Search

Background: New methods such as machine learning techniques have been increasingly used to enhance the performance of risk predictions for clinical decision-making. However, commonly reported performance metrics may not be sufficient to capture the advantages of these newly proposed models for their adoption by health care professionals to improve care. Machine learning models often improve risk estimation for certain subpopulations that may be missed by these metrics.Methods and Results: This article addresses the limitations of commonly reported metrics for performance comparison and proposes additional metrics. Our discussions cover metrics related to overall performance, discrimination, calibration, resolution, reclassification, and model implementation. Models for predicting acute kidney injury after percutaneous coronary intervention are used to illustrate the use of these metrics.Conclusions: We demonstrate that commonly reported metrics may not have sufficient sensitivity to identify improvement of machine learning models and propose the use of a comprehensive list of performance metrics for reporting and comparing clinical risk prediction models. [ABSTRACT FROM AUTHOR]

Background: Intraoperative data may improve models predicting postoperative events. We evaluated the effect of incorporating intraoperative variables to the existing preoperative model on the predictive performance of the model for coronary artery bypass graft.Methods: We analyzed 378 572 isolated coronary artery bypass graft cases performed across 1083 centers, using the national Society of Thoracic Surgeons Adult Cardiac Surgery Database between 2014 and 2016. Outcomes were operative mortality, 5 postoperative complications, and composite representation of all events. We fitted models by logistic regression or extreme gradient boosting (XGBoost). For each modeling approach, we used preoperative only, intraoperative only, or pre+intraoperative variables. We developed 84 models with unique combinations of the 3 variable sets, 2 variable selection methods, 2 modeling approaches, and 7 outcomes. Each model was tested in 20 iterations of 70:30 stratified random splitting into development/testing samples. Model performances were evaluated on the testing dataset using the C statistic, area under the precision-recall curve, and calibration metrics, including the Brier score.Results: The mean patient age was 65.3 years, and 24.7% were women. Operative mortality, excluding intraoperative death, occurred in 1.9%. In all outcomes, models that considered pre+intraoperative variables demonstrated significantly improved Brier score and area under the precision-recall curve compared with models considering pre or intraoperative variables alone. XGBoost without external variable selection had the best C statistics, Brier score, and area under the precision-recall curve values in 4 of the 7 outcomes (mortality, renal failure, prolonged ventilation, and composite) compared with logistic regression models with or without variable selection. Based on the calibration plots, risk restratification for mortality showed that the logistic regression model underestimated the risk in 11 114 patients (9.8%) and overestimated in 12 005 patients (10.6%). In contrast, the XGBoost model underestimated the risk in 7218 patients (6.4%) and overestimated in 0 patients (0%).Conclusions: In isolated coronary artery bypass graft, adding intraoperative variables to preoperative variables resulted in improved predictions of all 7 outcomes. Risk models based on XGBoost may provide a better prediction of adverse events to guide clinical care. [ABSTRACT FROM AUTHOR]

BACKGROUND: The National Cardiovascular Data Registry Left Atrial Appendage Occlusion Registry (LAAO) includes the vast majority of transcatheter LAAO procedures performed in the United States. The objective of this study was to develop a model predicting adverse events among patients undergoing LAAO with Watchman FLX. METHODS: Data from 41 001 LAAO procedures with Watchman FLX from July 2020 to September 2021 were used to develop and validate a model predicting in-hospital major adverse events. Randomly selected development (70%, n=28 530) and validation (30%, n=12 471) cohorts were analyzed with 1000 bootstrapped samples, using forward stepwise logistic regression to create the final model. A simplified bedside risk score was also developed using this model. RESULTS: Increased age, female sex, low preprocedure hemoglobin, no prior attempt at atrial fibrillation termination, and increased fall risk most strongly predicted in-hospital major adverse events and were included in the final model along with other clinically relevant variables. The median in-hospital risk-standardized adverse event rate was 1.50% (range, 1.03%–2.84%; interquartile range, 1.42%–1.64%). The model demonstrated moderate discrimination (development C-index, 0.67 [95% CI, 0.65–0.70] and validation C-index, 0.66 [95% CI, 0.62–0.70]) with good calibration. The simplified risk score was well calibrated with risk of in-hospital major adverse events ranging from 0.26% to 3.90% for a score of 0 to 8, respectively. CONCLUSIONS: A transcatheter LAAO risk model using National Cardiovascular Data Registry and LAAO Registry data can predict in-hospital major adverse events, demonstrated consistency across hospitals and can be used for quality improvement efforts. A simple bedside risk score was similarly predictive and may inform shared decision-making. [ABSTRACT FROM AUTHOR]

