic brain lesions (sIBL), detected by magnetic resonance imaging (MRI).

Methods

Patients presented to a neurology clinic with transient ischemic attack (TIA) symptoms and underwent brain MRI were included to the study. sIBL were defined as lesions without corresponding clinical symptoms regarding lesion localization evaluated by two neurologists. A 12-lead surface ECG was obtained from each patient. IAB was defined as P-wave duration > 120 ms with (advanced IAB) or without (partial IAB) biphasic morphology in the inferior leads.

Results

sIBL was detected in 61 (49.6%) patients. Patients with sIBL were older (P<0.001), had more left ventricular hypertrophy (LVH) (P=0.02) and higher CHA2DS2-VASc score compared to those without (P<0.001). P-wave duration was significantly longer in patients with sIBL (124 [110.5 - 129] msvs 107 [102 - 116.3] ms) (P<0.001). IAB was diagnosed in 36 patients (59%) with sIBL (+) and in 11 patients (18%) with sIBL (-); p<0.001. Multivariate logistic regression analysis identified age [Odds ratio (OR), 1.061; 95% confidence interval (CI), 1.012 - 1.113; p=0.014], CHA2DS2-VASc score (OR, 1.758; 95% CI, 1.045 - 2.956; p=0.034), LVH (OR, 3.062; 95% CI, 1.161 - 8.076; p=0.024) and IAB (including both partial and advanced) (OR, 5.959; 95% CI, 2.269 - 15.653; p<0.001) as independent predictors of sIBL.

Conclusion

IAB is a strong predictor of sIBL and can be easily diagnosed by performing surface 12-lead ECG.

Simulation of inter atrial block based on a human atrial model

Inter atrial block (IAB) is a prevailing cardiac conduction abnormality that is under-recognized in clinical practice. IAB has strong association with atrial arrhythmia, left atrial enlargement, and electromechanical discordance, increasing the risk of atrial fibrillation (AF) and myocardial ischemia. IAB was generally believed to be caused by impaired conduction along the Bachmann bundle (BB). However, there are three other conduction pathways, including the fibers posteriorly in the vicinity of the right pulmonary veins (VRPV), transseptal fibers in the fossa ovalis (FO), and muscular bundles on the inferior atrial surface near the coronary sinus (CS). We hypothesized that the importance of BB on IAB might have been overestimated. To test this hypothesis, various combinations of conduction pathway blocks were simulated based on a realistic human atrial model to investigate their effects on the index of clinical diagnosis standard of IAB using a simulated 12-lead electrocardiogram (ECG). Firstly, the results showed that the BB block alone could not generate typical P wave morphology of IAB, and that the combination of BB and VRPV pathway block played important roles in the occurrence of IAB. Secondly, although single FO and CS pathways play subordinate roles in inter atrial conduction, their combination with BB and VRPV block could also produce severe IAB. In summary, this simulation study has demonstrated that the combinations of different inter atrial conduction pathways, rather than BB alone, resulted in ECG morphology of IAB. Attention needs to be paid to this in future pathophysiological and clinical studies of IAB.

Update on Bayés' syndrome: the association between an interatrial block and supraventricular arrhythmias

INTRODUCTION

The past few years have given rise to extensive research on an interatrial block and its clinical relevance, mainly its association with supraventricular arrhythmias. In 2015, the authors of this article reviewed the Bayes syndrome for the first time and after three years there has been so much evidence accumulated that it seems reasonable to rewrite an update, based fundamentally on the new findings. Focused on its relationship with cardioembolic strokes, today efforts are being targeted at understanding its pathophysiology, its diagnosis, and its prognostic implications, in order to learn if it should be treated. Areas covered: A non-systematic review of the literature was developed using the Pubmed and Cochrane databases, focusing on randomized clinical trials and large observational studies that evaluated new physiopathological and epidemiological aspects, new clinical scenarios in which it has been assessed and its association with dementia. Finally, those studies that proposed new possible treatments were reviewed. Expert commentary: Interatrial block is not only a predictor of supraventricular arrhythmias, is a subclinical disease that might be considered as a marker of risk for adverse outcomes. Although there is some evidence to suggest that early treatment may be beneficial, potential therapies have yet to be investigated.

