Predictive Accuracy of a Clinical and Genetic Risk Model for Atrial Fibrillation

Prediction of incident atrial fibrillation using deep learning, clinical models, and polygenic scores

BACKGROUND AND AIMS

Deep learning applied to electrocardiograms (ECG-AI) is an emerging approach for predicting atrial fibrillation or flutter (AF). This study introduces an ECG-AI model developed and tested at a tertiary cardiac centre, comparing its performance with clinical models and AF polygenic score (PGS).

METHODS

Electrocardiograms in sinus rhythm from the Montreal Heart Institute were analysed, excluding those from patients with pre-existing AF. The primary outcome was incident AF at 5 years. An ECG-AI model was developed by splitting patients into non-overlapping data sets: 70% for training, 10% for validation, and 20% for testing. The performance of ECG-AI, clinical models, and PGS was assessed in the test data set. The ECG-AI model was externally validated in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) hospital data set.

RESULTS

A total of 669 782 ECGs from 145 323 patients were included. Mean age was 61 ± 15 years, and 58% were male. The primary outcome was observed in 15% of patients, and the ECG-AI model showed an area under the receiver operating characteristic (AUC-ROC) curve of .78. In time-to-event analysis including the first ECG, ECG-AI inference of high risk identified 26% of the population with a 4.3-fold increased risk of incident AF (95% confidence interval: 4.02-4.57). In a subgroup analysis of 2301 patients, ECG-AI outperformed CHARGE-AF (AUC-ROC = .62) and PGS (AUC-ROC = .59). Adding PGS and CHARGE-AF to ECG-AI improved goodness of fit (likelihood ratio test P < .001), with minimal changes to the AUC-ROC (.76-.77). In the external validation cohort (mean age 59 ± 18 years, 47% male, median follow-up 1.1 year), ECG-AI model performance remained consistent (AUC-ROC = .77).

CONCLUSIONS

ECG-AI provides an accurate tool to predict new-onset AF in a tertiary cardiac centre, surpassing clinical and PGS.

Validation, bias assessment, and optimization of the UNAFIED 2-year risk prediction model for undiagnosed atrial fibrillation using national electronic health data

Background

Prediction models for atrial fibrillation (AF) may enable earlier detection and guideline-directed treatment decisions. However, model bias may lead to inaccurate predictions and unintended consequences.

Objective

The purpose of this study was to validate, assess bias, and improve generalizability of "UNAFIED-10," a 2-year, 10-variable predictive model of undiagnosed AF in a national data set (originally developed using the Indiana Network for Patient Care regional data).

Methods

UNAFIED-10 was validated and optimized using Optum de-identified electronic health record data set. AF diagnoses were recorded in the January 2018-December 2019 period (outcome period), with January 2016-December 2017 as the baseline period. Validation cohorts (patients with AF and non-AF controls, aged ≥40 years) comprised the full imbalanced and randomly sampled balanced data sets. Model performance and bias in patient subpopulations based on sex, insurance, race, and region were evaluated.

Results

Of the 6,058,657 eligible patients (mean age 60 ± 12 years), 4.1% (n = 246,975) had their first AF diagnosis within the outcome period. The validated UNAFIED-10 model achieved a higher C-statistic (0.85 [95% confidence interval 0.85-0.86] vs 0.81 [0.80-0.81]) and sensitivity (86% vs 74%) but lower specificity (66% vs 74%) than the original UNAFIED-10 model. During retraining and optimization, the variables insurance, shock, and albumin were excluded to address bias and improve generalizability. This generated an 8-variable model (UNAFIED-8) with consistent performance.

Conclusion

UNAFIED-10, developed using regional patient data, displayed consistent performance in a large national data set. UNAFIED-8 is more parsimonious and generalizable for using advanced analytics for AF detection. Future directions include validation on additional data sets.

