Predicting Out-of-Hospital Cardiac Arrest in the General Population Using Electronic Health Records.

BACKGROUND: The majority of out-of-hospital cardiac arrests (OHCAs) occur among individuals in the general population, for whom there is no established strategy to identify risk. In this study, we assess the use of electronic health record (EHR) data to identify OHCA in the general population and define salient factors contributing to OHCA risk. METHODS: The analytical cohort included 2366 individuals with OHCA and 23 660 age- and sex-matched controls receiving health care at the University of Washington. Comorbidities, electrocardiographic measures, vital signs, and medication prescription were abstracted from the EHR. The primary outcome was OHCA. Secondary outcomes included shockable and nonshockable OHCA. Model performance including area under the receiver operating characteristic curve and positive predictive value were assessed and adjusted for observed rate of OHCA across the health system. RESULTS: There were significant differences in demographic characteristics, vital signs, electrocardiographic measures, comorbidities, and medication distribution between individuals with OHCA and controls. In external validation, discrimination in machine learning models (area under the receiver operating characteristic curve 0.80-0.85) was superior to a baseline model with conventional cardiovascular risk factors (area under the receiver operating characteristic curve 0.66). At a specificity threshold of 99%, correcting for baseline OHCA incidence across the health system, positive predictive value was 2.5% to 3.1% in machine learning models compared with 0.8% for the baseline model. Longer corrected QT interval, substance abuse disorder, fluid and electrolyte disorder, alcohol abuse, and higher heart rate were identified as salient predictors of OHCA risk across all machine learning models. Established cardiovascular risk factors retained predictive importance for shockable OHCA, but demographic characteristics (minority race, single marital status) and noncardiovascular comorbidities (substance abuse disorder) also contributed to risk prediction. For nonshockable OHCA, a range of salient predictors, including comorbidities, habits, vital signs, demographic characteristics, and electrocardiographic measures, were identified. CONCLUSIONS: In a population-based case-control study, machine learning models incorporating readily available EHR data showed reasonable discrimination and risk enrichment for OHCA in the general population. Salient factors associated with OCHA risk were myriad across the cardiovascular and noncardiovascular spectrum. Public health and tailored strategies for OHCA prediction and prevention will require incorporation of this complexity.

Academic cardiac electrophysiologists' perspectives on sleep apnea care.

PURPOSE: Obstructive sleep apnea syndrome (OSAS) is an important, modifiable risk factor in the pathophysiology of arrhythmias including atrial fibrillation (AF). The purpose of the study was to evaluate cardiac electrophysiologists' (EPs) perception of OSAS. METHODS: We designed a 27-item online Likert scale-based survey instrument entailing several domains: (1) relevance of OSAS in EP practice, (2) OSAS screening and diagnosis, (3) perception on treatments for OSAS, (4) opinion on the OSAS care model. The survey was distributed to 89 academic EP programs in the USA and Canada. While the survey instrument questions refer to the term sleep apnea (SA), our discussion of the diagnosis, management, and research on the sleep disorder is more accurately described with the term OSAS. RESULTS: A total of 105 cardiac electrophysiologists from 49 institutions responded over a 9-month period. The majority of respondents agreed that sleep apnea (SA) is a major concern in their practice (94%). However, 42% reported insufficient education on SA during training. Many (58%) agreed that they would be comfortable managing SA themselves with proper training and education and 66% agreed cardiac electrophysiologists should become more involved in management. Half of EPs (53%) were not satisfied with the sleep specialist referral process. Additionally, a majority (86%) agreed that trained advanced practice providers should be able to assess and manage SA. Time constraints, lack of knowledge, and the referral process are identified as major barriers to EPs becoming more involved in SA care. CONCLUSIONS: We found that OSAS is widely recognized as a major concern for EP. However, incorporation of OSAS care in training and routine practice lags. Barriers to increased involvement include time constraints and education. This study can serve as an impetus for innovation in the cardiology OSAS care model.

Prediction of Sudden Cardiac Death: Looking Beyond Ejection Fraction.

