Contemporary outpatient management of patients with worsening heart failure with reduced ejection fraction: Clinical outcome results from the CHART-HF study.

Background: Based on available data from randomized clinical trials, patients with heart failure with reduced ejection fraction (HFrEF) and worsening HF events (WHFE) have substantial disease burden and poor outcomes. WHFE clinical outcome data in non-clinical trial patients, more representative of the US clinical practice, has not been demonstrated. Methods and results: CHART-HF collected data from two complementary, non-clinical trial cohort with HFrEF (LVEF <45 %): 1) 1,000 patients from an integrated delivery network and 2) 458 patients from a nationwide physician panel. CHART-HF included patients with WHFE between 2017 and 2019 followed by an index outpatient cardiology visit ≤6 months, and patients without WHFE in a given year between 2017 and 2019, with the last outpatient cardiology visit in the same year as the index visit. Compared to patients without WHFE (after covariate adjustment, all p < 0.05), patients with WHFE had a greater risk of HF-related hospitalization (hazard ratio [HR]: 1.53-2.40) and next WHFE event (HR: 1.67-2.41) following index visits in both cohorts. Conclusion: HFrEF patients with recent WHFE consistently had worse clinical outcomes in these non-clinical trial cohorts. Despite advances in therapies, unmet need to improve clinical outcomes in HFrEF patients with WHFE remains.

Treatment patterns of patients with worsening heart failure with reduced ejection fraction.

AIMS: Patients with HFrEF and worsening HF events (WHFE) are at particularly high risk and urgently need disease-modifying therapy. CHART-HF assessed treatment patterns and reasons for medication decisions among HFrEF patients with and without WHFE. METHODS AND RESULTS: CHART-HF collected retrospective electronic medical records of outpatients with HF and EF < 45% between 2017-2019 from a nationwide panel of 238 cardiologists (458 patients) and the Geisinger Health System (GHS) medical record (1000 patients). The index visit in the WHFE cohort was the first outpatient cardiologist visit ≤6 months following the WHFE, and in the reference cohort was the last visit in a calendar year without WHFE. Demographic characteristics were similar between patients with and without WHFE in both the nationwide panel and GHS. In the nationwide panel, the proportion of patients with versus without WHFE receiving ≥50% of guideline-recommended dose on index visit was 35% versus 40% for beta blocker, 74% versus 83% for ACEI/ARB/ARNI, and 48% versus 49% for MRA. The proportion of patients receiving ≥50% of guideline-recommended dose was lower in the GHS: 29% versus 34% for beta-blocker, 16% versus 31% for ACEI/ARB/ARNI, and 18% versus 22% for MRA. For patients with and without WHFE, triple therapy on index date was 42% and 44% of patients from the nationwide panel, and 14% and 17% in the GHS. Comparing end of index clinic visit with 12-month follow-up in the GHS, the proportion of patients on no GDMT increased from 14% to 28% in the WHFE cohort and from 14 to 21% in the non-WHFE group. CONCLUSIONS: Major gaps in use of GDMT, particularly combination therapy, remain among US HFrEF patients. These gaps persist during longitudinal follow-up and are particularly large among patients with recent WHFE.

Wild-type Transthyretin Amyloid Deposition in an Ascending Aortic Aneurysm.

Amyloid deposition in aortic tissue is associated with increased stiffness. We report a patient with ascending aortic aneurysm and chronic abdominal aortic dissection who had significant wild-type transthyretin amyloid deposition on surgical pathology. The patient did not have cardiac involvement on further workup.

SUCCESSIVE SUBSPACE LEARNING FOR CARDIAC DISEASE CLASSIFICATION WITH TWO-PHASE DEFORMATION FIELDS FROM CINE MRI.