Background: Left ventricular ejection fraction remains the primary risk stratification tool used in the selection of patients for implantable cardioverter defibrillator therapy. However, this solitary marker fails to identify a substantial portion of patients experiencing sudden cardiac arrest. In this study, we examined the incremental value of considering right ventricular ejection fraction for the prediction of future arrhythmic events in patients with systolic dysfunction using the gold standard of cardiovascular magnetic resonance.Methods and Results: Three hundred fourteen consecutive patients with ischemic cardiomyopathy or nonischemic dilated cardiomyopathy undergoing cardiovascular magnetic resonance were followed for the primary outcome of sudden cardiac arrest or appropriate implantable cardioverter defibrillator therapy. Blinded quantification of left ventricular and right ventricular (RV) volumes was performed from standard cine imaging. Quantification of fibrosis from late gadolinium enhancement imaging was incrementally performed. RV dysfunction was defined as right ventricular ejection fraction ≤45%. Among all patients (164 ischemic cardiomyopathy, 150 nonischemic dilated cardiomyopathy), the mean left ventricular ejection fraction was 32±12% (range, 6-54%) with mean right ventricular ejection fraction of 48±15% (range, 7-78%). At a median of 773 days, 49 patients (15.6%) experienced the primary outcome (9 sudden cardiac arrest, 40 appropriate implantable cardioverter defibrillator therapies). RV dysfunction was independently predictive of the primary outcome (hazard ratio=2.98; P=0.002). Among those with a left ventricular ejection fraction >35% (N=121; mean left ventricular ejection fraction, 45±6%), RV dysfunction provided an adjusted hazard ratio of 4.2 (P=0.02).Conclusions: RV dysfunction is a strong, independent predictor of arrhythmic events. Among patients with mild to moderate LV dysfunction, a cohort greatly contributing to global sudden cardiac arrest burden, this marker provides robust discrimination of high- versus low-risk subjects. [ABSTRACT FROM AUTHOR]

Background: Visit-to-visit variability (VVV) in blood pressure values has been reported in clinical studies. However, little is known about VVV in clinical practice and whether it is associated with patient characteristics in real-world setting. Methods: We conducted a retrospective cohort study to quantify VVV in systolic blood pressure (SBP) values in a real-world setting. We included adults (age ≥18 years) with at least 2 outpatient visits between January 1, 2014 and October 31, 2018 from Yale New Haven Health System. Patient-level measures of VVV included SD and coefficient of variation of a given patient's SBP across visits. We calculated patient-level VVV overall and by patient subgroups. We further developed a multilevel regression model to assess the extent to which VVV in SBP was explained by patient characteristics. Results: The study population included 537 218 adults, with a total of 7 721 864 SBP measurements. The mean age was 53.4 (SD 19.0) years, 60.4% were women, 69.4% were non-Hispanic White, and 18.1% were on antihypertensive medications. Patients had a mean body mass index of 28.4 (5.9) kg/m2 and 22.6%, 8.0%, 9.7%, and 5.6% had a history of hypertension, diabetes, hyperlipidemia, and coronary artery disease, respectively. The mean number of visits per patient was 13.3, over an average period of 2.4 years. The mean (SD) intraindividual SD and coefficient of variation of SBP across visits were 10.6 (5.1) mm Hg and 0.08 (0.04). These measures of blood pressure variation were consistent across patient subgroups defined by demographic characteristics and medical history. In the multivariable linear regression model, only 4% of the variance in absolute standardized difference was attributable to patient characteristics. Conclusions: The VVV in real-world practice poses challenges for management of patients with hypertension based on blood pressure readings in outpatient settings and suggest the need to go beyond episodic clinic evaluation. [ABSTRACT FROM AUTHOR]

Background: The current ability to predict readmissions in patients with heart failure is modest at best. It is unclear whether machine learning techniques that address higher dimensional, nonlinear relationships among variables would enhance prediction. We sought to compare the effectiveness of several machine learning algorithms for predicting readmissions.Methods and Results: Using data from the Telemonitoring to Improve Heart Failure Outcomes trial, we compared the effectiveness of random forests, boosting, random forests combined hierarchically with support vector machines or logistic regression (LR), and Poisson regression against traditional LR to predict 30- and 180-day all-cause readmissions and readmissions because of heart failure. We randomly selected 50% of patients for a derivation set, and a validation set comprised the remaining patients, validated using 100 bootstrapped iterations. We compared C statistics for discrimination and distributions of observed outcomes in risk deciles for predictive range. In 30-day all-cause readmission prediction, the best performing machine learning model, random forests, provided a 17.8% improvement over LR (mean C statistics, 0.628 and 0.533, respectively). For readmissions because of heart failure, boosting improved the C statistic by 24.9% over LR (mean C statistic 0.678 and 0.543, respectively). For 30-day all-cause readmission, the observed readmission rates in the lowest and highest deciles of predicted risk with random forests (7.8% and 26.2%, respectively) showed a much wider separation than LR (14.2% and 16.4%, respectively).Conclusions: Machine learning methods improved the prediction of readmission after hospitalization for heart failure compared with LR and provided the greatest predictive range in observed readmission rates. [ABSTRACT FROM AUTHOR]