PURPOSE OF REVIEW: Sudden cardiac death (SCD) is a major public health burden accounting for 15-20% of global mortality. Contemporary guidelines for SCD prevention are centered around the presence of low left ventricular ejection fraction, although the majority of SCD accrues in those not meeting contemporary criteria for SCD prevention. The goal of this review is to elaborate on the contemporary landscape of SCD prediction tools and further highlight gaps and opportunities in SCD prediction and prevention. RECENT FINDINGS: There have been considerable advancements in both non-invasive and invasive measures for SCD risk prediction including clinical morbidities, electrocardiographic measures, cardiac imaging (nuclear, magnetic resonance, computed tomography), serum biomarkers, genetics, and invasively assessed electrophysiological characteristics. Novel methodological approaches including application of machine learning, incorporation of competing risk, and use of computational modeling have underscored a future of personalized risk prediction. SCD remains a vital public health challenge. Emerging methods highlight opportunities to improve SCD prediction in the majority of those at risk who do not meet contemporary criteria for SCD prevention therapies. Future efforts will need to focus on easily deployed, multi-parametric risk models that enrich for SCD risk and not for competing mortality.

Autonomic modulation and cardiac arrhythmias: old insights and novel strategies.

The autonomic nervous system (ANS) plays a critical role in both health and states of cardiovascular disease. There has been a long-recognized role of the ANS in the pathogenesis of both atrial and ventricular arrhythmias (VAs). This historical understanding has been expanded in the context of evolving insights into the anatomy and physiology of the ANS, including dysfunction of the ANS in cardiovascular disease such as heart failure and myocardial infarction. An expanding armamentarium of therapeutic strategies-both invasive and non-invasive-have brought the potential of ANS modulation to contemporary clinical practice. Here, we summarize the integrative neuro-cardiac anatomy underlying the ANS, review the physiological rationale for autonomic modulation in atrial and VAs, highlight strategies for autonomic modulation, and finally frame future challenges and opportunities for ANS therapeutics.

Validation of electrocardiographic criteria for identifying left ventricular dysfunction in patients with previous myocardial infarction.

BACKGROUND: Eleven criteria correlating electrocardiogram (ECG) findings with reduced left ventricular ejection fraction (LVEF) have been previously published. These have not been compared head-to-head in a single study. We studied their value as a screening test to identify patients with reduced LVEF estimated by cardiac magnetic resonance (CMR) imaging. METHODS: ECGs and CMR from 548 patients (age 61 + 11 years, 79% male) with previous myocardial infarction (MI), from the DETERMINE and PRE-DETERMINE studies, were analyzed. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of each criterion for identifying patients with LVEF ≤ 30% and ≤ 40% were studied. A useful screening test should have high sensitivity and NPV. RESULTS: Mean LVEF was 40% (SD = 11%); 264 patients (48.2%) had LVEF ≤ 40%, and 96 patients (17.5%) had LVEF ≤ 30%. Six of 11 criteria were associated with a significant lower LVEF, but had poor sensitivity to identify LVEF ≤ 30% (range 2.1%-55.2%) or LVEF ≤ 40% (1.1%-51.1%); NPVs were good for LVEF ≤ 30% (range 82.8%-85.9%) but not for LVEF ≤ 40% (range 52.1%-60.6%). Goldberger's third criterion (RV4/SV4 < 1) and combinations of maximal QRS duration > 124 ms + either Goldberger's third criterion or Goldberger's first criterion (SV1 or SV2 + RV5 or RV6 ≥ 3.5 mV) had high specificity (95.4%-100%) for LVEF ≤ 40%, although seen in only 48 (8.8%) patients; predictive values were similar on subgroup analysis. CONCLUSIONS: None of the ECG criteria qualified as a good screening test. Three criteria had high specificity for LVEF ≤ 40%, although seen in < 9% of patients. Whether other ECG criteria can better identify LV dysfunction remains to be determined.

Simple electrocardiographic measures improve sudden arrhythmic death prediction in coronary disease.