Research Priorities for Atrial Fibrillation in Australia: A Statement From the Australian Cardiovascular Alliance Clinical Arrhythmia Theme

Atrial fibrillation (AF) is highly prevalent in the Australian community, ranking amongst the highest globally. The consequences of AF are significant. Stroke, dementia and heart failure risk are increased substantially, hospitalisations are amongst the highest for all cardiovascular causes, and Australians living with AF suffer from substantial symptoms that impact quality of life. Australian research has made a significant impact at the global level in advancing the care of patients living with AF. However, new strategies are required to reduce the growing incidence of AF and its associated healthcare demand. The Australian Cardiovascular Alliance (ACvA) has led the development of an arrhythmia clinical theme with the objective of tackling major research priorities to achieve a reduction in AF burden across Australia. In this summary, we highlight these research priorities with particular focus on the strengths of Australian research and the strategies needed to move forward in reducing incident AF and improving outcomes for those who live with this chronic condition.

Integrating Clinical, Genetic, and Electrocardiogram-Based Artificial Intelligence to Estimate Risk of Incident Atrial Fibrillation

Background

AF risk estimation is feasible using clinical factors, inherited predisposition, and artificial intelligence (AI)-enabled electrocardiogram (ECG) analysis.

Objective

To test whether integrating these distinct risk signals improves AF risk estimation.

Methods

In the UK Biobank prospective cohort study, we estimated AF risk using three models derived from external populations: the well-validated Cohorts for Aging in Heart and Aging Research in Genomic Epidemiology AF (CHARGE-AF) clinical score, a 1,113,667-variant AF polygenic risk score (PRS), and a published AI-enabled ECG-based AF risk model (ECG-AI). We estimated discrimination of 5-year incident AF using time-dependent area under the receiver operating characteristic (AUROC) and average precision (AP).

Results

Among 49,293 individuals (mean age 65±8 years, 52% women), 825 (2.4%) developed AF within 5 years. Using single models, discrimination of 5-year incident AF was higher using ECG-AI (AUROC 0.705 [95%CI 0.686-0.724]; AP 0.085 [0.071-0.11]) and CHARGE-AF (AUROC 0.785 [0.769-0.801]; AP 0.053 [0.048-0.061]) versus the PRS (AUROC 0.618, [0.598-0.639]; AP 0.038 [0.028-0.045]). The inclusion of all components ("Predict-AF3") was the best performing model (AUROC 0.817 [0.802-0.832]; AP 0.11 [0.091-0.15], p<0.01 vs CHARGE-AF+ECG-AI), followed by the two component model of CHARGE-AF+ECG-AI (AUROC 0.802 [0.786-0.818]; AP 0.098 [0.081-0.13]). Using Predict-AF3, individuals at high AF risk (i.e., 5-year predicted AF risk >2.5%) had a 5-year cumulative incidence of AF of 5.83% (5.33-6.32). At the same threshold, the 5-year cumulative incidence of AF was progressively higher according to the number of models predicting high risk (zero: 0.67% [0.51-0.84], one: 1.48% [1.28-1.69], two: 4.48% [3.99-4.98]; three: 11.06% [9.48-12.61]), and Predict-AF3 achieved favorable net reclassification improvement compared to both CHARGE-AF+ECG-AI (0.039 [0.015-0.066]) and CHARGE-AF+PRS (0.033 [0.0082-0.059]).

Conclusions

Integration of clinical, genetic, and AI-derived risk signals improves discrimination of 5-year AF risk over individual components. Models such as Predict-AF3 have substantial potential to improve prioritization of individuals for AF screening and preventive interventions.

Innovative approaches to atrial fibrillation prediction: should polygenic scores and machine learning be implemented in clinical practice?

