Identification of Incident Atrial Fibrillation From Electronic Medical Records

Validity of International Classification of Diseases, Tenth Revision, codes for atrial fibrillation/flutter in critically ill patients with sepsis

BACKGROUND

Atrial fibrillation (AF) and atrial flutter (AFL) are frequently seen in critically ill sepsis patients and are associated with poor outcomes. There is a need for further research, however, studies are limited due to challenges in identifying patient cohorts. Administrative data using the International Classification of Diseases, Tenth Revision (ICD-10) are routinely used for identifying disease cohorts in large datasets. However, the validity of ICD-10 for AF/AFL remains unexplored in these populations.

METHODS

This validation study included 6554 adults with sepsis and septic shock admitted to the intensive care unit. We sought to determine whether ICD-10 coding could accurately identify patients with and without AF/AFL compared to manual chart review. We also evaluated whether the date of ICD-10 code entry could distinguish prevalent from incident AF/AFL, presuming codes dated during the index admission to be incident AF/AFL. A manual chart review was performed on 400 randomly selected patients for confirmation of AF/AFL, and validity was measured using sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

RESULTS

Among the 400 randomly selected patients, 293 lacked ICD-10 codes for AF/AFL. The manual chart review confirmed the absence of AF/AFL in 286 patients (NPV 97.3%, specificity 99.7%). Among the 107 patients with ICD-10 codes for AF/AFL, 106 were confirmed to have AF/AFL by manual chart review (PPV 99.1%, sensitivity 93.0%). Out of the 114 patients with confirmed AF/AFL, 44 had ICD-10 codes dated during the index admission. All 44 were confirmed to have AF/AFL, however, 18 patients had prior documentation of AF/AFL (incident AF/AFL: PPV 59.1%). Specificity for incident (95.1%) and prevalent (99.7%) AF/AFL were high; however, sensitivity was 76.5% and 77.5%, respectively.

DISCUSSION/CONCLUSION

ICD-10 codes perform well in identifying clinical AF/AFL in critically ill sepsis. However, their temporal specificity in distinguishing incidents from prevalent AF/AFL is limited.

From sleep patterns to heart rhythm: Predicting atrial fibrillation from overnight polysomnograms

BACKGROUND

Atrial fibrillation (AF) is often asymptomatic and thus under-observed. Given the high risks of stroke and heart failure among patients with AF, early prediction and effective management are crucial. Given the prevalence of obstructive sleep apnea among AF patients, electrocardiogram (ECG) analysis from polysomnography (PSG) offers a unique opportunity for early AF prediction. Our aim is to identify individuals at high risk of AF development from single‑lead ECGs during standard PSG.

METHODS

We analyzed 18,782 single‑lead ECG recordings from 13,609 subjects undergoing PSG at the Massachusetts General Hospital sleep laboratory. AF presence was identified using ICD-9/10 codes. The dataset included 15,913 recordings without AF history and 2054 recordings from patients diagnosed with AF between one month to fifteen years post-PSG. Data were partitioned into training, validation, and test cohorts ensuring that individual patients remained exclusive to each cohort. The test set was held out during the training process. We employed two different methods for feature extraction to build a final model for AF prediction: Extraction of hand-crafted ECG features and a deep learning method. For extraction of ECG-hand-crafted features, recordings were split into 30-s windows, and those with a signal quality index (SQI) below 0.95 were discarded. From each remaining window, 150 features were extracted from the time, frequency, time-frequency domains, and phase-space reconstructions of the ECG. A compilation of 12 statistical features summarized these window-specific features per recording, resulting in 1800 features (12 × 150). A pre-trained deep neural network from the PhysioNet Challenge 2021 was updated using transfer learning to discriminate recordings with and without AF. The model processed PSG ECGs in 16-s windows to generate AF probabilities, from which 13 statistical features were extracted. Combining 1800 features from feature extraction with 13 from the deep learning model, we performed a feature selection and subsequently trained a shallow neural network to predict future AF and evaluated its performance on the test cohort.