AIMS: To determine whether the combination of standard electrocardiographic (ECG) markers reflecting domains of arrhythmic risk improves sudden and/or arrhythmic death (SAD) risk stratification in patients with coronary heart disease (CHD). METHODS AND RESULTS: The association between ECG markers and SAD was examined in a derivation cohort (PREDETERMINE; N = 5462) with adjustment for clinical risk factors, left ventricular ejection fraction (LVEF), and competing risk. Competing outcome models assessed the differential association of ECG markers with SAD and competing mortality. The predictive value of a derived ECG score was then validated (ARTEMIS; N = 1900). In the derivation cohort, the 5-year cumulative incidence of SAD was 1.5% [95% confidence interval (CI) 1.1-1.9] and 6.2% (95% CI 4.5-8.3) in those with a low- and high-risk ECG score, respectively (P for Δ < 0.001). A high-risk ECG score was more strongly associated with SAD than non-SAD mortality (adjusted hazard ratios = 2.87 vs. 1.38 respectively; P for Δ = 0.003) and the proportion of deaths due to SAD was greater in the high vs. low risk groups (24.9% vs. 16.5%, P for Δ = 0.03). Similar findings were observed in the validation cohort. The addition of ECG markers to a clinical risk factor model inclusive of LVEF improved indices of discrimination and reclassification in both derivation and validation cohorts, including correct reclassification of 28% of patients in the validation cohort [net reclassification improvement 28 (7-49%), P = 0.009]. CONCLUSION: For patients with CHD, an externally validated ECG score enriched for both absolute and proportional SAD risk and significantly improved risk stratification compared to standard clinical risk factors including LVEF. CLINICAL TRIAL REGISTRATION: https://clinicaltrials.gov/ct2/show/NCT01114269. ClinicalTrials.gov ID NCT01114269.

Genetic Obesity and the Risk of Atrial Fibrillation: Causal Estimates from Mendelian Randomization.

BACKGROUND: Observational studies have identified an association between body mass index (BMI) and incident atrial fibrillation (AF). Inferring causality from observational studies, however, is subject to residual confounding, reverse causation, and bias. The primary objective of this study was to evaluate the causal association between BMI and AF by using genetic predictors of BMI. METHODS: We identified 51 646 individuals of European ancestry without AF at baseline from 7 prospective population-based cohorts initiated between 1987 and 2002 in the United States, Iceland, and the Netherlands with incident AF ascertained between 1987 and 2012. Cohort-specific mean follow-up ranged from 7.4 to 19.2 years, over which period there was a total of 4178 cases of incident AF. We performed a Mendelian randomization with instrumental variable analysis to estimate a cohort-specific causal hazard ratio for the association between BMI and AF. Two genetic instruments for BMI were used: FTO genotype (rs1558902) and a BMI gene score comprising 39 single-nucleotide polymorphisms identified by genome-wide association studies to be associated with BMI. Cohort-specific estimates were combined by random-effects, inverse variance-weighted meta-analysis. RESULTS: In age- and sex-adjusted meta-analysis, both genetic instruments were significantly associated with BMI (FTO: 0.43 [95% confidence interval, 0.32-0.54] kg/m2 per A-allele, P<0.001; BMI gene score: 1.05 [95% confidence interval, 0.90-1.20] kg/m2 per 1-U increase, P<0.001) and incident AF (FTO, hazard ratio, 1.07 [1.02-1.11] per A-allele, P=0.004; BMI gene score, hazard ratio, 1.11 [1.05-1.18] per 1-U increase, P<0.001). Age- and sex-adjusted instrumental variable estimates for the causal association between BMI and incident AF were hazard ratio, 1.15 (1.04-1.26) per kg/m2, P=0.005 (FTO) and 1.11 (1.05-1.17) per kg/m2, P<0.001 (BMI gene score). Both of these estimates were consistent with the meta-analyzed estimate between observed BMI and AF (age- and sex-adjusted hazard ratio 1.05 [1.04-1.06] per kg/m2, P<0.001). Multivariable adjustment did not significantly change findings. CONCLUSIONS: Our data are consistent with a causal relationship between BMI and incident AF. These data support the possibility that public health initiatives targeting primordial prevention of obesity may reduce the incidence of AF.

Increasing sex differences in the use of cardiac resynchronization therapy with or without implantable cardioverter-defibrillator.