Background--Diabetics remain at high risk of cardiovascular disease and mortality despite advancements in medical therapy. Noninvasive cardiac risk profiling is often more difficult in diabetics owing to the prevalence of silent ischemia with unrecognized myocardial infarction, reduced exercise capacity, nondiagnostic electrocardiographic changes, and balanced ischemia from diffuse epicardial coronary atherosclerosis and microvascular dysfunction. Methods and Results--A consecutive cohort of 173 patients with diabetes mellitus (mean age, 61.7±11.9 years; 37% women) with suspected myocardial ischemia underwent stress perfusion cardiac magnetic resonance imaging. Patients were evaluated for adverse cardiac events after cardiac magnetic resonance imaging with mean follow-up time of 2.9±2.5 years. Mean hemoglobin A1c for the population was 7.9±1.8%. Primary end point was a composite of cardiac death and nonfatal myocardial infarction. Diabetics with no inducible ischemia (n=94) experienced an annualized event rate of 1.4% compared with 8.2% (P=0.0003) in those with inducible ischemia (n=79). Diabetics without late gadolinium enhancement or inducible ischemia had a low annual cardiac event rate (0.5% per year). The presence of inducible ischemia was the strongest unadjusted predictor (hazard ratio, 4.86; P<0.01) for cardiac death and nonfatal myocardial infarction. This association remained robust in adjusted stepwise multivariable Cox regression analysis (hazard ratio, 4.28; P=0.02). In addition, categorical net reclassification index using 5-year risk cutoffs of 5% and 10% resulted in reclassification of 43.4% of the diabetic cohort with net reclassification index of 0.38 (95% confidence interval, 0.20-0.56; P<0.0001). Conclusions--Stress perfusion cardiac magnetic resonance imaging provided independent prognostic utility and effectively reclassified risk in patients with diabetes mellitus referred for ischemic assessment. Further evaluation is required to determine whether a noninvasive imaging strategy with cardiac magnetic resonance imaging can favorably affect downstream outcomes and improve cost-effectiveness of care in diabetics. [ABSTRACT FROM AUTHOR]

Strategies for prevention of sudden cardiac death focus on severe left ventricular (LV) dysfunction, although most sudden cardiac death postmyocardial infarction occurs in patients with mild/moderate LV dysfunction. We tested the hypothesis that infarct heterogeneity by cardiac magnetic resonance is associated with mortality beyond LV ejection fraction (LVEF) in patients with coronary artery disease and LV dysfunction. In addition, we examined the association between infarct heterogeneity and mortality in those with LVEF >35%.We studied 301 patients with coronary artery disease and LV dysfunction referred for cardiac magnetic resonance. We quantified total infarct mass, infarct core mass, and peri-infarct zone (PIZ) normalized for total infarct mass (%PIZ) using signal-intensity criteria of >2 SDs, >3 SDs, and 2- to -3 SDs above remote myocardium, respectively. Mean LVEF was 41±14%. After 3.9 years median follow-up, 66 (22%) patients died (13 sudden cardiac death; 33 with LVEF >35%). In patients with LVEF >35%, below-median %PIZ carried an annual death rate of 2.8% versus 12% in patients with above-median %PIZ (P<0.001). In a multivariable model, %PIZ maintained strong association with mortality adjusted to patient age, LVEF, right ventricular ejection fraction, prolonged QT interval, and total infarct size and resulted in improve risk reclassification 0.492 (95% confidence interval, 0.183-0.817).Cardiac magnetic resonance infarct heterogeneity has a strong association with mortality independent of LVEF in patients with coronary artery disease and LV dysfunction, particularly in patients with mild or moderate LV dysfunction. Further studies incorporating cardiac magnetic resonance in clinical decision making for defibrillator therapy are warranted. [ABSTRACT FROM AUTHOR]