RESULTS

On the test set, our model exhibited sensitivity, specificity, and precision of 0.67, 0.81, and 0.3, respectively, for AF prediction. Survival analysis revealed a hazard ratio of 8.36 (p-value: 1.93 × 10-52) for AF outcomes using the log-rank test.

CONCLUSIONS

Our proposed ECG analysis method, utilizing overnight PSG data, shows promise in AF prediction despite modest precision, suggesting false positives. This approach could enable low-cost screening and proactive treatment for high-risk patients. Refinements, including additional physiological parameters, may reduce false positives, enhancing clinical utility and accuracy.

From Sleep Patterns to Heart Rhythms: Predicting Atrial Fibrillation from Overnight Polysomnograms

Background

Atrial fibrillation (AF) is often asymptomatic and thus under-observed. Given the high risks of stroke and heart failure among patients with AF, early prediction and effective management are crucial. Importantly, obstructive sleep apnea is highly prevalent among AF patients (60-90%); therefore, electrocardiogram (ECG) analysis from polysomnography (PSG), a standard diagnostic tool for subjects with suspected sleep apnea, presents a unique opportunity for the early prediction of AF. Our goal is to identify individuals at a high risk of developing AF in the future from a single-lead ECG recorded during standard PSGs.

Methods

We analyzed 18,782 single-lead ECG recordings from 13,609 subjects at Massachusetts General Hospital, identifying AF presence using ICD-9/10 codes in medical records. Our dataset comprises 15,913 recordings without a medical record for AF and 2,056 recordings from patients who were first diagnosed with AF between 1 day to 15 years after the PSG recording. The PSG data were partitioned into training, validation, and test cohorts. In the first phase, a signal quality index (SQI) was calculated in 30-second windows and those with SQI < 0.95 were removed. From each remaining window, 150 hand-crafted features were extracted from time, frequency, time-frequency domains, and phase-space reconstructions of the ECG. A compilation of 12 statistical features summarized these window-specific features per recording, resulting in 1,800 features. We then updated a pre-trained deep neural network and data from the PhysioNet Challenge 2021 using transfer-learning to discriminate between recordings with and without AF using the same Challenge data. The model was applied to the PSG ECGs in 16-second windows to generate the probability of AF for each window. From the resultant probability sequence, 13 statistical features were extracted. Subsequently, we trained a shallow neural network to predict future AF using the extracted ECG and probability features.

Results

On the test set, our model demonstrated a sensitivity of 0.67, specificity of 0.81, and precision of 0.3 for predicting AF. Further, survival analysis for AF outcomes, using the log-rank test, revealed a hazard ratio of 8.36 (p-value of 1.93 × 10 -52 ).

Conclusions

Our proposed ECG analysis method, utilizing overnight PSG data, shows promise in AF prediction despite a modest precision indicating the presence of false positive cases. This approach could potentially enable low-cost screening and proactive treatment for high-risk patients. Ongoing refinement, such as integrating additional physiological parameters could significantly reduce false positives, enhancing its clinical utility and accuracy.

Observational study protocol for an arrhythmia notification feature

INTRODUCTION

Screening for atrial fibrillation (AF) in the general population may help identify individuals at risk, enabling further assessment of risk factors and institution of appropriate treatment. Algorithms deployed on wearable technologies such as smartwatches and fitness bands may be trained to screen for such arrhythmias. However, their performance needs to be assessed for safety and accuracy prior to wide-scale implementation.

METHODS AND ANALYSIS

This study will assess the ability of the WHOOP strap to detect AF using its WHOOP Arrhythmia Notification Feature (WARN) algorithm in an enriched cohort with a 2:1 distribution of previously diagnosed AF (persistent and paroxysmal) and healthy controls. Recruited participants will collect data for 7 days with the WHOOP wrist-strap and BioTel ePatch (electrocardiography gold-standard). Primary outcome will be participant level sensitivity and specificity of the WARN algorithm in detecting AF in analysable windows compared with the ECG gold-standard. Similar analyses will be performed on an available epoch-level basis as well as comparison of these findings in important subgroups.

