Machine learning–based biomarker profile derived from 4210 serially measured proteins predicts clinical outcome of patients with heart failure
Ontology highlight
ABSTRACT:
SUBMITTER:
PROVIDER: S-EPMC10689916 | biostudies-literature | 2023 Oct
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Aim The severity of cardiac impairment in acute heart failure (AHF) predicts outcome, but challenges remain to identify prognostically important non-invasive parameters of cardiac function. Left ventricular ejection fraction (LVEF) is relevant, but only in those with reduced LV systolic function. We aimed to assess the standard and advanced parameters of left and right ventricular (RV) function from echocardiography in predicting long-term outcomes in AHF. Methods A total of 418 consecutive AHF patients presenting over 12 months were prospectively recruited and underwent bedside echocardiography within 24 h of recruitment. We retrospectively assessed 8 RV and 5 LV echo parameters of the cardiac systolic function to predict 2-year mortality, using both guideline-directed and study-specific cutoffs, based on the maximum Youden indices via ROC analysis. For the RV, these were the tricuspid annular plane systolic excursion, RV fractional area change, tissue Doppler imaging (TDI) peak tricuspid annular systolic wave velocity, both peak- and end-systolic RV free wall global longitudinal strain (RV GLS) and strain rate (mean RV GLSR), RV ejection fraction (RVEF) derived from a 2D ellipsoid model and the ratio of the TAPSE to systolic pulmonary artery pressure (SPAP). For the LV, these were the LVEF, mitral regurgitant ΔP/Δt (MR dP/dt), the lateral mitral annular TDI peak systolic wave velocity, LV GLS, and the LV GLSR. Results A total of 7/8 parameters of RV systolic function were predictive of 2-year outcome, with study cutoffs like international guidelines. A cutoff of < −1.8 s–1 mean RV GLSR was associated with worse outcome compared to > −1.8 s–1 [HR 2.13 95% CI 1.33–3.40 (p = 0.002)]. TAPSE:SPAP of > 0.027 cm/mmHg (vs. < 0.027 cm/mmHg) predicted worse outcome [HR 2.12 95% CI 1.53–2.92 (p < 0.001)]. A 3-way comparison of 2-year mortality by LVEF from the European Society of Cardiology (ESC) guideline criteria of LVEF > 50, 41–49, and < 40% was not prognostic [38.6% vs. 30.9 vs. 43.9% (p = 0.10)]. Of the 5 parameters of LV systolic function, only an MR dP/dt cutoff of < 570 mmHg was predictive of adverse outcome [HR 1.63 95% CI 1.01–2.62 (p = 0.047)]. Conclusion With cutoffs broadly like the ESC guidelines, we identified RV dysfunction to be associated with adverse prognosis, whereas LVEF could not identify patients at risk.
| S-EPMC9157539 | biostudies-literature
Project description:AimsSurveillance imaging is often used to detect remodelling, a change in cardiac geometry, and/or function; however, there are limited data in patients with chronic heart failure (HF). We sought to characterize cardiac remodelling in patients with chronic HF and evaluate its association with outcome.Methods and resultsA prospective cohort of patients at risk for HF or with chronic HF underwent cardiac magnetic resonance (CMR) at baseline and 1 year. Ventricular function, volumes, mass, left atrial volume, global longitudinal strain, and myocardial scar were measured. The primary outcome was a composite of death or cardiovascular hospitalization up to 5 years from the 1 year scan. Cox regression was used to identify 1 year CMR predictors of outcome after adjusting for baseline risk. A total of 262 patients (median age 68 years, 57% males) including 96 at risk for HF, 97 with HF and preserved ejection fraction, and 69 with HF and reduced ejection fraction were included. In the patients with HF, 55 events were identified during follow-up. After adjustment for baseline clinical risk, Cox proportion hazard regressions only identified 1 year change in left ventricular (LV) mass index as a CMR predictor of outcome, adjusted hazard ratio 1.21 (1.02, 1.44) per 10% increase, P = 0.031. Cardiac remodelling defined as a 1 year change in LV mass index ≥15% was observed in 35% of patients with HF. Patients with adverse remodelling of LV mass index had more events on Kaplan-Meier analyses compared to those with no remodelling, log-rank P = 0.004 for overall cohort, P = 0.035 for heart failure with preserved ejection fraction and P = 0.035 for heart failure and reduced ejection fraction.ConclusionsCardiac remodelling is common during serial CMR assessment of patients with chronic HF. Change in LV mass predicted long-term outcomes whereas change in left ventricular ejection fraction did not.