Myocardial fibrosis is a hallmark of hypertrophic cardiomyopathy (HCM) and a potential substrate for arrhythmias and heart failure. Sarcomere mutations seem to induce profibrotic changes before left ventricular hypertrophy (LVH) develops. To further evaluate these processes, we used cardiac magnetic resonance with T1 measurements on a genotyped HCM population to quantify myocardial extracellular volume (ECV).Sarcomere mutation carriers with LVH (G+/LVH+, n=37) and without LVH (G+/LVH-, n=29), patients with HCM without mutations (sarcomere-negative HCM, n=11), and healthy controls (n=11) underwent contrast cardiac magnetic resonance, measuring T1 times pre- and postgadolinium infusion. Concurrent echocardiography and serum biomarkers of collagen synthesis, hemodynamic stress, and myocardial injury were also available in a subset. Compared with controls, ECV was increased in patients with overt HCM, as well as G+/LVH- mutation carriers (ECV=0.36±0.01, 0.33±0.01, 0.27±0.01 in G+/LVH+, G+/LVH-, controls, respectively; P0.001 for all comparisons). ECV correlated with N-terminal probrain natriuretic peptide levels (r=0.58; P<0.001) and global E’ velocity (r=-0.48; P<0.001). Late gadolinium enhancement was present in >60% of overt patients with HCM but absent from G+/LVH- subjects. Both ECV and late gadolinium enhancement were more extensive in sarcomeric HCM than sarcomere-negative HCM.Myocardial ECV is increased in HCM sarcomere mutation carriers even in the absence of LVH. These data provide additional support that fibrotic remodeling is triggered early in disease pathogenesis. Quantifying ECV may help characterize the development of myocardial fibrosis in HCM and ultimately assist in developing novel disease-modifying therapy, targeting interstitial fibrosis. [ABSTRACT FROM AUTHOR]

Objectives: It would be useful to have a quick and cost-effective method to study individual-level determinants of mortality in countries where reliable data are not available. We have modified indirect demographic methods and applied them to a population sample to investigate predictors of mortality in Russia.Methods: A national sample of the Russian population was interviewed in a cross-sectional survey. The participants were asked about characteristics of their eldest siblings, including their vital status, year of birth, and year of death (if deceased). The association between personal characteristics and mortality risk was estimated for 682 male and 698 female siblings (of whom 122 and 81, respectively, had died).Results: In both sexes, mortality was strongly associated with smoking and low education. After adjustment for smoking and education, mortality was elevated in men and women who drank alcohol at least once a month. Mortality was also higher among in men who had been binge drinking (more than half a bottle of vodka per drinking session) at least once a week (adjusted risk ratio [RR] = 2.5; 95% confidence interval [CI] = 1.2-4.9) and in women who were binging at least once a month (RR = 3.9; CI = 1.1-14.5) compared with nonbinging.Subjects: Similar associations with drinking were seen for cardiovascular deaths in men. Childhood social circumstances were not associated with mortality.Conclusions: The study of siblings appears to be a cost-effective alternative for estimating risk factors for mortality in literate populations. This study identified smoking, low education, and alcohol consumption (especially binge drinking) as risk factors for mortality in Russia. [ABSTRACT FROM AUTHOR]

SUMMARY: OBJECTIVES It would be useful to have a quick and cost-effective method to study individual-level determinants of mortality in countries where reliable data are not available. We have modified indirect demographic methods and applied them to a population sample to investigate predictors of mortality in Russia.METHODS A national sample of the Russian population was interviewed in a cross-sectional survey. The participants were asked about characteristics of their eldest siblings, including their vital status, year of birth, and year of death (if deceased). The association between personal characteristics and mortality risk was estimated for 682 male and 698 female siblings (of whom 122 and 81, respectively, had died).RESULTS In both sexes, mortality was strongly associated with smoking and low education. After adjustment for smoking and education, mortality was elevated in men and women who drank alcohol at least once a month. Mortality was also higher among in men who had been binge drinking (more than half a bottle of vodka per drinking session) at least once a week (adjusted risk ratio [RR] = 2.5; 95% confidence interval [CI] = 1.2-4.9) and in women who were binging at least once a month (RR = 3.9; CI = 1.1-14.5) compared with nonbinging.SUBJECTS Similar associations with drinking were seen for cardiovascular deaths in men. Childhood social circumstances were not associated with mortality.CONCLUSIONS The study of siblings appears to be a cost-effective alternative for estimating risk factors for mortality in literate populations. This study identified smoking, low education, and alcohol consumption (especially binge drinking) as risk factors for mortality in Russia. [ABSTRACT FROM AUTHOR]