ETHICS AND DISSEMINATION

The study was approved by the ethics board at the study site. Participants will be enrolled after signing an online informed consent document. Updates will be shared via clinicaltrials.gov. The data obtained from the conclusion of this study will be presented in national and international conferences with publication in clinical research journals.

TRIAL REGISTRATION NUMBER

NCT05809362.

Glucagon-like peptide-1 receptor agonists and the risk of atrial fibrillation in adults with diabetes: a real-world study

BACKGROUND

Glucagon-like peptide-1 receptor agonists (GLP-1RA) have cardiovascular benefits in type 2 diabetes, but none of the cardiovascular trials studied atrial fibrillation/atrial flutter (AF) as a primary endpoint. Data from post-marketing surveillance studies remains sparse.

OBJECTIVE

To examine the real-world risk of AF comparing GLP-1RA with other non-insulin glucose-lowering agents.

DESIGN

Cohort study using de-identified electronic health record data from the Optum Labs Data Warehouse.

PARTICIPANTS

Adult patients with diabetes who were newly prescribed add-on non-insulin glucose-lowering agents and were on metformin between 2005-2020.

EXPOSURES

New users of GLP-1RA were separately compared with new users of dipeptidyl peptidase-4 inhibitors (DPP4i) and sodium-glucose cotransporter 2 inhibitors (SGLT2i), using 1:1 propensity score matching to adjust for differences in patient characteristics.

MAIN MEASURES

The primary outcome was incident AF, defined and captured by diagnosis code for AF. Incidence rate difference (IRD) and hazard ratio (HR) were estimated in the matched cohorts.

KEY RESULTS

In the matched cohort of 14,566 pairs of GLP-1RA and DPP4i followed for a median of 3.8 years, GLP-1RA use was associated with a lower risk of AF (IRD, -1.0; 95% CI, -1.8 to -0.2 per 1000 person-years; HR, 0.82; 95% CI, 0.70 to 0.96). In the matched cohort of 9,424 pairs of patients on GLP-1RA and SGLT2i with a median follow-up of 2.9 years, there was no difference in the risk for AF (IRD, 0.4; 95% CI -0.7 to 1.5 per 1000 person-years; HR, 1.12; 95% CI, 0.89 to 1.42).

CONCLUSIONS

In this real-word study, GLP-1RA was associated with a lower risk of AF compared with DPP4i, but no difference compared with SGLT2i, suggesting that cardiovascular benefits of GLP-1RA use may extend to prevention for AF in patients with diabetes. Our findings call for future randomized controlled trials to focus on the effects of GLP-1RA on AF prevention.

Assessing ascertainment bias in atrial fibrillation across US minority groups

The aim of this study is to define atrial fibrillation (AF) prevalence and incidence rates across minority groups in the United States (US), to aid in diversity enrollment target setting for randomized controlled trials. In AF, US minority groups have lower clinically detected prevalence compared to the non-Hispanic or Latino White (NHW) population. We assess the impact of ascertainment bias on AF prevalence estimates. We analyzed data from adults in Optum's de-identified Clinformatics® Data Mart Database from 2017-2020 in a cohort study. Presence of AF at baseline was identified from inpatient and/or outpatient encounters claims using validated ICD-10-CM diagnosis algorithms. AF incidence and prevalence rates were determined both in the overall population, as well as in a population with a recent stroke event, where monitoring for AF is assumed. Differences in prevalence across cohorts were assessed to determine if ascertainment bias contributes to the variation in AF prevalence across US minority groups. The period prevalence was respectively 4.9%, 3.2%, 2.1% and 5.9% in the Black or African American, Asian, Hispanic or Latino, and NHW population. In patients with recent ischemic stroke, the proportion with AF was 32.2%, 24.3%, 25%, and 24.5%, respectively. The prevalence of AF among the stroke population was approximately 7 to 10 times higher than the prevalence among the overall population for the Asian and Hispanic or Latino population, compared to approximately 5 times higher for NHW patients. The relative AF prevalence difference of the Asian and Hispanic or Latino population with the NHW population narrowed from respectively, -46% and -65%, to -22% and -24%. The study findings align with previous observational studies, revealing lower incidence and prevalence rates of AF in US minority groups. Prevalence estimates of the adult population, when routine clinical practice is assumed, exhibit higher prevalence differences compared to settings in which monitoring for AF is assumed, particularly among Asian and Hispanic or Latino subgroups.