| S-EPMC8712825 | biostudies-literature
Project description:The perception of acute heart failure (AHF) as a single entity is increasingly outdated, as distinct patient profiles can be discerned. Key heart failure (HF) studies have previously highlighted the difference in both the course and prognosis of de novo AHF and acute decompensated chronic HF (ADHF). Accordingly, distinct AHF profiles with differing underlying pathophysiologies of disease progression can be shown. We compared a range of selected biomarkers in order to better describe the profile of de novo AHF and ADHF, including the inter alia-serum lactate, bilirubin, matrix metallopeptidase 9 (MMP-9), follistatin, intercellular adhesion molecule 1 (ICAM-1), lipocalin and galectin-3. The study comprised 248 AHF patients (de novo = 104), who were followed up for one year. The biomarker data of the de novo AHF and ADHF profiles was then compared in order to link biomarkers to their prognosis. Our study demonstrated that, although there are similarities between each patient profile, key biomarker differences do exist-predominantly in terms of NTproBNP, serum lactate, bilirubin, ICAM-1, follistatin, ferritin and sTfR (soluble transferrin receptor). ADHF tended to have compromised organ function and higher risks of both one-year mortality and composite endpoint (one-year mortality or rehospitalization for heart failure) hazard ratios (HR) (95% CI): 3.4 (1.8-6.3) and 2.8 (1.6-4.6), respectively, both p < 0.0001. Among the biomarkers of interest: sTfR HR (95% CI): 1.4 (1.04-1.8), NGAL(log) (neutrophil gelatinase-associated lipocalin) HR (95% CI): 2.0 (1.3-3.1) and GDF-15(log) (growth/differentiation factor-15) HR (95% CI): 4.0 (1.2-13.0) significantly impacted the one-year survival, all p < 0.05.
| S-EPMC8615401 | biostudies-literature
Project description:BackgroundIt remains unclear whether the plasma proteome adds value to established predictors in heart failure (HF) with reduced ejection fraction (HFrEF). We sought to derive and validate a plasma proteomic risk score (PRS) for survival in patients with HFrEF (HFrEF-PRS).MethodsPatients meeting Framingham criteria for HF with EF<50% were enrolled (N=1017) and plasma underwent SOMAscan profiling (4453 targets). Patients were randomly divided 2:1 into derivation and validation cohorts. The HFrEF-PRS was derived using Cox regression of all-cause mortality adjusted for clinical score and NT-proBNP (N-terminal pro-B-type natriuretic peptide), then was tested in the validation cohort. Risk stratification improvement was evaluated by C statistic, integrated discrimination index, continuous net reclassification index, and median improvement in risk score for 1-year and 3-year mortality.ResultsParticipants' mean age was 68 years, 48% identified as Black, 35% were female, and 296 deaths occurred. In derivation (n=681), 128 proteins associated with mortality, 8 comprising the optimized HFrEF-PRS. In validation (n=336) the HFrEF-PRS associated with mortality (hazard ratio, 2.27 [95% CI, 1.84-2.82], P=6.3×10-14), Kaplan-Meier curves differed significantly between HFrEF-PRS quartiles (P=2.2×10-6), and it remained significant after adjustment for clinical score and NT-proBNP (hazard ratio, 1.37 [95% CI, 1.05-1.79], P=0.021). The HFrEF-PRS improved metrics of risk stratification (C statistic change, 0.009, P=0.612; integrated discrimination index, 0.041, P=0.010; net reclassification index=0.391, P=0.078; median improvement in risk score=0.039, P=0.016) and associated with cardiovascular death and HF phenotypes (eg, 6-minute walk distance, EF change). Most HFrEF-PRS proteins had little known connection to HFrEF.ConclusionsA plasma multiprotein score improved risk stratification in patients with HFrEF and identified novel candidates.
