Abstract 18382: Clinical and Genetic Associations of Deep-Learning Estimated Peak Oxygen Consumption From the Resting 12-Lead Electrocardiogram

Circulation, Volume 148, Issue Suppl_1, Page A18382-A18382, November 6, 2023. Introduction:Oxygen consumption at peak exercise (VO2peak) is the gold standard for cardiorespiratory fitness, but prior genome-wide association studies (GWAS) have been limited by the availability of cardiopulmonary exercise testing (CPET). Recent deep learning methods to estimate VO2peakfrom the resting 12-lead electrocardiogram (ECG) may enhance genetic studies of cardiorespiratory fitness.Methods:We applied a validated deep learning model (Deep ECG-VO2) to estimate VO2peakamong UK Biobank participants with a 12-lead ECG. We assessed for associations between estimated VO2peakand incident hypertension, diabetes, and atrial fibrillation (AF) using Cox proportional hazards models adjusted for age and sex, and plotted cumulative risk of each outcome stratified by tertile of estimated VO2peak. We then performed a multi-ancestry GWAS of estimated VO2peakusing BOLT-LMM, adjusted for age, sex, array, and the first five principal components of ancestry. Candidate genes were prioritized based on proximity to the lead variant.Results:We applied Deep ECG-VO2 to estimate VO2peakusing the resting 12-lead ECG of 40,801 UK Biobank participants (age 65±8, 52% women). Greater estimated VO2peakwas associated with lower risks of hypertension (hazard ratio per 1-standard deviation 0.76, 95% CI 0.70-0.83), diabetes (0.60, 95% CI 0.53-0.67) and AF (0.82, 95% CI 0.74-0.90). Cumulative risk of each outcome was higher with decreasing estimated VO2peak(Figure). In a GWAS of estimated VO2peakwithin 39,716 participants with genetic data (age 65±8, 52% women, 90% European), we identified 10 novel genome-wide significant loci, including variants near genes involved in cardiac structure (CCDC141/TTN, BAG3), cardiac conduction (SCN5A), and adiposity (FTO).Conclusions:Leveraging artificial intelligence-enabled estimation of VO2peakfrom the resting 12-lead ECG, we identify 10 novel common genetic variants associated with cardiorespiratory fitness.

Leggi
Novembre 2023

Abstract 13690: Diagnosis of Left Ventricular Hypertrophy on 12-leads Electrocardiogram by Historical Criterion and Machine Learning Models

Circulation, Volume 148, Issue Suppl_1, Page A13690-A13690, November 6, 2023. Introduction:Diagnostic criterion of left ventricular hypertrophy (LVH) on 12-leads electrocardiogram (ECG) were established. We verified them comparing with artificial intelligence (AI) method.Hypothesis:Machine learning on 12-leads ECG show higher diagnostic performance comparing with historical criterion.Methods:First, consecutive 60 patients with LVH were recruited, and one to one matching with age and sex to patients with normal cardiac function was performed. Finally, 120 patients (69.6 ± 12.6years, 38men per group) were enrolled. LVH was defined as at least one LV wall (septum, posterior wall, apex) showed thickness over 15mm on ultrasound echocardiography. No sinus rhythm, and wide QRS cases were excluded.Results:By logistic regression analysis, 77 significant predictors were extracted. Among historical criterion, Cornell voltage showed high accuracy (0.783) and area under receiver operating characteristics curve analysis (AUROC; 0.808). Conversely, among AI methods, light gradient boosting machine demonstrated higher accuracy (0.843) and random forest method higher AUROC (0.882). V2/V2 S-wave amplitude and I/V5 T-wave amplitude played essential roles to build the AI models.Conclusions:AI diagnosis on ECG for LVH showed powerful diagnostic performance comparing historical criterion.