Supplemental Digital Content is available in the text. Background: There is increasing evidence that right ventricular ejection fraction (RVEF) may provide incremental value to left ventricular (LV) ejection fraction for the prediction of major adverse cardiovascular events. To date, generalizable utility for RVEF quantification in patients with cardiovascular disease has not been established. Using a large prospective clinical outcomes registry, we investigated the prognostic value of RVEF for the prediction of major adverse cardiovascular events- and heart failure-related outcomes. Methods: Seven thousand one hundred thirty-one consecutive patients with known or suspected cardiovascular disease undergoing cardiovascular magnetic resonance imaging were prospectively enrolled. Multichamber volumetric quantification was performed by standardized operational procedures. Patients were followed for the primary composite outcome of all-cause death, survived cardiac arrest, admission for heart failure, need for transplantation or LV assist device, acute coronary syndrome, need for revascularization, stroke, or transient ischemic attack. A secondary, heart failure focused outcome of heart failure admission, need for transplantation/LV assist device or death was also studied. Results: Mean age was 54±15 years. The mean LV ejection fraction was 55±14% (range 6%–90%) with a mean RVEF of 54±10% (range 9%–87%). At a median follow-up of 908 days, 870 (12%) patients experienced the primary composite outcome and 524 (7%) the secondary outcome. Each 10% drop in RVEF was associated with a 1.3-fold increased risk of the primary outcome (P <0.001) and 1.5-fold increased risk of the secondary outcome (P <0.001). RVEF was an independent predictor following comprehensive covariate adjustment, inclusive of LV ejection fraction. Patients with an RVEF<40% experienced a 3.1-fold risk of the primary outcome (P <0.001) with a 1-year cumulative event rate of 22% versus 7% above this cutoff. Conclusions: RVEF is a powerful and independent predictor of major adverse cardiac events with broad generalizability across patients with known or suspected cardiovascular disease. These findings support migration towards biventricular phenotyping for the classification of risk in clinical practice. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT04367220. [ABSTRACT FROM AUTHOR]

Use of machine learning (ML) in clinical research is growing steadily given the increasing availability of complex clinical data sets. ML presents important advantages in terms of predictive performance and identifying undiscovered subpopulations of patients with specific physiology and prognoses. Despite this popularity, many clinicians and researchers are not yet familiar with evaluating and interpreting ML analyses. Consequently, readers and peer-reviewers alike may either overestimate or underestimate the validity and credibility of an ML-based model. Conversely, ML experts without clinical experience may present details of the analysis that are too granular for a clinical readership to assess. Overwhelming evidence has shown poor reproducibility and reporting of ML models in clinical research suggesting the need for ML analyses to be presented in a clear, concise, and comprehensible manner to facilitate understanding and critical evaluation. We present a recommendation for transparent and structured reporting of ML analysis results specifically directed at clinical researchers. Furthermore, we provide a list of key reporting elements with examples that can be used as a template when preparing and submitting ML-based manuscripts for the same audience. [ABSTRACT FROM AUTHOR]

Supplemental Digital Content is available in the text. Background: Cardiovascular magnetic resonance (CMR) imaging is commonly used to diagnose acute myocarditis. However, the natural history of CMR-based tissue markers and their association with left ventricular recovery is poorly explored. We prospectively investigated the natural history of CMR-based myocardial injury and chamber remodeling over 12 months in patients with suspected acute myocarditis. Methods: One hundred patients with suspected acute myocarditis were enrolled. All underwent CMR evaluations at baseline and 12 months, inclusive of T2 and late gadolinium enhancement. Blinded quantitative analyses compared left ventricular chamber volumes, function, myocardial edema, and necrosis at each time point using predefined criteria. The predefined primary outcomes were improvement in left ventricular ejection fraction ≥10% and improvement in the indexed left ventricular end diastolic volume ≥10% at 12 months. Results: The mean age was 39.9±14.5 years (82 male) with baseline left ventricular ejection fraction of 57.1±11.2%. A total of 72 patients (72%) showed late gadolinium enhancement at baseline with 57 (57%) having any T2 signal elevation. Left ventricular volumes and EF improved significantly at 12 months. Global late gadolinium enhancement extent dropped from 8.5±9.2% of left ventricular mass to 3.0±5.2% (P =0.0001) with prevalence of any late gadolinium enhancement dropping to 48%. Reductions in global T2 signal ratio occurred at 12 months (1.85±0.3 to 1.56±0.2; P =0.0001) with prevalence of T2 ratio ≥2.0 dropping to 7%. Neither marker provided associations with the primary outcomes. Conclusions: In clinically suspected acute myocarditis, significant reductions in tissue injury markers occur during the first 12 months of convalescence. Neither the presence nor extent of the investigated CMR-based tissue injury markers were predictive of our pre-defined function or remodeling outcomes at 12 months in this referral population. [ABSTRACT FROM AUTHOR]