Search

ImportanceEarly detection of atrial fibrillation (AF) may help prevent adverse cardiovascular events such as stroke. Deep learning applied to electrocardiograms (ECGs) has been successfully used for early identification of several cardiovascular diseases.ObjectiveTo determine whether deep learning models applied to outpatient ECGs in sinus rhythm can predict AF in a large and diverse patient population.Design, setting, and participantsThis prognostic study was performed on ECGs acquired from January 1, 1987, to December 31, 2022, at 6 US Veterans Affairs (VA) hospital networks and 1 large non-VA academic medical center. Participants included all outpatients with 12-lead ECGs in sinus rhythm.Main outcomes and measuresA convolutional neural network using 12-lead ECGs from 2 US VA hospital networks was trained to predict the presence of AF within 31 days of sinus rhythm ECGs. The model was tested on ECGs held out from training at the 2 VA networks as well as 4 additional VA networks and 1 large non-VA academic medical center.ResultsA total of 907 858 ECGs from patients across 6 VA sites were included in the analysis. These patients had a mean (SD) age of 62.4 (13.5) years, 6.4% were female, and 93.6% were male, with a mean (SD) CHA2DS2-VASc (congestive heart failure, hypertension, age, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age, sex category) score of 1.9 (1.6). A total of 0.2% were American Indian or Alaska Native, 2.7% were Asian, 10.7% were Black, 4.6% were Latinx, 0.7% were Native Hawaiian or Other Pacific Islander, 62.4% were White, 0.4% were of other race or ethnicity (which is not broken down into subcategories in the VA data set), and 18.4% were of unknown race or ethnicity. At the non-VA academic medical center (72 483 ECGs), the mean (SD) age was 59.5 (15.4) years and 52.5% were female, with a mean (SD) CHA2DS2-VASc score of 1.6 (1.4). A total of 0.1% were American Indian or Alaska Native, 7.9% were Asian, 9.4% were Black, 2.9% were Latinx, 0.03% were Native Hawaiian or Other Pacific Islander, 74.8% were White, 0.1% were of other race or ethnicity, and 4.7% were of unknown race or ethnicity. A deep learning model predicted the presence of AF within 31 days of a sinus rhythm ECG on held-out test ECGs at VA sites with an area under the receiver operating characteristic curve (AUROC) of 0.86 (95% CI, 0.85-0.86), accuracy of 0.78 (95% CI, 0.77-0.78), and F1 score of 0.30 (95% CI, 0.30-0.31). At the non-VA site, AUROC was 0.93 (95% CI, 0.93-0.94); accuracy, 0.87 (95% CI, 0.86-0.88); and F1 score, 0.46 (95% CI, 0.44-0.48). The model was well calibrated, with a Brier score of 0.02 across all sites. Among individuals deemed high risk by deep learning, the number needed to screen to detect a positive case of AF was 2.47 individuals for a testing sensitivity of 25% and 11.48 for 75%. Model performance was similar in patients who were Black, female, or younger than 65 years or who had CHA2DS2-VASc scores of 2 or greater.Conclusions and relevanceDeep learning of outpatient sinus rhythm ECGs predicted AF within 31 days in populations with diverse demographics and comorbidities. Similar models could be used in future AF screening efforts to reduce adverse complications associated with this disease.

BackgroundDrugs and high-risk medical devices are increasingly likely to receive Food and Drug Administration (FDA) approval through expedited pathways, which has implications for informed treatment consent (i.e., consent in clinical practice).ObjectiveTo obtain expert opinion about the clinical and ethical implications of the increasing availability of new drugs and devices approved through expedited development and regulatory review pathways.DesignQualitative study using individual semi-structured videoconference interviews.ParticipantsNational leaders in medicine, ethics, and law (n=12) with expertise in medical product regulation, payor policymaking, bioethics, physician practice, patient advocacy, public health expertise/advocacy, clinical trials, the pharmaceutical and device industry, institutional review board oversight, and real-world evidence.Main measuresPrincipal themes in 3 domains: expedited regulatory pathways, physician and patient understanding of and reliance on FDA approval, and informed treatment consent.Key resultsRespondents pointed out that more common use of expedited pathways translates to increased reliance on surrogate measures, some with uncertain clinical significance. While expedited development and review can have advantages, participants expressed worry that physicians were unaware when medical products were expedited and did not communicate about uncertainties in knowledge about new drug or device approvals effectively with patients. Many participants felt that informed treatment consent discussions about new drugs or devices should include some explanations of expedited pathways and use of surrogate measures.ConclusionsExperts identified advantages of expediting development and of FDA flexibility in applying its standards to new drugs and medical devices, but highlighted concerns that patients may not be adequately informed about the risks of shorter review times or about uncertainties in the evidence that result. There is a need to identify approaches to ensure effective clinical use of drugs and devices when approved through expedited pathways.