Association Between CHADS2 Score and the Development of Interatrial Block

Interatrial block (IAB) is associated with a multitude of medical conditions. The aim of this retrospective study was to investigate whether CHADS₂ (congestive heart failure, hypertension, age ≥ 75 years, diabetes mellitus, prior stroke) score is positively associated with the development of IAB. A total of 1072 patients (men, 555; women, 517; mean age, 61 ± 14 years) were included in the study. P-wave duration was measured manually using a caliper. IAB was defined as a P-wave duration of ≥ 120 ms on a 12-lead electrocardiogram. CHADS₂ scores were calculated retrospectively. Among the 1072 patients, the prevalence of IAB was 36.1% (387/1072). In multivariate analysis, increased CHADS₂ score (odds ratio [OR], 1.810; 95% confidence interval [CI], 1.577-2.077; P < 0.001), coronary artery disease (OR, 1.536; 95% CI, 1.065-2.216; P = 0.022), and increased left atrial diameter (OR, 1.039; 95% CI, 1.008-1.071; P = 0.013) were independently associated with IAB. The percentages of patients with IAB among those with a CHADS₂ score of 0, 1, 2, 3, 4, 5, and 6 were 20.6%, 33.0%, 45.0%, 55.9%, 61.9%, 77.8%, and 100%, respectively (P < 0.001). There was a greater percentage of patients with a CHADS₂ score of ≥ 2 with IAB compared with a CHADS₂ score of < 2 (26.5% vsrsus 52.0%; P < 0.001). In receiver operating curve (ROC) analysis, CHADS₂ score (area under the curve, 0.670; 95% CI, 0.636-0.704; P < 0.001) was predictive of IAB. In conclusion, CHADS₂ score was significantly associated with the development of IAB in this study population.

Second-degree interatrial block: Brief review and concept

The advanced interatrial block (A-IAB) (P ≥ 120 ms plus ± pattern in II, III and aVF) corresponds at atrial level, to right or left advanced bundle branch block at ventricular level, and it is well known that these patterns may be seen transiently in relation to taquicardia or bradycardia (tachycardia or bradycardia dependent right or left bundle branch block). We present for the first time, the same phenomenon at atrial level. In one case, the A-IAB appears in relation to tachycardization and in the other disappears during a pause induced by ventricular premature complex.

Bayés' syndrome: Time to consider early anticoagulation?

In the past few decades, extensive research has been conducted on atrial conduction disorders and their clinical relevance. An association between interatrial block (IAB) and supraventricular arrhythmias [most commonly atrial fibrillation (AF)] has been discovered and extensively investigated. We coined the term "Bayés Syndrome" to describe this association, and the medical community has accepted the eponym in recognition to the scientist who discovered most of the aspects associated with it. In this non-systematic review, we will focus on the association between IAB and AF, with special emphasis on the value of the surface 12-lead ECG as a valid tool to predict AF.

P-wave and interatrial block: New predictor for atrial high rate episodes in patients with cardiac implantable electronic devices

BACKGROUND

The presence of interatrial block (IAB) is associated with the development of atrial fibrillation (AF). The aim of this study was to determine whether P-wave duration and presence of IAB before the implantation of a cardiac implantable electronic device (CIED) are associated with the presence of atrial high rate episodes (AHRE), during long-term follow-up.

METHODS

380 patients (57% men; 75 ± 10 years) were included. IAB was defined according to the International Consensus Criteria. AHRE was defined as an episode of atrial rate ≥225 beats/min with a minimum duration of 5 minutes.

RESULTS

Documented paroxysmal AF before the implantation was present in 24% of the patients; 80% had hypertension and 32% structural heart disease. Mean P-wave duration was 123 ± 23 ms, and 39% of the patients had IAB (32% partial, 7% advanced). After a mean follow-up of 18 ± 12 months, 33% of the patients presented AHRE. Patients with AHRE had a P-wave duration significantly longer (130 ± 24 ms vs 119 ± 21 ms; P < 0.001) and a greater prevalence of IAB (53% vs 32%; P < 0.001). In a multivariate analysis, predictors of AHRE were: IAB (odds ratio [OR] 2.1; 95% confidence interval [CI] [1.3-3.4], P < 0.001) and previous paroxysmal AF (OR 2.6; 95% CI [1.5-4.3], P < 0.001). In patients without previous AF, the presence of IAB was also a significant predictor of AHRE (OR 3.1; 95% CI [1.8-5.5], P < 0.001).

CONCLUSIONS

IAB is a strong predictor of AHRE in patients with CIED. This finding is independent of the presence of prior paroxysmal AF.

The Frequency and Risk Factors for Ischemic Stroke in Myotonic Dystrophy Type 1 Patients

INTRODUCTION

Patients with myotonic dystrophy type 1 have several cardiac abnormalities, especially myocardial conduction disorders. Few studies have investigated cerebral infarction. We investigated the frequency of both symptomatic and asymptomatic ischemic strokes in patients with myotonic dystrophy type 1.