Progression of stroke risk in patients aged <65 years diagnosed with atrial fibrillation: a cohort study in general practice

BACKGROUND

As a result of new technologies, atrial fibrillation (AF) is more likely to be diagnosed in people aged <65 years.

AIM

To investigate the risk of someone diagnosed with AF aged <65 years developing an indication for anticoagulation before they reach 65 years.

DESIGN AND SETTING

Population-based cohort study of patients from English practices using the Clinical Practice Research Datalink, a primary care database of electronic medical records.

METHOD

The study included patients aged <65 years newly diagnosed with AF. The CHA2DS2-VASc score was derived at time of diagnosis based on patients' medical records. Patients not eligible for anticoagulation were followed up until they became eligible or turned 65 years old. The primary outcome of interest was development of a risk factor for stroke in AF.

RESULTS

Among 18 178 patients aged <65 years diagnosed with AF, 9188 (50.5%) were eligible for anticoagulation at the time of diagnosis. Among the 8990 patients not eligible for anticoagulation, 1688 (18.8%) developed a risk factor during follow-up before reaching 65 years of age or leaving the cohort for other reasons, at a rate of 6.1 per 100 patient-years. Hypertension and heart failure were the most common risk factors to occur, with rates of 2.65 (95% CI = 2.47 to 2.84) and 1.58 (95% CI = 1.45 to 1.72) per 100 patient-years, respectively. The rate of new diabetes was 0.95 (95% CI = 0.85 to 1.06) per 100 patient-years.

CONCLUSION

People aged <65 years with AF are at higher risk of developing hypertension, heart failure, and diabetes than the general population, so may warrant regular review to identify new occurrence of such risk factors.

An economic evaluation of first-line cryoballoon ablation vs antiarrhythmic drug therapy for the treatment of paroxysmal atrial fibrillation from a U.S. Medicare perspective

Background

Three recent randomized controlled trials have demonstrated that, as an initial rhythm control strategy, first-line cryoballoon ablation (cryoablation) reduces atrial arrhythmia recurrence compared with antiarrhythmic drugs (AADs) in patients with symptomatic paroxysmal atrial fibrillation (PAF).

Objective

The study sought to evaluate the cost-effectiveness of first-line cryoablation compared with first-line AADs for treating symptomatic PAF from a U.S. Medicare payer perspective.

Methods

Individual patient-level data from 703 participants with PAF enrolled into the Cryo-FIRST (NCT01803438), STOP AF First (NCT03118518), and EARLY-AF (NCT02825979) trials were used to derive parameters for the cost-effectiveness model. The cost-effectiveness model used a hybrid decision tree and Markov structure. The decision tree had a 1-year time horizon and was used to inform the initial health state allocation in the first cycle of the Markov model. The Markov model used a 40-year time horizon (3-month cycle length). Health benefits were expressed in quality-adjusted life years (QALYs). Costs and benefits were discounted at 3% per year.

Results

Cryoablation was estimated to yield higher QALYs (+0.17) and higher costs (+$4274) per patient over a 40-year time horizon than AADs. Ultimately, this produced an average incremental cost-effectiveness ratio of $24,637 per QALY gained. Independent of initial treatment, individuals were expected to receive ∼1.2 ablations over a lifetime. There was a 45% relative reduction in time spent in atrial fibrillation health states for those initially treated with cryoablation compared with AADs.

Conclusion

Initial rhythm control with first-line cryoballoon ablation is highly cost-effective compared with first-line AADs from a U.S. Medicare payer perspective.

Risk of Cancer After Diagnosis of Cardiovascular Disease

Background

Cardiovascular disease (CVD) and cancer share several risk factors. Although preclinical models show that various types of CVD can accelerate cancer progression, clinical studies have not determined the impact of atherosclerosis on cancer risk.