Leggi
Novembre 2023

Abstract 14025: Health Status Profiles in Patients With New or Worsening Peripheral Artery Disease Symptoms and 12-month Hospitalization Risk From the Portrait Registry

Circulation, Volume 148, Issue Suppl_1, Page A14025-A14025, November 6, 2023. Introduction:Risk stratification for healthcare utilization in PAD is critical given rising costs. The association between health status measures and hospitalization is unknown. We examined the association between a disease-specific patient-reported outcome measure and risk of hospitalization at 12 months.Methods:Patients with new or worsened lower extremity claudication enrolled at US sites in the PORTRAIT registry from 2011 to 2015 were included. The Peripheral Artery Questionnaire, a PAD-specific patient-reported outcome measure, was used to measure health status. PAQ summary scores range from 0 to 100 (better health status). Kaplan-Meier failure curves and adjusted Cox proportional hazards models assessed the association between baseline PAQ summary score and (1) the combined endpoint of all-cause hospital admission and ED visit (AD-ED) and (2) all-cause hospital admission (AD) at 12 months.Results:Of the 796 patients (mean age 69 ± 10 years, 42% female, 72% white, mean baseline PAQ summary score 46.8 ± 22.0) included, 349 (44%) had a hospital admission or ED visit at 12 months, with a total of 661 visits. Patients in the lowest PAQ quartile had higher rates of AD-ED at 30 days (16.1% vs. 4.3%), 90 days (29.8% vs. 12.8%), and 12 months (53.3% vs. 22.4%). In the fully adjusted model, lower PAQ score was associated with higher risk for AD-ED (HR per 10-point decrease, 1.12, 95% CI, 1.06-1.19, P

Leggi
Novembre 2023

Abstract 16682: Artificial Intelligence Predicts All-Cause and Cardiovascular Mortalities Using 12-Lead Electrocardiography

Circulation, Volume 148, Issue Suppl_1, Page A16682-A16682, November 6, 2023. Introduction:Electrocardiography (ECG) can be easily obtained at a low cost and includes voltage and time interval representing heart conditions. We hypothesized that artificial intelligence (AI) detects a subtle abnormality in 12-lead ECG and may predict individual mortality.Methods:Among 502,411 population in UK Biobank, 42,096 individuals had 12-lead ECG from 2013 to 2022. Among 41,572 survival group, after adjusting the following inclusion criteria; normal sinus rhythm, age under 60 years, PR interval 120~200ms, QTc interval 350~460ms, and QRS duration 70~100ms, 4,512 individuals were enrolled in this study. We developed and tested convolutional neural network (CNN) model to predict all cause death, cardiovascular (CV) death, or sudden cardiac arrest (SCA). The study population were divided into train (80%), validation (10%), and test (20%) set.Results:Among 4,512 patients with median 3.7 years [IQR; 2.7-5.1] of follow-up, the rate of all-cause mortality was 11.6% (524). In overall study population, median age was 55.5 years and proportion of male sex was 42.2%. The patients with all-cause death were older (p

Leggi
Novembre 2023

Use of CT, ED presentation and hospitalisations 12 months before and after a diagnosis of cancer in Western Australia: a population-based retrospective cohort study

Objective
To examine the use of CT, emergency department (ED)-presentation and hospitalisation and in 12 months before and after a diagnosis of cancer.

Design
Population-based retrospective cohort study.

Setting
West Australian linked administrative records at individual level.

Participants
104 009 adults newly diagnosed with cancer in 2004–2014.

Main outcome measures
CT use, ED presentations, hospitalisations.

Results
As compared with the rates in the 12th month before diagnosis, the rate of CT scans started to increase from 2 months before diagnosis with an increase in both ED presentations and hospitalisation from 1 month before the diagnosis. These rates peaked in the month of diagnosis for CT scans (477 (95% CI 471 to 482) per 1000 patients), and for hospitalisations (910 (95% CI 902 to 919) per 1000 patients), and the month prior to diagnosis for ED (181 (95% CI 178 to 184) per 1000 patients) then rapidly reduced after diagnosis but remained high for the next 12 months. While the patterns of the health services used were similar between 2004 and 2014, the rate of the health services used during after diagnosis was higher in 2014 versus 2004 except for CT use in patients with lymphohaematopoietic cancer with a significant reduction.

Conclusion
Our results showed an increase in demand for health services from 2 months before diagnosis of cancer. Increasing use of health services during and post cancer diagnosis may warrant further investigation to identify factors driving this change.

Leggi
Ottobre 2023