Cardiac cine magnetic resonance imaging (MRI) has been used to characterize cardiovascular diseases (CVD), often providing a noninvasive phenotyping tool. While recently flourished deep learning based approaches using cine MRI yield accurate characterization results, the performance is often degraded by small training samples. In addition, many deep learning models are deemed a "black box," for which models remain largely elusive in how models yield a prediction and how reliable they are. To alleviate this, this work proposes a lightweight successive subspace learning (SSL) framework for CVD classification, based on an interpretable feedforward design, in conjunction with a cardiac atlas. Specifically, our hierarchical SSL model is based on (i) neighborhood voxel expansion, (ii) unsupervised subspace approximation, (iii) supervised regression, and (iv) multi-level feature integration. In addition, using two-phase 3D deformation fields, including end-diastolic and end-systolic phases, derived between the atlas and individual subjects as input offers objective means of assessing CVD, even with small training samples. We evaluate our framework on the ACDC2017 database, comprising one healthy group and four disease groups. Compared with 3D CNN-based approaches, our framework achieves superior classification performance with 140× fewer parameters, which supports its potential value in clinical use.

Clinical profiling of end-stage heart failure with preserved ejection fraction: The National Readmission Database.

BACKGROUND: End-stage (Stage D) heart failure with preserved ejection fraction (HFpEF) is a poorly characterized syndrome that has heterogeneous underlying pathophysiology. A better characterization of the various clinical profiles of Stage D HFpEF is needed. METHOD: 1066 patients with Stage D HFpEF were selected from National Readmission Database. A Bayesian clustering algorithm based on a Dirichlet process mixture model was implemented. Cox proportional hazard regression model was used to relate the risk of in-hospital mortality with each identified clinical cluster. RESULT: 4 distinct clinical clusters were recognized. Group 1 had a higher prevalence of obesity (84.5%) and sleep disorders (62.0%). Group 2 had a higher prevalence of diabetes mellitus (92%), chronic kidney disease (98.3%), anemia (72.6%), and coronary artery disease (59.0%). Group 3 had a higher prevalence of advanced age (82.1%), hypothyroidism (28.9%), dementia (17.0%), atrial fibrillation (63.8%) and valvular disease (30.5%) and Group 4 had a higher prevalence of liver disease (44.5%), right-sided HF (20.2%) and amyloidosis (4.5%). During 2019, 193 (18.1%) in-hospital mortality events occurred. Considering Group 1 (with mortality rate of 4.1%) as a reference, the hazard ratio of in-hospital mortality was 5.4 [95% confidence interval (CI): 2.2-13.6] for Group 2, 6.4 (95% CI: 2.6-15.8) for Group 3 and 9.1 (95% CI: 3.5-23.8) for Group 4. CONCLUSION: End-stage HFpEF presents with different clinical profiles with varied upstream causes. This may help provide evidence toward the development of targeted therapies.

Performance of a multi-biomarker panel for prediction of cardiovascular event in patients with chronic kidney disease.

BACKGROUND: Patients with chronic kidney disease (CKD) undergoing coronary catheterization are at increased risk of cardiovascular events (CVE). Measuring biomarkers before the procedure may guide clinicians in identifying patients at higher risk of future cardiovascular events. METHODS: In this sub-study the Catheter Sampled Blood Archive in Cardiovascular Diseases (CASABLANCA), 927 patients underwent coronary catheterization and were followed up for two years. Using machine learning algorithm and targeted proteomics from samples of patients with CKD, 4 biomarkers (kidney injury molecule-1, N-terminal pro B-type natriuretic peptide, osteopontin, and tissue inhibitor of metalloproteinase-1) were integrated into a prognostic algorithm to predict CVE. Results from the panel are expressed in a graded fashion (CVE higher risk and lower risk) using a data-driven cutoff optimized for balanced sensitivity and specificity. RESULTS: During the 2-year follow-up, 74 CVE were ascertained. 51 (rate: 51/378 = 13.5%) events occurred in stage 1-2 CKD and 23 (rate: 23/68 = 33.8%) events occurred in stage 3-5 CKD. The C-statistic for predicting 2-years cardiovascular events in all 446 patients was 0.77 (0.72, 0.82). The model was well-calibrated (Hosmer-Lemeshow test p-value >0.40). Considering patients at CVE lower-risk within each CKD staging group as a reference, the hazard ratio (95% confidence interval) of cardiovascular events was 2.82 (1.53, 5.22) for CKD stage 1-2/CVE higher-risk, and 8.32 (1.12, 61.76) for CKD stage 3-5/CVE higher-risk. CONCLUSION: Measuring biomarker panel prior to coronary catheterization may be useful to individualize CVE risk assessment among patients with CKD.