Objectives

The objective of this study was to determine whether CVD, especially atherosclerotic CVD, is independently associated with incident cancer.

Methods

Using IBM MarketScan claims data from over 130 million individuals, 27 million cancer-free subjects with a minimum of 36 months of follow-up data were identified. Individuals were stratified by presence or absence of CVD, time-varying analysis with multivariable adjustment for cardiovascular risk factors was performed, and cumulative risk of cancer was calculated. Additional analyses were performed according to CVD type (atherosclerotic vs nonatherosclerotic) and cancer subtype.

Results

Among 27,195,088 individuals, those with CVD were 13% more likely to develop cancer than those without CVD (HR: 1.13; 95% CI: 1.12-1.13). Results were more pronounced for individuals with atherosclerotic CVD (aCVD), who had a higher risk of cancer than those without CVD (HR: 1.20; 95% CI: 1.19-1.21). aCVD also conferred a higher risk of cancer compared with those with nonatherosclerotic CVD (HR: 1.11; 95% CI: 1.11-1.12). Cancer subtype analyses showed specific associations of aCVD with several malignancies, including lung, bladder, liver, colon, and other hematologic cancers.

Conclusions

Individuals with CVD have an increased risk of developing cancer compared with those without CVD. This association may be driven in part by the relationship of atherosclerosis with specific cancer subtypes, which persists after controlling for conventional risk factors.

Delayed vs early cardioversion in patients with paroxysmal atrial fibrillation: a population-based study (2015-2020)

Aim: There is limited data on clinical outcomes of delayed cardioversion (DCV) compared with early cardioversion (ECV) in paroxysmal atrial fibrillation (AF) patients. Methods: We utilized data from National Inpatient Sample (2015-2020) and propensity-score matched analysis to determine adjusted odds ratio (aOR) of major clinical outcomes, including 17,879 AF cases: 9725 and 8154 underwent ECV and DCV, respectively. Results: Compared with ECV, DCV was associated with higher odds of acute heart failure (AHF; aOR 1.79 [1.67-1.92]; p < 0.01), median length of stay (4 vs 2 days; p < 0.01) and cost of hospitalization ($33,410 vs $21,738; p < 0.01) with no significant difference in inpatient mortality and other cardiovascular and neurological outcomes. Conclusion: Compared with ECV, DCV was associated with more AHF and resource utilization.

Trends in Oral Anticoagulant Use Among 436 864 Patients With Atrial Fibrillation in Community Practice, 2011 to 2020

Background

Among patients with nonvalvular atrial fibrillation (AF) and an elevated stroke risk, guidelines recommend direct oral anticoagulants (DOACs) over warfarin for stroke prevention. Changes in DOAC use over the past decade have not been well described.

Methods and Results

We evaluated trends in use of DOACs and warfarin from 2011 to 2020 among adults with AF and a CHA 2 DS 2 ‐VASc score ≥2 based on electronic health record data from 88 health systems in the United States contributing to Cerner Real World Data. The use of DOACs and warfarin was described over time, by age, sex, race, and ethnicity, and at the health‐system level. We identified 436 864 patients with AF at risk for stroke (median age, 78 years; 52.1% men). From 2011 to 2020, overall anticoagulation rates increased from 56.3% to 64.7%, as DOAC use increased steadily (from 4.7% to 47.9%), while warfarin use declined (from 52.4% to 17.7%). DOAC uptake was similar across age, sex, and race and ethnicity groups but varied by health system. In 2020, the median health‐system‐level proportion of patients with AF on a DOAC was 49% (interquartile range, 40%–54%).

Conclusions

Over the past decade, anticoagulation rates for patients with AF have increased modestly as DOACs largely replaced warfarin, though significant gaps remain: One in 3 high‐risk patients with AF is not on any anticoagulant. While DOAC adoption was generally consistent across major demographic groups, use between health systems remained highly variable, suggesting that provider and system factors influence DOAC uptake use more than patient‐level factors.