Atrial fibrillation (AF) prediction and screening are of important clinical interest because of the potential to prevent serious adverse events. Devices capable of detecting short episodes of arrhythmia are now widely available. Although it has recently been suggested that some high-risk patients with AF detected on implantable devices may benefit from anticoagulation, long-term management remains challenging in lower-risk patients and in those with AF detected on monitors or wearable devices as the development of clinically meaningful arrhythmia burden in this group remains unknown. Identification and prediction of clinically relevant AF is therefore of unprecedented importance to the cardiologic community. Family history and underlying genetic markers are important risk factors for AF. Recent studies suggest a good predictive ability of polygenic risk scores, with a possible additive value to clinical AF prediction scores. Artificial intelligence, enabled by the exponentially increasing computing power and digital data sets, has gained traction in the past decade and is of increasing interest in AF prediction using a single or multiple lead sinus rhythm electrocardiogram. Integrating these novel approaches could help predict AF substrate severity, thereby potentially improving the effectiveness of AF screening and personalizing the management of patients presenting with conditions such as embolic stroke of undetermined source or subclinical AF. This review presents current evidence surrounding deep learning and polygenic risk scores in the prediction of incident AF and provides a futuristic outlook on possible ways of implementing these modalities into clinical practice, while considering current limitations and required areas of improvement.

Interaction of genetic risk and lifestyle on the incidence of atrial fibrillation

BACKGROUND

The relationship between combined genetic predisposition and lifestyle and the risk of incident atrial fibrillation (AF) is unclear. Therefore, we aimed to assess a possible interaction between lifestyle and genetics on AF risk.

METHODS

We included AF cases and a randomly drawn subcohort of 4040 participants from the Danish Diet, Cancer and Health cohort. Lifestyle risk factors were assessed, a score was calculated, and participants were categorised as having a poor, intermediate, or ideal lifestyle. We calculated a genetic risk score comprising 142 variants, and categorised participants into low (quintile 1), intermediate (quintiles 2-4) or high (quintile 5) genetic risk of AF.

RESULTS

3094 AF cases occurred during a median follow-up of 12.9 years. Regardless of genetic risk, incidence rates per 1000 person-years were gradually higher with worse lifestyle. For participants with high genetic risk, the incidence rates of AF per 1000 person-years were 5.0 (95% CI 3.4 to 7.3) among individuals with ideal lifestyle, 6.6 (95% CI 5.4 to 8.1) among those with intermediate lifestyle and 10.4 (95% CI 9.2 to 11.8) among participants with poor lifestyle. On an additive scale, there was a positive statistically significant interaction between genetic risk and lifestyle (relative excess risk due to interaction=0.86, 95% CI 0.68 to 1.03, p<0.001).

CONCLUSIONS

The rates of AF increased gradually with worse lifestyle within each category of genetic risk. We found a positive interaction on an additive scale between genetic risk and lifestyle, suggesting that risk factor modification is especially important in individuals with a high genetic risk of AF.

Clonal haematopoiesis of indeterminate potential and atrial fibrillation: an east Asian cohort study

BACKGROUND AND AIMS

Both clonal haematopoiesis of indeterminate potential (CHIP) and atrial fibrillation (AF) are age-related conditions. This study investigated the potential role of CHIP in the development and progression of AF.

METHODS

Deep-targeted sequencing of 24 CHIP mutations (a mean depth of coverage = 1000×) was performed in 1004 patients with AF and 3341 non-AF healthy subjects. Variant allele fraction ≥ 2.0% indicated the presence of CHIP mutations. The association between CHIP and AF was evaluated by the comparison of (i) the prevalence of CHIP mutations between AF and non-AF subjects and (ii) clinical characteristics discriminated by CHIP mutations within AF patients. Furthermore, the risk of clinical outcomes-the composite of heart failure, ischaemic stroke, or death-according to the presence of CHIP mutations in AF was investigated from the UK Biobank cohort.