AIMS: Previous studies have identified sex disparities in the use of cardiac resynchronization therapy (CRT) and implantable cardioverter defibrillators (ICD), although the basis of underutilization in women remains poorly understood. The aim of this study was to assess sex differences in patterns of CRT use with our without ICD. METHODS AND RESULTS: In this cross-sectional study using the National Inpatient Sample database we identified 311 009 patients undergoing CRT implantation in the United States between 2006 and 2012. Demographic and clinical characteristics were compared between men and women undergoing CRT implantation, with special attention to clinical predictors of left ventricular reverse remodelling (CRT response, score range: 0-4) and reduced ICD efficacy (score range: 0-7). When compared to men, women undergoing CRT implantation were significantly more likely to have ≥ 3 predictors of CRT response (47.3 vs. 33.2%, P < 0.001) and less likely to have ≥3 predictors of reduced ICD efficacy (27.0 vs. 37.3%, P < 0.001). Despite this, men were significantly more likely to undergo CRT with ICD (CRT-D) as the type of CRT (88.6 vs. 80.1% of all CRT implants). Compared to those with the greatest likelihood of CRT response (score ≥ 3), those with the least likelihood of CRT response had a significant decreased odds of CRT-D implant (adj odds ratio 0.27 [0.24-0.31], P < 0.001), with a greater decreased odds in women compared to men (P, for sex interaction <0.001). The difference in the % of CRT-D implant in men vs. women increased over the study period (P, sex Δ time trend = 0.012). CONCLUSION: In this large, contemporary cohort, sex differences in CRT-D implantation were inversely related to predicted CRT efficacy and have increased over time. Future efforts to narrow the gap in CRT-D implantation in men and women may help better align device selection with those most likely to benefit.

National Trends in the Use of Cardiac Resynchronization Therapy With or Without Implantable Cardioverter-Defibrillator.

BACKGROUND: Candidates for cardiac resynchronization therapy (CRT) receive either a biventricular pacemaker or a biventricular pacemaker with an implantable cardioverter-defibrillator (CRT-D). Optimal device selection remains challenging because the benefit of implantable cardioverter-defibrillator therapy may not be uniform, particularly in patients at competing risk of nonsudden death. METHODS AND RESULTS: In this serial cross-sectional study using the National Inpatient Sample database, we identified 311,086 admissions associated with CRT implant between 2006 to 2012. CRT-D was the most common device type (86.1%), including in patients ≥ 75 years of age with ≥ 5 Elixhauser comorbidities (75.5%). Multivariate predictors of CRT-D implant included demographic, clinical, and geographic factors: prior ventricular arrhythmia (rate ratio [RR], 1.14; 95% CI, 1.13-1.14), ischemic heart disease (RR, 1.11; 95% CI, 1.10-1.11), male sex (RR, 1.10; 95% CI, 1.09-1.10), black race (RR, 1.06; 95% CI: 1.04-1.07), and Northeast geographic region (RR, 1.06; 95% CI, 1.04-1.09). There was significant interhospital variation in the use of CRT-D (10-90 percentile range, 72.9%-98.0% CRT-D). CONCLUSIONS: The majority of patients in this contemporary US cohort underwent implantation of CRT-D. Predictors of CRT-D implant included demographic, clinical, and geographic factors. In patient subgroups predicted to have an attenuated benefit from implantable cardioverter-defibrillator therapy (older adults with multiple comorbidities), CRT-D remained the dominant device type. An improved understanding of the determinants of device selection may aid in decision making and ultimately better align patient risk with device benefit at the time of CRT implantation.

Reduced appropriate implantable cardioverter-defibrillator therapy after cardiac resynchronization therapy-induced left ventricular function recovery: a meta-analysis and systematic review.