| S-EPMC8221080 | biostudies-literature
Project description:AimsThe development of acute heart failure (AHF) is a critical decision point in the natural history of the disease and carries a dismal prognosis. The lack of appropriate risk-stratification tools at hospital discharge of AHF patients significantly limits clinical ability to precisely tailor patient-specific therapeutic regimen at this pivotal juncture. Machine learning-based strategies may improve risk stratification by incorporating analysis of high-dimensional patient data with multiple covariates and novel prediction methodologies. In the current study, we aimed at evaluating the drivers for success in prediction models and establishing an institute-tailored artificial Intelligence-based prediction model for real-time decision support.Methods and resultsWe used a cohort of all 10 868 patients AHF patients admitted to a tertiary hospital during a 12 years period. A total of 372 covariates were collected from admission to the end of the hospitalization. We assessed model performance across two axes: (i) type of prediction method and (ii) type and number of covariates. The primary outcome was 1-year survival from hospital discharge. For the model-type axis, we experimented with seven different methods: logistic regression (LR) with either L1 or L2 regularization, random forest (RF), Cox proportional hazards model (Cox), extreme gradient boosting (XGBoost), a deep neural-net (NeuralNet) and an ensemble classifier of all the above methods. We were able to achieve an area under receiver operator curve (AUROC) prediction accuracy of more than 80% with most prediction models including L1/L2-LR (80.4%/80.3%), Cox (80.2%), XGBoost (80.5%), NeuralNet (80.4%). RF was inferior to other methods (78.8%), and the ensemble model was slightly superior (81.2%). The number of covariates was a significant modifier (P < 0.001) of prediction success, the use of multiplex-covariates preformed significantly better (AUROC 80.4% for L1-LR) compared with a set of known clinical covariates (AUROC 77.8%). Demographics followed by lab-tests and administrative data resulted in the largest gain in model performance.ConclusionsThe choice of the predictive modelling method is secondary to the multiplicity and type of covariates for predicting AHF prognosis. The application of a structured data pre-processing combined with the use of multiple-covariates results in an accurate, institute-tailored risk prediction in AHF.
| S-EPMC10232285 | biostudies-literature
Project description:Natriuretic peptides (BNP and NT-proBNP) are recognized as gold-standard predictive markers in Heart Failure (HF). However, currently ST2 (member of the interleukin 1 receptor family) has emerged as marker of inflammation, fibrosis and cardiac stress. We evaluated ST2 and CRP as prognostic markers in 178 patients with chronic heart failure in comparison with other classical markers such as clinical established parameters but also biological markers: NT-proBNP, hs-cTnT alone or in combination. In multivariate analysis, subsequent addition of ST2 led to age, CRP and ST2 as the only remaining predictors of all-cause mortality (HR 1.03, HR 1.61 and HR 2.75, respectively) as well as of cardiovascular mortality (HR 1.00, HR 2.27 and HR 3.78, respectively). The combined increase of ST2 and CRP was significant for predicting worsened outcomes leading to identify a high risk subgroup that individual assessment of either marker. The same analysis was performed with ST2 in combination with Barcelona score. Overall, our findings extend previous data demonstrating that ST2 in combination with CRP as a valuable tool for identifying patients at risk of death.
| S-EPMC4911159 | biostudies-literature
Project description:High mortality and morbidity rates associated with ST-elevation myocardial infarction (STEMI) and post-STEMI heart failure (HF) necessitate proper risk stratification for coronary artery disease (CAD). A prediction model that combines specificity and convenience is highly required. This study aimed to design a monocyte-based gene assay for predicting STEMI and post-STEMI HF. A total of 1,956 monocyte expression profiles and corresponding clinical data were integrated from multiple sources. Meta-results were obtained through the weighted gene co-expression network analysis (WGCNA) and differential analysis to identify characteristic genes for STEMI. Machine learning models based on the decision tree (DT), support vector machine (SVM), and random forest (RF) algorithms were trained and validated. Five genes overlapped and were subjected to the model proposal. The discriminative performance of the DT model outperformed the other two methods. The established four-gene panel (HLA-J, CFP, STX11, and NFYC) could discriminate STEMI and HF with an area under the curve (AUC) of 0.86 or above. In the gene set enrichment analysis (GSEA), several cardiac pathogenesis pathways and cardiovascular disorder signatures showed statistically significant, concordant differences between subjects with high and low expression levels of the four-gene panel, affirming the validity of the established model. In conclusion, we have developed and validated a model that offers the hope for accurately predicting the risk of STEMI and HF, leading to optimal risk stratification and personalized management of CAD, thereby improving individual outcomes.