RESULTS

The mean age was 67.6 ± 6.9 vs. 58.5 ± 6.5 years in AF (paroxysmal, 39.0%; persistent, 61.0%) and non-AF cohorts, respectively. CHIP mutations with a variant allele fraction of ≥2.0% were found in 237 (23.6%) AF patients (DNMT3A, 13.5%; TET2, 6.6%; and ASXL1, 1.5%) and were more prevalent than non-AF subjects [356 (10.7%); P < .001] across the age. After multivariable adjustment (age, sex, smoking, body mass index, diabetes, and hypertension), CHIP mutations were 1.4-fold higher in AF [adjusted odds ratio (OR) 1.38; 95% confidence interval 1.10-1.74, P < .01]. The ORs of CHIP mutations were the highest in the long-standing persistent AF (adjusted OR 1.50; 95% confidence interval 1.14-1.99, P = .004) followed by persistent (adjusted OR 1.44) and paroxysmal (adjusted OR 1.33) AF. In gene-specific analyses, TET2 somatic mutation presented the highest association with AF (adjusted OR 1.65; 95% confidence interval 1.05-2.60, P = .030). AF patients with CHIP mutations were older and had a higher prevalence of diabetes, a longer AF duration, a higher E/E', and a more severely enlarged left atrium than those without CHIP mutations (all P < .05). In UK Biobank analysis of 21 286 AF subjects (1297 with CHIP and 19 989 without CHIP), the CHIP mutation in AF is associated with a 1.32-fold higher risk of a composite clinical event (heart failure, ischaemic stroke, or death).

CONCLUSIONS

CHIP mutations, primarily DNMT3A or TET2, are more prevalent in patients with AF than non-AF subjects whilst their presence is associated with a more progressive nature of AF and unfavourable clinical outcomes.

Atrial fibrillation: pathophysiology, genetic and epigenetic mechanisms

Atrial fibrillation (AF) is the most common supraventricular arrhythmia affecting up to 1% of the general population. Its prevalence dramatically increases with age and could reach up to ∼10% in the elderly. The management of AF is a complex issue that is object of extensive ongoing basic and clinical research, it depends on its genetic and epigenetic causes, and it varies considerably geographically and also according to the ethnicity. Mechanistically, over the last decade, Genome Wide Association Studies have uncovered over 100 genetic loci associated with AF, and have shown that European ancestry is associated with elevated risk of AF. These AF-associated loci revolve around different types of disturbances, including inflammation, electrical abnormalities, and structural remodeling. Moreover, the discovery of epigenetic regulatory mechanisms, involving non-coding RNAs, DNA methylation and histone modification, has allowed unravelling what modifications reshape the processes leading to arrhythmias. Our review provides a current state of the field regarding the identification and functional characterization of AF-related genetic and epigenetic regulatory networks, including ethnic differences. We discuss clear and emerging connections between genetic regulation and pathophysiological mechanisms of AF.

Advances in basic and translational research in atrial fibrillation

Atrial fibrillation (AF) is a very common cardiac arrhythmia with an estimated prevalence of 33.5 million patients globally. It is associated with an increased risk of death, stroke and peripheral embolism. Although genetic studies have identified a growing number of genes associated with AF, the definitive impact of these genetic findings is yet to be established. Several mechanisms, including electrical, structural and neural remodelling of atrial tissue, have been proposed to contribute to the development of AF. Despite over a century of exploration, the molecular and cellular mechanisms underlying AF have not been fully established. Current antiarrhythmic drugs are associated with a significant rate of adverse events and management of AF using ablation is not optimal, especially in cases of persistent AF. This review discusses recent advances in our understanding and management of AF, including new concepts of epidemiology, genetics and pathophysiological mechanisms. We review the current status of antiarrhythmic drug therapy for AF, new potential agents, as well as mechanism-based AF ablation. This article is part of the theme issue 'The heartbeat: its molecular basis and physiological mechanisms'.