AIMS: For patients undergoing cardiac resynchronization therapy (CRT) with implantable cardioverter-defibrillator (ICD; CRT-D), the effect of an improvement in left ventricular ejection fraction (LVEF) on appropriate ICD therapy may have significant implications regarding management at the time of ICD generator replacement. METHODS AND RESULTS: We conducted a meta-analysis to determine the effect of LVEF recovery following CRT on the incidence of appropriate ICD therapy. A search of multiple electronic databases identified 709 reports, of which 6 retrospective cohort studies were included (n = 1740). In patients with post-CRT LVEF ≥35% (study n = 4), the pooled estimated rate of ICD therapy (5.5/100 person-years) was significantly lower than patients with post-CRT LVEF <35% [incidence rate difference (IRD): -6.5/100 person-years, 95% confidence interval (95% CI): -8.8 to -4.2, P < 0.001]. Similarly, patients with post-CRT LVEF ≥45% (study n = 4) demonstrated lower estimated rates of ICD therapy (2.3/100 person-years) compared with patients without such recovery (IRD: -5.8/100 person-years, 95% CI: -7.6 to -4.0, P < 0.001). Restricting analysis to studies discounting ICD therapies during LVEF recovery (study n = 3), patients with LVEF recovery (≥35 or ≥45%) had significantly lower rates of ICD therapy compared with patients without such recovery (P for both <0.001). Patients with primary prevention indication for ICD, regardless of LVEF recovery definition, had very low rates of ICD therapy (0.4 to 0.8/100-person years). CONCLUSION: Recovery of LVEF post-CRT is associated with significantly reduced appropriate ICD therapy. Patients with improvement of LVEF ≥45% and those with primary prevention indication for ICD appear to be at lowest risk.

Renal Response in Patients with Chronic Kidney Disease Predicts Outcome Following Cardiac Resynchronization Therapy.

BACKGROUND: Chronic kidney disease (CKD) severity is associated with increased morbidity and mortality in congestive heart failure. There is a paucity of data regarding renal improvement after cardiac resynchronization therapy (CRT) and its potential impact on clinical outcomes, especially in patients with severe CKD. METHOD: This was a retrospective analysis of a prospectively collected cohort of 260 patients with CKD undergoing CRT at a single center. Renal function was compared before and after CRT. The primary end point was a composite of death, heart transplant, and left ventricular assist device (LVAD), assessed at 5 years. RESULTS: Patients with more severe CKD demonstrated increased risk of death, transplant, or LVAD following CRT (P = 0.015). Renal response (estimated glomerular filtration rate improvement ≥10 mL/min/1.73 m(2) ) was observed in 14% of all patients and 28% of patients with stage IV CKD. Independent predictors of renal response included left ventricular ejection fraction improvement (odds ratio [OR] 1.06, confidence interval [CI] 1.01-1.10), angiotensin-converting enzyme inhibitor/angiotensin receptor blocker use (OR 4.31, CI 1.08-17.23), and advanced CKD stage (OR 2.19, CI 1.14-4.23). Renal response independently decreased hazard of the primary outcome (HR 0.24, CI 0.08-0.73, P = 0.01). Renal responders with stage IV CKD had 80% 5-year event-free survival, compared to 0% for nonrenal responders in stage IV (P = 0.03). CONCLUSION: Although severity of CKD is associated with poorer outcome after CRT, improvement in renal function can occur in patients across all CKD stages. Renal responders, including those with stage IV CKD, demonstrate favorable 5-year outcomes. Assessment of renal response may help better prognostic outcomes following CRT.

Cardiac resynchronization therapy: past, present, and future.

Cardiac resynchronization therapy (CRT), or biventricular pacing, has become a standard therapeutic modality for patients with symptomatic heart failure (HF), depressed left ventricular (LV) function, and electrical dyssynchrony. Despite the overall success of CRT in improving morbidity and mortality in selected patients with HF, a significant minority demonstrates nonresponse. This review describes the electrical and physiologic rationale for biventricular pacing therapy, summarizes landmark clinical trials assessing CRT efficacy, highlights strategies to optimize the response to CRT, and frames future challenges in the use, delivery, and care of patients undergoing CRT.

Racial differences in breast cancer stage at diagnosis in the mammography era.