| S-EPMC10950350 | biostudies-literature
Project description:AimsThe lack of effective therapies for patients with heart failure with preserved ejection fraction (HFpEF) is often ascribed to the heterogeneity of patients with HFpEF. We aimed to identify distinct pathophysiologic clusters of HFpEF based on circulating biomarkers.Methods and resultsWe performed an unsupervised cluster analysis using 363 biomarkers from 429 patients with HFpEF. Relative differences in expression profiles of the biomarkers between clusters were assessed and used for pathway over-representation analyses. We identified four distinct patient subgroups based on their biomarker profiles: cluster 1 with the highest prevalence of diabetes mellitus and renal disease; cluster 2 with oldest age and frequent age-related comorbidities; cluster 3 with youngest age, largest body size, least symptoms and lowest N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels; and cluster 4 with highest prevalence of ischaemic aetiology, smoking and chronic lung disease, most symptoms, as well as highest NT-proBNP and troponin levels. Over a median follow-up of 21 months, the occurrence of death or heart failure hospitalization was highest in clusters 1 and 4 (62.1% and 62.8%, respectively) and lowest in cluster 3 (25.6%). Pathway over-representation analyses revealed that the biomarker profile of patients in cluster 1 was associated with activation of inflammatory pathways while the biomarker profile of patients in cluster 4 was specifically associated with pathways implicated in cell proliferation regulation and cell survival.ConclusionUnsupervised cluster analysis based on biomarker profiles identified mutually exclusive subgroups of patients with HFpEF with distinct biomarker profiles, clinical characteristics and outcomes, suggesting different underlying pathophysiological pathways.
| S-EPMC8360080 | biostudies-literature
Project description:BackgroundWe evaluated whether cardiac troponin T (cTnT) measured with a new highly sensitive assay was associated with incident coronary heart disease (CHD), mortality, and hospitalization for heart failure (HF) in a general population of participants in the Atherosclerosis Risk in Communities (ARIC) Study.Methods and resultsAssociations between increasing cTnT levels and CHD, mortality, and HF hospitalization were evaluated with Cox proportional hazards models adjusted for traditional CHD risk factors, kidney function, high-sensitivity C-reactive protein, and N-terminal pro-B-type natriuretic peptide in 9698 participants aged 54 to 74 years who at baseline were free from CHD and stroke (and HF in the HF analysis). Measurable cTnT levels (≥0.003 μg/L) were detected in 66.5% of individuals. In fully adjusted models, compared with participants with undetectable levels, those with cTnT levels in the highest category (≥0.014 μg/L; 7.4% of the ARIC population) had significantly increased risk for CHD (hazard ratio=2.29; 95% confidence interval, 1.81 to 2.89), fatal CHD (hazard ratio=7.59; 95% confidence interval, 3.78 to 15.25), total mortality (hazard ratio=3.96; 95% confidence interval, 3.21 to 4.88), and HF (hazard ratio=5.95; 95% confidence interval, 4.47 to 7.92). Even minimally elevated cTnT (≥0.003 μg/L) was associated with increased risk for mortality and HF (P<0.05). Adding cTnT to traditional risk factors improved risk prediction parameters; the improvements were similar to those with N-terminal pro-B-type natriuretic peptide and better than those with the addition of high-sensitivity C-reactive protein.ConclusionscTnT detectable with a highly sensitive assay was associated with incident CHD, mortality, and HF in individuals from a general population without known CHD/stroke.
| S-EPMC3072024 | biostudies-literature
Project description:AimsSevere deficiency of the essential trace element selenium can cause myocardial dysfunction although the mechanism at cellular level is uncertain. Whether, in clinical practice, moderate selenium deficiency is associated with worse symptoms and outcome in patients with heart failure is unknown.Methods and resultsBIOSTAT-CHF is a multinational, prospective, observational cohort study that enrolled patients with worsening heart failure. Serum concentrations of selenium were measured by inductively coupled plasma mass spectrometry. Primary endpoint was a composite of all-cause mortality and hospitalization for heart failure; secondary endpoint was all-cause mortality. To investigate potential mechanisms by which selenium deficiency might affect prognosis, human cardiomyocytes were cultured in absence of selenium, and mitochondrial function and oxidative stress were assessed. Serum selenium concentration (deficiency) was <70 μg/L in 485 (20.4%) patients, who were older, more often women, had worse New York Heart Association class, more severe signs and symptoms of heart failure and poorer exercise capacity (6-min walking test) and quality of life (Kansas City Cardiomyopathy Questionnaire). Selenium deficiency was associated with higher rates of the primary endpoint [hazard ratio (HR) 1.23; 95% confidence interval (CI) 1.06-1.42] and all-cause mortality (HR 1.52; 95% CI 1.26-1.86). In cultured human cardiomyocytes, selenium deprivation impaired mitochondrial function and oxidative phosphorylation, and increased intracellular reactive oxygen species levels.ConclusionsSelenium deficiency in heart failure patients is independently associated with impaired exercise tolerance and a 50% higher mortality rate, and impaired mitochondrial function in vitro, in human cardiomyocytes. Clinical trials are needed to investigate the effect of selenium supplements in patients with heart failure, especially if they have low plasma concentrations of selenium.
| S-EPMC7540257 | biostudies-literature