Inflammation across universal definition of heart failure stages: the CASABLANCA study.

AIM: We sought to investigate the association of inflammatory biomarkers with incident heart failure (HF) events in patients at different stages of HF. METHODS AND RESULTS: Overall, 1231 study participants undergoing diagnostic coronary and/or peripheral angiography were categorized by Universal Definition of HF (UDHF) stage A (at risk), stage B (pre-HF), and stages C or D (HF, including end-stage). Twenty-four inflammatory biomarkers were collected prior to angiography and unsupervised machine learning categorized levels of inflammation into three groups (low, medium, and high). Cox proportional hazard regression was implemented to assess the associations of inflammation level with incident HF hospitalization in each UDHF stage. Using machine learning, study participants were grouped into low (n = 443), medium (n = 570) and high inflammation categories (n = 230). Significantly higher concentrations of natriuretic peptide, troponin, and soluble ST2 were observed among those with high inflammation levels (p < 0.001). During 3.7 years of follow-up, 123 (15.1%) HF hospitalizations occurred in stage A/B and 180 (41.8%) HF hospitalizations occurred in stage C/D. In multivariable model considering low inflammation level as a reference, among patients with stage A/B, the hazard ratio (HR) (95% confidence interval [CI]) of incident HF was 2.31 (1.40-3.80) for moderate inflammation level, and 4.16 (2.35-7.37) for high inflammation level. Among patients with stage C/D, the corresponding HR (95% CI) of HF hospitalization was 1.98 (1.28-3.04) for moderate inflammation level, and 2.69 (1.69-4.28) for high inflammation level. CONCLUSION: Patterns of inflammation severity may have differing prognostic meaning across UDHF stages.

Multinational Federated Learning Approach to Train ECG and Echocardiogram Models for Hypertrophic Cardiomyopathy Detection.

BACKGROUND: Novel targeted treatments increase the need for prompt hypertrophic cardiomyopathy (HCM) detection. However, its low prevalence (0.5%) and resemblance to common diseases present challenges that may benefit from automated machine learning-based approaches. We aimed to develop machine learning models to detect HCM and to differentiate it from other cardiac conditions using ECGs and echocardiograms, with robust generalizability across multiple cohorts. METHODS: Single-institution HCM ECG models were trained and validated on external data. Multi-institution models for ECG and echocardiogram were trained on data from 3 academic medical centers in the United States and Japan using a federated learning approach, which enables training on distributed data without data sharing. Models were validated on held-out test sets for each institution and from a fourth academic medical center and were further evaluated for discrimination of HCM from aortic stenosis, hypertension, and cardiac amyloidosis. Last, automated detection was compared with manual interpretation by 3 cardiologists on a data set with a realistic HCM prevalence. RESULTS: We identified 74 376 ECGs for 56 129 patients and 8392 echocardiograms for 6825 patients at the 4 academic medical centers. Although ECG models trained on data from each institution displayed excellent discrimination of HCM on internal test data (C statistics, 0.88-0.93), the generalizability was limited, most notably for a model trained in Japan and tested in the United States (C statistic, 0.79-0.82). When trained in a federated manner, discrimination of HCM was excellent across all institutions (C statistics, 0.90-0.96 and 0.90-0.96 for ECG and echocardiogram model, respectively), including for phenotypic subgroups. The models further discriminated HCM from hypertension, aortic stenosis, and cardiac amyloidosis (C statistics, 0.84, 0.83, and 0.88, respectively, for ECG and 0.93, 0.94, 0.85, respectively, for echocardiogram). Analysis of electrocardiography-echocardiography paired data from 11 823 patients from an external institution indicated a higher sensitivity of automated HCM detection at a given positive predictive value compared with cardiologists (0.98 versus 0.81 at a positive predictive value of 0.01 for ECG and 0.78 versus 0.59 at a positive predictive value of 0.24 for echocardiogram). CONCLUSIONS: Federated learning improved the generalizability of models that use ECGs and echocardiograms to detect and differentiate HCM from other causes of hypertrophy compared with training within a single institution.