OBJECTIVES: We assessed racial differences in breast cancer mortality by stage at diagnosis, since mammography became available. METHODS: We calculated adjusted odds of distant (versus local or regional) tumors for 143,249 White and 13,571 Black women aged 50 to 69 years, diagnosed with breast cancer between 1982 and 2007 and living in a Surveillance, Epidemiology, and End Results region. We compared linear trends in stage at diagnosis before and after 1998. RESULTS: Distant-stage cancer was diagnosed in 5.8% of White and 10.2% of Black participants. The Black-White disparity in distant tumors narrowed until 1998 (1998 adjusted difference = 0.65%), before increasing. Between 1982 and 1997, the proportion of distant tumors decreased for Blacks (adjusted odds ratio [AOR]/y = 0.973; 95% confidence interval [CI] = 0.960, 0.987) and Whites (AOR/y = 0.978; 95% CI = 0.973, 0.983), with no racial differences (P = .47). From 1998 to 2007, the odds of distant versus local or regional tumors increased for Blacks (AOR/y = 1.036; 95% CI = 1.013, 1.060) and Whites (AOR/y = 1.011; 95% CI = 1.002, 1.021); the rate of increase was greater for Blacks than Whites (P = .04). CONCLUSIONS: In the mammography era, racial disparities remain in stage at diagnosis.

Presentation and Outcomes of Adults With Overdose-Related Out-of-Hospital Cardiac Arrest.

Importance: Drug overdose (OD) is a public health challenge and an important cause of out-of-hospital cardiac arrest (OHCA). Existing studies evaluating OD-related OHCA (OD-OHCA) either aggregate all drugs or focus on opioids. The epidemiology, presentation, and outcomes of drug-specific OHCA are largely unknown. Objective: To evaluate the temporal pattern, clinical presentation, care, and outcomes of adult patients with OHCA overall and according to the drug-specific profile. Design, Setting, and Participants: This cohort study of adults with OHCA in King County Washington was conducted between January 1, 2015, and December 31, 2021. Etiology of OHCA was determined using emergency medical service, hospital, and medical examiner records. Etiology was classified as non-OD OHCA or OD-OHCA, with drug-specific profiles categorized as (1) opioid without stimulant, (2) stimulant without opioid, (3) opioid and stimulant, or (4) all other nonstimulant, nonopioid drugs. Statistical analysis was performed on July 1, 2023. Exposure: Out-of-hospital cardiac arrest. Main Outcomes and Measures: The primary outcome was survival to hospital discharge. The secondary outcome was survival with favorable functional status defined by Cerebral Performance Category 1 or 2 based on review of the hospital record. Results: In this cohort study, there were 6790 adult patients with emergency medical services-treated OHCA from a US metropolitan system. During the 7-year study period, there were 702 patients with OD-OHCA (median age, 41 years [IQR, 29-53 years]; 64% male [n = 450] and 36% female [n = 252]) and 6088 patients with non-OD OHCA (median age, 66 years [IQR, 56-77 years]; 65% male [n = 3944] and 35% female [n = 2144]). The incidence of OD-OHCA increased from 5.2 (95% CI, 3.8-6.6) per 100 000 person-years in 2015 to 13.0 (95% CI, 10.9-15.1) per 100 000 person-years in 2021 (P < .001 for trend), whereas there was no significant temporal change in the incidence of non-OD OHCA (P = .30). OD-OHCA were more likely to be unwitnessed (66% [460 of 702] vs 41% [2515 of 6088]) and less likely to be shockable (8% [56 of 702] vs 25% [1529 of 6088]) compared with non-OD OHCA. Unadjusted survival was not different (20% [138 of 702] for OD vs 18% [1095 of 6088] for non-OD). When stratified by drug profile, combined opioid-stimulant OHCA demonstrated the greatest relative increase in incidence. Presentation and outcomes differed by drug profile. Patients with stimulant-only OHCA were more likely to have a shockable rhythm (24% [31 of 129]) compared with patients with opioid-only OHCA (4% [11 of 295]) or patients with combined stimulant-opioid OHCA 5% [10 of 205]), and they were more likely to have a witnessed arrest (50% [64 of 129]) compared with patients with OHCA due to other drugs (19% [14 of 73]) or patients with combined stimulant-opioid OHCA (23% [48 of 205]). Patients with a combined opioid-stimulant OHCA had the lowest survival to hospital discharge (10% [21 of 205]) compared with patients with stimulant-only OHCA (22% [29 of 129]) or patients with OHCA due to other drugs (26% [19 of 73]), a difference that persisted after multivariable adjustment. Conclusions and Relevance: In a population-based cohort study, the incidence of OD-OHCA increased significantly from 2015 to 2021, with the greatest increase observed among patients with a combined stimulant-opioid OHCA. Presentation and outcome differed according to the drug-specific profile. The combination of increasing incidence and lower survival among among patients with a opioid-stimulant OHCA supports prevention and treatment initiatives that consider the drug-specific profile.