Epidemiology and modifiable risk factors for atrial fibrillation

The global prevalence of atrial fibrillation (AF) has increased substantially over the past three decades and is currently approximately 60 million cases. Incident AF and its clinical consequences are largely the result of risk factors that can be modified by lifestyle changes. In this Review, we provide evidence that the lifetime risk of AF is modified not only by sex and race but also through the clinical risk factor and comorbidity burden of individual patients. We begin by summarizing the epidemiology of AF, focusing on non-modifiable and modifiable risk factors, as well as targets and strategies for the primary prevention of AF. Furthermore, we evaluate the role of modifiable risk factors in the secondary prevention of AF as well as the potential effects of risk factor interventions on the frequency and severity of subsequent AF episodes. We end the Review by proposing strategies that require evaluation as well as global policy changes that are needed for the prevention of incident AF and the management of recurrent episodes in patients already affected by AF.

Natural Language Processing to Improve Prediction of Incident Atrial Fibrillation Using Electronic Health Records

Background Models predicting atrial fibrillation (AF) risk, such as Cohorts for Heart and Aging Research in Genomic Epidemiology AF (CHARGE-AF), have not performed as well in electronic health records. Natural language processing (NLP) may improve models by using narrative electronic health record text. Methods and Results From a primary care network, we included patients aged ≥65 years with visits between 2003 and 2013 in development (n=32 960) and internal validation cohorts (n=13 992). An external validation cohort from a separate network from 2015 to 2020 included 39 051 patients. Model features were defined using electronic health record codified data and narrative data with NLP. We developed 2 models to predict 5-year AF incidence using (1) codified+NLP data and (2) codified data only and evaluated model performance. The analysis included 2839 incident AF cases in the development cohort and 1057 and 2226 cases in internal and external validation cohorts, respectively. The C-statistic was greater (P<0.001) in codified+NLP model (0.744 [95% CI, 0.735-0.753]) compared with codified-only (0.730 [95% CI, 0.720-0.739]) in the development cohort. In internal validation, the C-statistic of codified+NLP was modestly higher (0.735 [95% CI, 0.720-0.749]) compared with codified-only (0.729 [95% CI, 0.715-0.744]; P=0.06) and CHARGE-AF (0.717 [95% CI, 0.703-0.731]; P=0.002). Codified+NLP and codified-only were well calibrated, whereas CHARGE-AF underestimated AF risk. In external validation, the C-statistic of codified+NLP (0.750 [95% CI, 0.740-0.760]) remained higher (P<0.001) than codified-only (0.738 [95% CI, 0.727-0.748]) and CHARGE-AF (0.735 [95% CI, 0.725-0.746]). Conclusions Estimation of 5-year risk of AF can be modestly improved using NLP to incorporate narrative electronic health record data.

Prediction performance and fairness heterogeneity in cardiovascular risk models

Prediction models are commonly used to estimate risk for cardiovascular diseases, to inform diagnosis and management. However, performance may vary substantially across relevant subgroups of the population. Here we investigated heterogeneity of accuracy and fairness metrics across a variety of subgroups for risk prediction of two common diseases: atrial fibrillation (AF) and atherosclerotic cardiovascular disease (ASCVD). We calculated the Cohorts for Heart and Aging in Genomic Epidemiology Atrial Fibrillation (CHARGE-AF) score for AF and the Pooled Cohort Equations (PCE) score for ASCVD in three large datasets: Explorys Life Sciences Dataset (Explorys, n = 21,809,334), Mass General Brigham (MGB, n = 520,868), and the UK Biobank (UKBB, n = 502,521). Our results demonstrate important performance heterogeneity across subpopulations defined by age, sex, and presence of preexisting disease, with fairly consistent patterns across both scores. For example, using CHARGE-AF, discrimination declined with increasing age, with a concordance index of 0.72 [95% CI 0.72-0.73] for the youngest (45-54 years) subgroup to 0.57 [0.56-0.58] for the oldest (85-90 years) subgroup in Explorys. Even though sex is not included in CHARGE-AF, the statistical parity difference (i.e., likelihood of being classified as high risk) was considerable between males and females within the 65-74 years subgroup with a value of - 0.33 [95% CI - 0.33 to - 0.33]. We also observed weak discrimination (i.e., < 0.7) and suboptimal calibration (i.e., calibration slope outside of 0.7-1.3) in large subsets of the population; for example, all individuals aged 75 years or older in Explorys (17.4%). Our findings highlight the need to characterize and quantify the behavior of clinical risk models within specific subpopulations so they can be used appropriately to facilitate more accurate, consistent, and equitable assessment of disease risk.