Cardiovascular Disease Projections in the United States Based on the 2020 Census Estimates.

BACKGROUND: Understanding trends in cardiovascular (CV) risk factors and CV disease according to age, sex, race, and ethnicity is important for policy planning and public health interventions. OBJECTIVES: The goal of this study was to project the number of people with CV risk factors and disease and further explore sex, race, and ethnical disparities. METHODS: The prevalence of CV risk factors (diabetes mellitus, hypertension, dyslipidemia, and obesity) and CV disease (ischemic heart disease, heart failure, myocardial infarction, and stroke) according to age, sex, race, and ethnicity was estimated by using logistic regression models based on 2013-2018 National Health and Nutrition Examination Survey data and further combining them with 2020 U.S. Census projection counts for years 2025-2060. RESULTS: By the year 2060, compared with the year 2025, the number of people with diabetes mellitus will increase by 39.3% (39.2 million [M] to 54.6M), hypertension by 27.2% (127.8M to 162.5M), dyslipidemia by 27.5% (98.6M to 125.7M), and obesity by 18.3% (106.3M to 125.7M). Concurrently, projected prevalence will similarly increase compared with 2025 for ischemic heart disease by 31.1% (21.9M to 28.7M), heart failure by 33.0% (9.7M to 12.9M), myocardial infarction by 30.1% (12.3M to 16.0M), and stroke by 34.3% (10.8M to 14.5M). Among White individuals, the prevalence of CV risk factors and disease is projected to decrease, whereas significant increases are projected in racial and ethnic minorities. CONCLUSIONS: Large future increases in CV risk factors and CV disease prevalence are projected, disproportionately affecting racial and ethnic minorities. Future health policies and public health efforts should take these results into account to provide quality, affordable, and accessible health care.

Inflammatory biomarkers and risk of cardiovascular events in patients undergoing coronary angiography.

BACKGROUND: Inflammation, measured by traditional biomarkers such as C-reactive protein, has been linked to cardiovascular (CV) events. Recent technological advancement has allowed for measuring larger numbers of inflammatory biomarkers. A contemporary evaluation with established and novel biomarkers of inflammation is needed. METHODS: 1,090 individuals who underwent coronary angiography were enrolled. Twenty-four inflammatory biomarkers were collected prior to angiography. Unsupervised machine learning cluster analyses determined unique patterns of inflammatory biomarkers. Cox proportional hazard regression assessed both association of inflammatory biomarker clusters and individual biomarker associations with major adverse cardiovascular events (MACE; non-fatal myocardial infarction or stroke, and CV death) during a median follow-up of 3.67 years. RESULTS: Four distinct clusters were recognized. Incremental increases in inflammatory biomarkers were observed from cluster 1 to cluster 4. During follow-up, 263 MACE were ascertained. Considering cluster 1 as a reference, study participants with inflammatory cluster 2 (Hazard ratio [HR] 1.55, 95% confidence interval [CI]: 1.01-2.37), cluster 3 (HR 1.89, CI: 1.25-2.85), and cluster 4 (HR 2.93, CI: 1.95-4.42) were at increased risk of MACE. Interleukin (IL)-1α IL-6, IL-8, IL-10, IL-12, Adhesion molecule-1 high-sensitivity C-reactive protein, ferritin, myeloperoxidase, macrophage inflammatory protein (MIP)-1a, MIP 3, and macrophage colony-stimulating factor-1 were independently associated with MACE. CONCLUSIONS: Among persons undergoing coronary angiography procedures, distinct clusters of inflammatory biomarker distributions with significant prognostic meaning may be identified. These results may identify unique targets for anti-inflammatory treatments aimed at CV disease.