METHODS

Patients who were diagnosed with myotonic dystrophy type 1 using genetic testing or clinical examinations at Asahikawa Medical Center were included. We retrospectively reviewed their medical history, neuroradiological imaging, electrocardiograms, and treatment. Their CHADS2 and CHA2DS2-VASc scores were calculated.

RESULT

A total of 108 patients were diagnosed with myotonic dystrophy type 1. Magnetic resonance imaging was performed in 72 and 1 patient whose results were not available was excluded. Among these, 2 patients had atrial flutter and 3 had atrial fibrillation. Regarding the CHADS2 score, 11 patients scored more than 2. Regarding the CHA2DS2-VASc score, 22 patients scored more than 2. Ischemic strokes were found in 9 patients with 1 having an atrial flutter and 4 having atrial fibrillation. All patients with stroke had CHADS2 and CHA2DS2-VASc scores higher than 2. There were significant differences between the 2 groups in atrial fibrillation (P < .001), CHADS2 score (P < .001), and CHA2DS2-VASc score (P < .001).

CONCLUSIONS

Ischemic stroke in patients with myotonic dystrophy type 1 is associated with atrial fibrillation. The CHADS2 score seems to be useful for the management of patients with myotonic dystrophy type 1. Repeated electrocardiograms are necessary for managing these patients.

Association of P wave duration index with atrial fibrillation recurrence after cryoballoon catheter ablation

AIM

We aimed to investigate the relationship between the recurrence of AF and P wave duration index (PWDI) in patients with nonvalvular PAF.

METHODS

We included 114 patients who underwent cryoballoon catheter ablation with the diagnosis of paroxysmal atrial fibrillation (PAF) (55 male, 59 female; mean age 55.5±10,9years). PWDI was calculated by dividing the Pwd by the PR interval in DII lead of 12‑lead ECG. Patients had regular follow-up visits with 12-lead ECG, medical history and clinical evaluation. 24h Holter ECG monitoring had been recorded at least 12months after ablation.

RESULTS

AF recurrence was detected in 24 patients after 1year. Patients were divided into two groups according to the AF recurrence. All parameters were compared between the two groups. Age, DM, HT frequency, ACEI-ARB use, CHA2DS2VASc and HAS-BLED score, HsCRP, LA diameter, LA volume, LA volume index, Pwd and PWDI were related to AF recurrence. In binary logistic regression analysis, PWDI (OR=1.143, p=0.001) and HT (OR=0.194, p=0.020) were found to be independent parameters for predicting AF recurrence. Every 0,01 unit increase in PWDI was found to be associated with 14.3% increase in the risk of AF recurrence. The cut-off value of PWDI obtained by ROC curve analysis was 59,9 for prediction of AF recurrence (sensitivity: 75.0%, specificity: 69.0%). The area under the curve (AUC) was 0.760 (p<0.001).

CONCLUSION

Increased PWDI may help to identify those patients in whom electrical remodeling has already occurred and who will get less benefit from cryoablation.

Incidence rates, correlates, and prognosis of electrocardiographic P-wave abnormalities - a nationwide population-based study

BACKGROUND

Scant data exist on incidence rates, correlates, and prognosis of electrocardiographic P-wave abnormalities in the general population.

METHODS

We recorded ECG and measured conventional cardiovascular risk factors in 5667 Finns who were followed up for incident atrial fibrillation (AF). We obtained repeat ECGs from 3089 individuals 11years later.

RESULTS

The incidence rates of prolonged P-wave duration, abnormal P terminal force (PTF), left P-wave axis deviation, and right P-wave axis deviation were 16.0%, 7.4%, 3.4%, and 2.2%, respectively. Older age and higher BMI were associated with incident prolonged P-wave duration and abnormal PTF (P≤0.01). Higher blood pressure was associated with incident prolonged P-wave duration and right P-wave axis deviation (P≤0.01). During follow-up, only prolonged P-wave duration predicted AF (multivariable-adjusted hazard ratio, 1.38; P=0.001).

CONCLUSIONS

Modifiable risk factors associate with P-wave abnormalities that are common and may represent intermediate steps of atrial cardiomyopathy on a pathway leading to AF.