ObjectivesGenerating and using real-world evidence (RWE) is a pragmatic solution for evaluating health technologies. RWE is recognized by regulators, health technology assessors, clinicians, and manufacturers as a valid source of information to support their decision-making. Well-designed registries can provide RWE and become more powerful when linked with electronic health records and administrative databases in coordinated registry networks (CRNs). Our objective was to create a framework of maturity of CRNs and registries, so guiding their development and the prioritization of funding.Design setting and participantsWe invited 52 stakeholders from diverse backgrounds including patient advocacy groups, academic, clinical, industry and regulatory experts to participate on a Delphi survey. Of those invited, 42 participated in the survey to provide feedback on the maturity framework for CRNs and registries. An expert panel reviewed the responses to refine the framework until the target consensus of 80% was reached. Two rounds of the Delphi were distributed via Qualtrics online platform from July to August 2020 and from October to November 2020.Main outcome measuresConsensus on the maturity framework for CRNs and registries consisted of seven domains (unique device identification, efficient data collection, data quality, product life cycle approach, governance and sustainability, quality improvement, and patient-reported outcomes), each presented with five levels of maturity.ResultsOf 52 invited experts, 41 (79.9%) responded to round 1; all 41 responded to round 2; and consensus was reached for most domains. The expert panel resolved the disagreements and final consensus estimates ranged from 80.5% to 92.7% for seven domains.ConclusionsWe have developed a robust framework to assess the maturity of any CRN (or registry) to provide reliable RWE. This framework will promote harmonization of approaches to RWE generation across different disciplines and health systems. The domains and their levels may evolve over time as new solutions become available.

BackgroundThere is an increasing burden of cardiovascular disease, including coronary artery disease (CAD) and heart failure (HF), among women Veterans. Clinical practice guidelines recommend multiple pharmacotherapies that can reduce risk of mortality and adverse cardiovascular outcomes.ObjectiveTo determine if there are disparities in the use of guideline-directed medical therapy by gender among Veterans with incident CAD and HF.DesignRetrospective.ParticipantsVeterans (934,504; 87.8% men and 129,469; 12.2% women) returning from Operations Enduring Freedom, Iraqi Freedom, and New Dawn.Main measuresDifferences by gender in the prescription of Class 1, Level of Evidence A guideline-directed medical therapy among patients who developed incident CAD and HF at 30 days, 90 days, and 12 months after diagnosis. For CAD, medications included statins and antiplatelet therapy. For HF, medications included beta-blockers and renin-angiotensin-aldosterone system inhibitors.Key resultsOverall, women developed CAD and HF at a younger average age than men (mean 45.8 vs. 47.7 years, p

IntroductionPersonal digital devices that provide health information, such as the Apple Watch, have developed an increasing array of cardiopulmonary tracking features which have received regulatory clearance and are directly marketed to consumers. Despite their widespread and increasing use, data about the impact of personal digital device use on patient-reported outcomes and healthcare utilisation are sparse. Among a population of patients with atrial fibrillation and/or atrial flutter undergoing cardioversion, our primary aim is to determine the impact of the heart rate measurement, irregular rhythm notification, and ECG features of the Apple Watch on quality of life and healthcare utilisation.Methods and analysisWe are conducting a prospective, open-label multicentre pragmatic randomised clinical trial, leveraging a unique patient-centred health data sharing platform for enrolment and follow-up. A total of 150 patients undergoing cardioversion for atrial fibrillation or atrial flutter will be randomised 1:1 to receive the Apple Watch Series 6 or Withings Move at the time of cardioversion. The primary outcome is the difference in the Atrial Fibrillation Effect on QualiTy-of-life global score at 6 months postcardioversion. Secondary outcomes include inpatient and outpatient healthcare utilisation. Additional secondary outcomes include a comparison of the Apple Watch ECG and pulse oximeter features with gold-standard data obtained in routine clinical care settings.Ethics and disseminationThe Institutional Review Boards at Yale University, Mayo Clinic, and Duke University Health System have approved the trial protocol. This trial will provide important data to policymakers, clinicians and patients about the impact of the heart rate, irregular rhythm notification, and ECG features of widely used personal digital devices on patient quality of life and healthcare utilisation. Findings will be disseminated to study participants, at professional society meetings and in peer-reviewed journals.Trial registration numberNCT04468321.