Development and validation of a decision support tool for the diagnosis of acute heart failure: systematic review, meta-analysis, and modelling study.

OBJECTIVES: To evaluate the diagnostic performance of N-terminal pro-B-type natriuretic peptide (NT-proBNP) thresholds for acute heart failure and to develop and validate a decision support tool that combines NT-proBNP concentrations with clinical characteristics. DESIGN: Individual patient level data meta-analysis and modelling study. SETTING: Fourteen studies from 13 countries, including randomised controlled trials and prospective observational studies. PARTICIPANTS: Individual patient level data for 10 369 patients with suspected acute heart failure were pooled for the meta-analysis to evaluate NT-proBNP thresholds. A decision support tool (Collaboration for the Diagnosis and Evaluation of Heart Failure (CoDE-HF)) that combines NT-proBNP with clinical variables to report the probability of acute heart failure for an individual patient was developed and validated. MAIN OUTCOME MEASURE: Adjudicated diagnosis of acute heart failure. RESULTS: Overall, 43.9% (4549/10 369) of patients had an adjudicated diagnosis of acute heart failure (73.3% (2286/3119) and 29.0% (1802/6208) in those with and without previous heart failure, respectively). The negative predictive value of the guideline recommended rule-out threshold of 300 pg/mL was 94.6% (95% confidence interval 91.9% to 96.4%); despite use of age specific rule-in thresholds, the positive predictive value varied at 61.0% (55.3% to 66.4%), 73.5% (62.3% to 82.3%), and 80.2% (70.9% to 87.1%), in patients aged <50 years, 50-75 years, and >75 years, respectively. Performance varied in most subgroups, particularly patients with obesity, renal impairment, or previous heart failure. CoDE-HF was well calibrated, with excellent discrimination in patients with and without previous heart failure (area under the receiver operator curve 0.846 (0.830 to 0.862) and 0.925 (0.919 to 0.932) and Brier scores of 0.130 and 0.099, respectively). In patients without previous heart failure, the diagnostic performance was consistent across all subgroups, with 40.3% (2502/6208) identified at low probability (negative predictive value of 98.6%, 97.8% to 99.1%) and 28.0% (1737/6208) at high probability (positive predictive value of 75.0%, 65.7% to 82.5%) of having acute heart failure. CONCLUSIONS: In an international, collaborative evaluation of the diagnostic performance of NT-proBNP, guideline recommended thresholds to diagnose acute heart failure varied substantially in important patient subgroups. The CoDE-HF decision support tool incorporating NT-proBNP as a continuous measure and other clinical variables provides a more consistent, accurate, and individualised approach. STUDY REGISTRATION: PROSPERO CRD42019159407.

Contemporary outpatient management of patients with worsening heart failure with reduced ejection fraction: Rationale and design of the CHART-HF study.

BACKGROUND: Patients with heart failure with reduced ejection fraction (HFrEF) and worsening HF events (WHFE) represent a distinct subset of patients with a substantial comorbidity burden, greater potential for intolerance to medical therapy, and high risk of subsequent death, hospitalization and excessive healthcare costs. Although multiple therapies have been shown to be efficacious and safe in this high-risk population, there are limited real-world data regarding factors that impact clinical decision-making when initiating or modifying therapy. Likewise, prior analyses of US clinical practice support major gaps in medical therapy for HFrEF and few medication changes during longitudinal follow-up, yet granular data on reasons why clinicians do not initiate or up-titrate guideline-directed medication are lacking. METHODS: We designed the CHART-HF study, an observational study of approximately 1,500 patients comparing patients with and without WHFE (WHFE defined as receipt of intravenous diuretics in the inpatient, outpatient, or emergency department setting) who had an index outpatient visit in the US between 2017 and 2019. Patient-level data on clinical characteristics, clinical outcomes, and therapy will be collected from 2 data sources: a single integrated health system, and a national panel of cardiologists. Furthermore, clinician-reported rationale for treatment decisions and the factors prioritized with selection and optimization of therapies in real-world practice will be obtained. To characterize elements of clinician decision-making not documented in the medical record, the panel of cardiologists will review records of patients seen under their care to explicitly note their primary reason for initiating, discontinuing, and titrating medications specific medications, as well as the reason for not making changes to each medication during the outpatient visit. CONCLUSIONS: Results from CHART-HF have the potential to detail real-world US practice patterns regarding care of patients with HFrEF with versus without a recent WHFE, to examine clinician-reported reasons for use and non-use of guideline-directed medical therapy, and to characterize the magnitude and nature of clinical inertia toward evidence-based medication changes for HFrEF.