Surface ECG interatrial block-guided treatment for stroke prevention: rationale for an attractive hypothesis

Atrial fibrillation (AF) is the most common sustained arrhythmia and is associated with stroke, cognitive impairment, and cardiovascular death. Some predisposing factors − as aging, diabetes, hypertension − induce and maintain electrophysiological and ultrastructural remodeling that usually includes fibrosis. Interatrial conduction disturbances play a crucial role in the initiation of atrial fibrosis and in its associated complications. The diagnosis of interatrial blocks (IABs) is easy to perform using the surface ECG. IAB is classified as partial when the P wave duration is ≥120 ms, and advanced if the P wave also presents a biphasic pattern in II, III and aVF. IAB is very frequent in the elderly and, particularly in the case of the advanced type, is associated with AF, AF recurrences, stroke, and dementia. The anticoagulation in elderly patients at high risk of AF without documented arrhythmias is an open issue but recent data suggest that it might have a role, particularly in elderly patients with structural heart disease, high CHA₂DS₂VASc (Congestive heart failure/left ventricular dysfunction, Hypertension, Age ≥ 75 [doubled], Diabetes, Stroke [doubled] – Vascular disease, Age 65–74, and Sex category [female]), and advanced IAB. In this debate, we discuss the association of surface ECG IAB, a marker of atrial fibrosis, with AF and stroke. We also present the rationale that justifies further studies regarding anticoagulation in some of these patients.

Bayés syndrome and acute cardioembolic ischemic stroke

Bayés syndrome is an under-recognized clinical condition characterized by advanced interatrial block. Bayés syndrome is a subclinical disease that manifests electrocardiographically as a prolonged P wave duration > 120 ms with biphasic morphology ± in the inferior leads. The clinical relevance of Bayés syndrome lies in the fact that is a clear arrhythmological syndrome and has a strong association with supraventricular arrhythmias, particularly atypical atrial flutter and atrial fibrillation. Likewise, Bayés syndrome has been recently identified as a novel risk factor for non-lacunar cardioembolic ischemic stroke and vascular dementia. Advanced interatrial block can be a risk for embolic stroke due to its known sequelae of left atrial dilation, left atrial electromechanical dysfunction or atrial tachyarrhythmia (paroxysmal or persistent atrial fibrillation), conditions predisposing to thromboembolism. Bayés syndrome may be responsible for some of the unexplained ischemic strokes and shall be considered and investigated as a possible cause for cryptogenetic stroke. In summary, Bayés syndrome is a poorly recognized cardiac rhythm disorder with important cardiologic and neurologic implications.

Anticoagulation in patients at high risk of stroke without documented atrial fibrillation. Time for a paradigm shift?

Atrial fibrillation (AF) is currently considered a risk factor for stroke. Depending on the severity of clinical factors (risk scores) a recommendation for full anticoagulation is made. Although AF is most certainly a risk factor for ischemic stroke, it is not necessarily the direct cause of it. The causality of association between AF and ischemic stroke is questioned by the reported lack of temporal relation between stroke events and AF paroxysms (or atrial high-rate episodes detected by devices). In different studies, only 2% of patients had subclinical AF > 6 minutes in duration at the time of stroke or systemic embolism. Is it time to consider AF only one more factor of endothelial disarray rather than the main contributor to stroke? In this "opinion paper" we propose to consider not only clinical variables predicting AF/stroke but also electrocardiographic markers of atrial fibrosis, as we postulate this as a strong indicator of risk of AF/stroke. We ask if it is time to change the paradigm and to consider, in some special situations, to protect patients (preventing stroke) who have no evidence of AF.

Rationale and design of the BAYES (Interatrial Block and Yearly Events) registry

The prevalence of interatrial block (IAB) is high in the elderly, particularly in those with heart disease. Despite this high prevalence-and the association of IAB with the risk of atrial fibrillation (AF), stroke, and cognitive decline-little information exists about the prognosis of older patients with IAB. P-wave duration and morphology are associated with risk of developing AF, stroke, and cognitive decline in elderly patients with structural heart disease. The aim of the Interatrial Block and Yearly Events (BAYES) registry is to assess the impact of IAB on the risk of AF and stroke during 3 years of follow-up. A series of 654 ambulatory patients age ≥70 years with heart disease from 35 centers will be included in 3 similar-size groups of patients. Group A: normal P-wave duration (<120 ms); Group B: partial IAB (P-wave duration ≥120 ms without biphasic [plus/minus] morphology in the inferior leads II, III, and aVF); and Group C: advanced IAB (P-wave duration ≥120 ms with biphasic [plus/minus] morphology in the inferior leads II, III, and aVF). Patients will be managed according to current recommendations. The 2 primary endpoints are defined as (1) AF duration >5 minutes and documented in any form of electrocardiographic recording; and (2) stroke. Results from this study might significantly improve the knowledge of IAB and its impact on the outcome of elderly patients with heart disease and could open the door to the use of anticoagulation therapy in some elderly patients with IAB.