Polygenic Risk Score Predicts Sudden Death in Patients With Coronary Disease and Preserved Systolic Function.

BACKGROUND: A familial predisposition to sudden and/or arrhythmic death (SAD) in the setting of coronary artery disease (CAD) exists; however, the genetic basis is poorly understood. OBJECTIVES: The purpose of this study was to determine whether a genome-wide polygenic score for coronary artery disease (GPSCAD) might have utility in SAD risk stratification in CAD patients without severe systolic dysfunction. METHODS: A previously validated GPSCAD was generated utilizing genome-wide genotyping in 4,698 PRE-DETERMINE participants of European ancestry with CAD and left ventricular ejection fraction >30%-35%. The population was dichotomized according to top GPSCAD decile as defined by the general population, and absolute, proportional, and relative risks for SAD and non-SAD were estimated using competing risk analyses. RESULTS: Over a median follow-up of 8.0 years, participants in the top GPSCAD decile were at elevated absolute SAD risk (8.0%; 95% CI: 5.1%-12.4% vs 4.8%; 95% CI: 3.3%-7.0%; P = 0.005) and proportional SAD risk (29% vs 16%; P = 0.0003) compared with the remainder. After controlling for left ventricular ejection fraction, clinical factors, and electrocardiogram parameters, the top GPSCAD decile was associated with SAD (subdistribution HR: 1.77; 95% CI: 1.23-2.54; P = 0.002) but not non-SAD (subdistribution HR: 1.00; 95% CI: 0.80-1.25; P = 0.98) (P for Δ = 0.003). The addition of the top GPSCAD decile to the multivariable model significantly improved net reclassification indexes (NRIs) (continuous NRI: 14.0%; P = 0.024; and categorical NRI: 6.6%; P = 0.005) but not the C-index (difference in C-index: 0.007; P = 0.143). CONCLUSIONS: Among CAD patients without severe systolic dysfunction, high GPSCAD specifically predicted SAD and enriched for both absolute and proportional SAD risk, identifying a population who might benefit from defibrillator therapy.

Identifying risk of adverse outcomes in COVID-19 patients via artificial intelligence-powered analysis of 12-lead intake electrocardiogram.

BACKGROUND: Adverse events in COVID-19 are difficult to predict. Risk stratification is encumbered by the need to protect healthcare workers. We hypothesize that artificial intelligence (AI) can help identify subtle signs of myocardial involvement in the 12-lead electrocardiogram (ECG), which could help predict complications. OBJECTIVE: Use intake ECGs from COVID-19 patients to train AI models to predict risk of mortality or major adverse cardiovascular events (MACE). METHODS: We studied intake ECGs from 1448 COVID-19 patients (60.5% male, aged 63.4 ± 16.9 years). Records were labeled by mortality (death vs discharge) or MACE (no events vs arrhythmic, heart failure [HF], or thromboembolic [TE] events), then used to train AI models; these were compared to conventional regression models developed using demographic and comorbidity data. RESULTS: A total of 245 (17.7%) patients died (67.3% male, aged 74.5 ± 14.4 years); 352 (24.4%) experienced at least 1 MACE (119 arrhythmic, 107 HF, 130 TE). AI models predicted mortality and MACE with area under the curve (AUC) values of 0.60 ± 0.05 and 0.55 ± 0.07, respectively; these were comparable to AUC values for conventional models (0.73 ± 0.07 and 0.65 ± 0.10). There were no prominent temporal trends in mortality rate or MACE incidence in our cohort; holdout testing with data from after a cutoff date (June 9, 2020) did not degrade model performance. CONCLUSION: Using intake ECGs alone, our AI models had limited ability to predict hospitalized COVID-19 patients' risk of mortality or MACE. Our models' accuracy was comparable to that of conventional models built using more in-depth information, but translation to clinical use would require higher sensitivity and positive predictive value. In the future, we hope that mixed-input AI models utilizing both ECG and clinical data may be developed to enhance predictive accuracy.