Circulating levels and prognostic cut-offs of sST2, hs-cTnT, and NT-proBNP in women vs. men with chronic heart failure.

AIMS: To define plasma concentrations, determinants, and optimal prognostic cut-offs of soluble suppression of tumorigenesis-2 (sST2), high-sensitivity cardiac troponin T (hs-cTnT), and N-terminal pro-B-type natriuretic peptide (NT-proBNP) in women and men with chronic heart failure (HF). METHODS AND RESULTS: Individual data of patients from the Biomarkers In Heart Failure Outpatient Study (BIOS) Consortium with sST2, hs-cTnT, and NT-proBNP measured were analysed. The primary endpoint was a composite of 1 year cardiovascular death and HF hospitalization. The secondary endpoints were 5 year cardiovascular and all-cause death. The cohort included 4540 patients (age 67 ± 12 years, left ventricular ejection fraction 33 ± 13%, 1111 women, 25%). Women showed lower sST2 (24 vs. 27 ng/mL, P < 0.001) and hs-cTnT level (15 vs. 20 ng/L, P < 0.001), and similar concentrations of NT-proBNP (1540 vs. 1505 ng/L, P = 0.408). Although the three biomarkers were confirmed as independent predictors of outcome in both sexes, the optimal prognostic cut-off was lower in women for sST2 (28 vs. 31 ng/mL) and hs-cTnT (22 vs. 25 ng/L), while NT-proBNP cut-off was higher in women (2339 ng/L vs. 2145 ng/L). The use of sex-specific cut-offs improved risk prediction compared with the use of previously standardized prognostic cut-offs and allowed to reclassify the risk of many patients, to a greater extent in women than men, and for hs-cTnT than sST2 or NT-proBNP. Specifically, up to 18% men and up to 57% women were reclassified, by using the sex-specific cut-off of hs-cTnT for the endpoint of 5 year cardiovascular death. CONCLUSIONS: In patients with chronic HF, concentrations of sST2 and hs-cTnT, but not of NT-proBNP, are lower in women. Lower sST2 and hs-cTnT and higher NT-proBNP cut-offs for risk stratification could be used in women.

Relation of High-Sensitivity Cardiac Troponin I and Obstructive Coronary Artery Disease in Patients Without Acute Myocardial Infarction.

The relation of high-sensitivity cardiac troponin I (hs-cTnI) concentration and presence of obstructive coronary artery disease (CAD) in patients without myocardial infarction (MI) is unclear. Study participants selected from patients free of MI who underwent coronary angiography with or without intervention were enrolled, and hs-cTnI measured. A gradient boosting model was implemented to build a model for detection of CAD. Cox proportional hazard regression was used to assess the association of hs-cTnI and adverse cardiovascular (CV) outcome. Among 978 study participants, 607 patients (62%) had CAD. Higher concentrations of hs-cTnI were associated with chronic kidney disease, heart failure, CAD, male gender, current tobacco use, anemia, age, and low-density lipoprotein cholesterol. History of CAD, male gender, type 2 diabetes mellitus, hs-cTnI, anemia, age, and high-density lipoprotein cholesterol were the most influential factors for detection of CAD. The gradient boosting model had an area under the curve of 0.82, accuracy of 75%, sensitivity of 88%, specificity of 52%, positive predictive value of 76%, and negative predictive value of 72% for detection of CAD. Increase in 1 log unit of hs-cTnI was significantly associated with increased risk of incident MI (hazard ratio [HR] 1.34, 95% confidence interval [CI] 1.22 to 1.47, p <0.001), CV mortality (HR 1. 24, 95% CI 1.11 to 1.39, p <0.001), and composite of incident MI or CV mortality (HR 1.29, 95% CI 1.20 to 1.40, p <0.001). In conclusion, among patients without acute MI and CAD, higher concentrations of hs-cTnI were associated with the presence of CAD and linked to increased risk of future CV events. ClinicalTrials.gov Identifier: NCT00842868.