ECG-Based Deep Learning and Clinical Risk Factors to Predict Atrial Fibrillation

BACKGROUND

Artificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it remains unclear whether AI provides meaningful and generalizable improvement in predictive accuracy beyond clinical risk factors for AF.

METHODS

We trained a convolutional neural network (ECG-AI) to infer 5-year incident AF risk using 12-lead ECGs in patients receiving longitudinal primary care at Massachusetts General Hospital (MGH). We then fit 3 Cox proportional hazards models, composed of ECG-AI 5-year AF probability, CHARGE-AF clinical risk score (Cohorts for Heart and Aging in Genomic Epidemiology-Atrial Fibrillation), and terms for both ECG-AI and CHARGE-AF (CH-AI), respectively. We assessed model performance by calculating discrimination (area under the receiver operating characteristic curve) and calibration in an internal test set and 2 external test sets (Brigham and Women's Hospital [BWH] and UK Biobank). Models were recalibrated to estimate 2-year AF risk in the UK Biobank given limited available follow-up. We used saliency mapping to identify ECG features most influential on ECG-AI risk predictions and assessed correlation between ECG-AI and CHARGE-AF linear predictors.

RESULTS

The training set comprised 45 770 individuals (age 55±17 years, 53% women, 2171 AF events) and the test sets comprised 83 162 individuals (age 59±13 years, 56% women, 2424 AF events). Area under the receiver operating characteristic curve was comparable using CHARGE-AF (MGH, 0.802 [95% CI, 0.767-0.836]; BWH, 0.752 [95% CI, 0.741-0.763]; UK Biobank, 0.732 [95% CI, 0.704-0.759]) and ECG-AI (MGH, 0.823 [95% CI, 0.790-0.856]; BWH, 0.747 [95% CI, 0.736-0.759]; UK Biobank, 0.705 [95% CI, 0.673-0.737]). Area under the receiver operating characteristic curve was highest using CH-AI (MGH, 0.838 [95% CI, 0.807 to 0.869]; BWH, 0.777 [95% CI, 0.766 to 0.788]; UK Biobank, 0.746 [95% CI, 0.716 to 0.776]). Calibration error was low using ECG-AI (MGH, 0.0212; BWH, 0.0129; UK Biobank, 0.0035) and CH-AI (MGH, 0.012; BWH, 0.0108; UK Biobank, 0.0001). In saliency analyses, the ECG P-wave had the greatest influence on AI model predictions. ECG-AI and CHARGE-AF linear predictors were correlated (Pearson r: MGH, 0.61; BWH, 0.66; UK Biobank, 0.41).

CONCLUSIONS

AI-based analysis of 12-lead ECGs has similar predictive usefulness to a clinical risk factor model for incident AF and the approaches are complementary. ECG-AI may enable efficient quantification of future AF risk.

Somatic Mutations in Cardiovascular Disease

Advances in population-scale genomic sequencing have greatly expanded the understanding of the inherited basis of cardiovascular disease (CVD). Reanalysis of these genomic datasets identified an unexpected risk factor for CVD, somatically acquired DNA mutations. In this review, we provide an overview of somatic mutations and their contributions to CVD. We focus on the most common and well-described manifestation, clonal hematopoiesis of indeterminate potential. We also review the currently available data regarding how somatic mutations lead to tissue mosaicism in various forms of CVD, including atrial fibrillation and aortic aneurism associated with Marfan Syndrome. Finally, we highlight future research directions given current knowledge gaps and consider how technological advances will enhance the discovery of somatic mutations in CVD and management of patients with somatic mutations.