Abstracted detailed demographic data about the age, sex, race, ethnicity, and socioeconomic position of study participants. We also assessed the presence and results of subgroup analyses to understand the safety and effectiveness of devices across trial populations. Results: Analysis of 92 pivotal studies revealed that age and sex were reported in 96.7% of the studies, while race and ethnicity were reported in 71.7% and 58.7%, respectively. However, only 7.9% of studies explicitly detailed the participation of older adults (≥ 65 years) and no studies reported patients' socioeconomic position. Subgroup analyses by sex were conducted in 70.7% of studies, with 12.3% reporting significant differences. In contrast, analyses by race and ethnicity were performed in only 12.0% of the studies, with 9.1% reporting significant differences. Conclusion: Approximately one-third of pivotal studies for high-risk cardiovascular devices approved by the FDA from 2014 to 2022 did not report the race of study participants, nearly 40% did not report ethnicity, and more than 90% did not report the participation of older adults (≥ 65 years). Subgroup analyses were infrequently conducted by age or race and ethnicity. There is a need for better trial demographic reporting and conduct of subgroup analyses in premarketing studies to ensure the safety and effectiveness of medical devices for all patients. [ABSTRACT FROM AUTHOR]

Background The current acute kidney injury (AKI) risk prediction model for patients undergoing percutaneous coronary intervention (PCI) from the American College of Cardiology (ACC) National Cardiovascular Data Registry (NCDR) employed regression techniques. This study aimed to evaluate whether models using machine learning techniques could significantly improve AKI risk prediction after PCI. Methods and findings We used the same cohort and candidate variables used to develop the current NCDR CathPCI Registry AKI model, including 947,091 patients who underwent PCI procedures between June 1, 2009, and June 30, 2011. The mean age of these patients was 64.8 years, and 32.8% were women, with a total of 69,826 (7.4%) AKI events. We replicated the current AKI model as the baseline model and compared it with a series of new models. Temporal validation was performed using data from 970,869 patients undergoing PCIs between July 1, 2016, and March 31, 2017, with a mean age of 65.7 years; 31.9% were women, and 72,954 (7.5%) had AKI events. Each model was derived by implementing one of two strategies for preprocessing candidate variables (preselecting and transforming candidate variables or using all candidate variables in their original forms), one of three variable-selection methods (stepwise backward selection, lasso regularization, or permutation-based selection), and one of two methods to model the relationship between variables and outcome (logistic regression or gradient descent boosting). The cohort was divided into different training (70%) and test (30%) sets using 100 different random splits, and the performance of the models was evaluated internally in the test sets. The best model, according to the internal evaluation, was derived by using all available candidate variables in their original form, permutation-based variable selection, and gradient descent boosting. Compared with the baseline model that uses 11 variables, the best model used 13 variables and achieved a significantly better area under the receiver operating characteristic curve (AUC) of 0.752 (95% confidence interval [CI] 0.749-0.754) versus 0.711 (95% CI 0.708-0.714), a significantly better Brier score of 0.0617 (95% CI 0.0615-0.0618) versus 0.0636 (95% CI 0.0634-0.0638), and a better calibration slope of observed versus predicted rate of 1.008 (95% CI 0.988-1.028) versus 1.036 (95% CI 1.015-1.056). The best model also had a significantly wider predictive range (25.3% versus 21.6%, p < 0.001) and was more accurate in stratifying AKI risk for patients. Evaluated on a more contemporary CathPCI cohort (July 1, 2015-March 31, 2017), the best model consistently achieved significantly better performance than the baseline model in AUC (0.785 versus 0.753), Brier score (0.0610 versus 0.0627), calibration slope (1.003 versus 1.062), and predictive range (29.4% versus 26.2%). The current study does not address implementation for risk calculation at the point of care, and potential challenges include the availability and accessibility of the predictors. Conclusions Machine learning techniques and data-driven approaches resulted in improved prediction of AKI risk after PCI. The results support the potential of these techniques for improving risk prediction models and identification of patients who may benefit from risk-mitigation strategies., Author(s): Chenxi Huang 1, Karthik Murugiah 2, Shiwani Mahajan 1, Shu-Xia Li 1, Sanket S. Dhruva 3,4, Julian S. Haimovich 5, Yongfei Wang 1, Wade L. Schulz 1,6, Jeffrey M. [...]