Multiple Cardiac Biomarker Testing Among Patients With Acute Dyspnea From the ICON-RELOADED Study.

BACKGROUND: Among patients with acute dyspnea, concentrations of N-terminal pro-B-type natriuretic peptide (NT-proBNP), high-sensitivity cardiac troponin T, and insulin-like growth factor binding protein-7 predict cardiovascular outcomes and death. Understanding the optimal means to interpret these elevated biomarkers in patients presenting with acute dyspnea remains unknown. METHODS AND RESULTS: Concentrations of NT-proBNP, high-sensitivity cardiac troponin T, and insulin-like growth factor binding protein-7 were analyzed in 1448 patients presenting with acute dyspnea from the prospective, multicenter International Collaborative of NT-proBNP-Re-evaluation of Acute Diagnostic Cut-Offs in the Emergency Department (ICON-RELOADED) Study. Eight biogroups were derived based upon patterns in biomarker elevation at presentation and compared for differences in baseline characteristics. Of 441 patients with elevations in all 3 biomarkers, 218 (49.4%) were diagnosed with acute heart failure (HF). The frequency of acute HF diagnosis in this biogroup was higher than those with elevations in 2 biomarkers (18.8%, 44 of 234), 1 biomarker (3.8%, 10 of 260), or no elevated biomarkers (0.4%, 2 of 513). The absolute number of elevated biomarkers on admission was prognostic of the composite end point of mortality and HF rehospitalization. In adjusted models, patients with one, 2, and 3 elevated biomarkers had 3.74 (95% confidence interval [CI], 1.26-11.1, P = .017), 12.3 (95% CI, 4.60-32.9, P < .001), and 12.6 (95% CI, 4.54-35.0, P < .001) fold increased risk of 180-day mortality or HF rehospitalization. CONCLUSIONS: A multimarker panel of NT-proBNP, hsTnT, and IGBFP7 provides unique clinical, diagnostic, and prognostic information in patients presenting with acute dyspnea. Differences in the number of elevated biomarkers at presentation may allow for more efficient clinical risk stratification of short-term mortality and HF rehospitalization.

Segmentation of Cardiac Structures via Successive Subspace Learning with Saab Transform from Cine MRI.

Assessment of cardiovascular disease (CVD) with cine magnetic resonance imaging (MRI) has been used to non-invasively evaluate detailed cardiac structure and function. Accurate segmentation of cardiac structures from cine MRI is a crucial step for early diagnosis and prognosis of CVD, and has been greatly improved with convolutional neural networks (CNN). There, however, are a number of limitations identified in CNN models, such as limited interpretability and high complexity, thus limiting their use in clinical practice. In this work, to address the limitations, we propose a lightweight and interpretable machine learning model, successive subspace learning with the subspace approximation with adjusted bias (Saab) transform, for accurate and efficient segmentation from cine MRI. Specifically, our segmentation framework is comprised of the following steps: (1) sequential expansion of near-to-far neighborhood at different resolutions; (2) channel-wise subspace approximation using the Saab transform for unsupervised dimension reduction; (3) class-wise entropy guided feature selection for supervised dimension reduction; (4) concatenation of features and pixel-wise classification with gradient boost; and (5) conditional random field for post-processing. Experimental results on the ACDC 2017 segmentation database, showed that our framework performed better than state-of-the-art U-Net models with 200× fewer parameters in delineating the left ventricle, right ventricle, and myocardium, thus showing its potential to be used in clinical practice.Clinical relevance- Delineation of the left ventricular cavity, myocardium, and right ventricle from cardiac MR images is a common clinical task to establish diagnosis and prognosis of CVD.