Performance of EHR classifiers for patient eligibility in a clinical trial of precision screening

BACKGROUND

Validated computable eligibility criteria use real-world data and facilitate the conduct of clinical trials. The Genomic Medicine at VA (GenoVA) Study is a pragmatic trial of polygenic risk score testing enrolling patients without known diagnoses of 6 common diseases: atrial fibrillation, coronary artery disease, type 2 diabetes, breast cancer, colorectal cancer, and prostate cancer. We describe the validation of computable disease classifiers as eligibility criteria and their performance in the first 16 months of trial enrollment.

METHODS

We identified well-performing published computable classifiers for the 6 target diseases and validated these in the target population using blinded physician review. If needed, classifiers were refined and then underwent a subsequent round of blinded review until true positive and true negative rates ≥80% were achieved. The optimized classifiers were then implemented as pre-screening exclusion criteria; telephone screens enabled an assessment of their real-world negative predictive value (NPV-RW).

RESULTS

Published classifiers for type 2 diabetes and breast and prostate cancer achieved desired performance in blinded chart review without modification; the classifier for atrial fibrillation required two rounds of refinement before achieving desired performance. Among the 1077 potential participants screened in the first 16 months of enrollment, NPV-RW of the classifiers ranged from 98.4% for coronary artery disease to 99.9% for colorectal cancer. Performance did not differ by gender or race/ethnicity.

CONCLUSIONS

Computable disease classifiers can serve as efficient and accurate pre-screening classifiers for clinical trials, although performance will depend on the trial objectives and diseases under study.

Association Between Sex-Specific Risk Factors and Risk of New-Onset Atrial Fibrillation Among Women

IMPORTANCE

Atrial fibrillation (AF) is the most common cardiac arrhythmia worldwide, with different epidemiological and pathophysiological processes for women vs men and a poorer prognosis for women. Further investigation of sex-specific risk factors associated with AF development in women is warranted.

OBJECTIVE

To investigate the linear and potential nonlinear associations between sex-specific risk factors and the risk of new-onset AF in women.

DESIGN, SETTING, AND PARTICIPANTS

This population-based cohort study obtained data from the 2006 to 2010 UK Biobank study, a cohort of more than 500 000 participants aged 40 to 69 years. Participants were women without AF and history of hysterectomy and/or bilateral oophorectomy at baseline. Median follow-up period for AF onset was 11.6 years, and follow-up ended on October 3, 2020.

EXPOSURES

Self-reported, sex-specific risk factors, including age at menarche, history of irregular menstrual cycle, menopause status, age at menopause, years after menopause, age at first live birth, years after last birth, history of spontaneous miscarriages, history of stillbirths, number of live births, and total reproductive years.

MAIN OUTCOMES AND MEASURES

The primary outcome was new-onset AF, which was defined by the use of International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code I48.

RESULTS

A total of 235 191 women (mean [SD] age, 55.7 [8.1] years) were included in the present study. During follow-up, 4629 (2.0%) women experienced new-onset AF. In multivariable-adjusted models, history of irregular menstrual cycle was associated with higher AF risk (hazard ratio [HR], 1.34; 95% CI, 1.01-1.79). Both early menarche (age 7-11 years; HR, 1.10 [95% CI, 1.00-1.21]) and late menarche (age 13-18 years; HR, 1.08 [95% CI, 1.00-1.17]) were associated with AF incidence. Early menopause (age 35-44 years; HR, 1.24 [95% CI, 1.10-1.39]) and delayed menopause (age ≥60 years; HR, 1.34 [95% CI, 1.10-1.78]) were associated with higher risk of AF. Compared with women with 1 to 2 live births, those with 0 live births (HR, 1.13; 95% CI, 1.04-1.24) or 7 or more live births (HR, 1.67; 95% CI, 1.03-2.70) both had significantly higher AF risk.

CONCLUSIONS AND RELEVANCE

Results of this study suggest that irregular menstrual cycles, nulliparity, and multiparity were associated with higher risk of new-onset AF among women. The results highlight the importance of taking into account the reproductive history of women in devising screening strategies for AF prevention.