FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2913887066> ?p ?o ?g. }

• W2913887066 endingPage "381" @default.
• W2913887066 startingPage "379" @default.
• W2913887066 abstract "Heart failure (HF) is a leading cause for hospitalization in Canada and globally. Readmission rates following HF hospitalization remain upward of 20% within 30 days and 50% within 6 months.1Virani S.A. Bains M. Code J. et al.The need for heart failure advocacy in Canada.Can J Cardiol. 2017; 33: 1450-1454Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar, 2Desai A.S. Stevenson L.W. Rehospitalization for heart failure: predict or prevent?.Circulation. 2012; 126: 501-506Crossref PubMed Scopus (425) Google Scholar Although predicting the need for HF hospitalization (including readmission) is recognized as a health care priority, developing accurate predictive tools to identify at-risk patients has remained an elusive challenge. Such analytics, if sufficiently reliable, could identify patients at risk and allow the deployment of preventive interventions to avert the need for hospitalization. Hospitalization for HF is typically preceded by a gradual rise in ventricular filling pressures.2Desai A.S. Stevenson L.W. Rehospitalization for heart failure: predict or prevent?.Circulation. 2012; 126: 501-506Crossref PubMed Scopus (425) Google Scholar As these hemodynamic changes often precede the onset of overt clinical manifestations by days to weeks,3Zile M.R. Bennett T.D. St John Sutton M. et al.Transition from chronic compensated to acute decompensated heart failure: pathophysiological insights obtained from continuous monitoring of intracardiac pressures.Circulation. 2008; 118: 1433-1441Crossref PubMed Scopus (377) Google Scholar detecting rising filling pressures has been the focus of many predictive approaches to date. Such approaches include implantable hemodynamic monitoring devices4Bohm M. Drexler H. Oswald H. et al.Fluid status telemedicine alerts for heart failure: a randomized controlled trial.Eur Heart J. 2016; 37: 3154-3163Crossref PubMed Scopus (139) Google Scholar, 5Bourge R.C. Abraham W.T. Adamson P.B. et al.Randomized controlled trial of an implantable continuous hemodynamic monitor in patients with advanced heart failure: the COMPASS-HF study.J Am Coll Cardiol. 2008; 51: 1073-1079Crossref PubMed Scopus (407) Google Scholar, 6Abraham W.T. Adamson P.B. Bourge R.C. et al.Wireless pulmonary artery haemodynamic monitoring in chronic heart failure: a randomised controlled trial.Lancet. 2011; 377: 658-666Abstract Full Text Full Text PDF PubMed Scopus (1058) Google Scholar, 7Adamson P.B. Abraham W.T. Stevenson L.W. et al.Pulmonary artery pressure-guided heart failure management reduces 30-day readmissions.Circ Heart Failure. 2016; 9e002600Crossref Scopus (60) Google Scholar and blood-based biomarkers of congestion (NT-proBNP, troponin, and soluble ST2).8Demissei B.G. Cotter G. Prescott M.F. et al.A multimarker multi-timer point-based risk stratification strategy in acute heart failure: results from the RELAX-AHF trial.Eur J Heart Fail. 2017; 19: 1001-1010Crossref PubMed Scopus (67) Google Scholar, 9Stiensen S. Salah K. Moons A.H. et al.NT-proBNP (N-Terminal pro-B-Type Natriuretic Peptide)-guided therapy in acute decompensated heart failure: PRIMA II randomized controlled trial (Can NT-ProBNP-Guided therapy during hospital admission for acute decompensated heart failure reduce mortality and readmissions?).Circulation. 2018; 137: 1671-1683Google Scholar, 10Braga J.R. Tu J.V. Austin P.C. et al.Outcomes and care of patients with acute heart failure syndromes and cardiac troponin elevation.Circ Heart Failure. 2013; 6: 193-202Crossref PubMed Scopus (43) Google Scholar, 11Felker G.M. Fiuzat M. Thompson V. et al.Soluble ST2 in ambulatory patients with heart failure: association with functional capacity and long-term outcomes.Circ Heart Failure. 2013; 6: 1172-1179Crossref PubMed Scopus (111) Google Scholar Clinical variables may be predictive. For example, hospital length of stay may be predictive, given that shorter lengths of stay are associated with increased rates of readmission for HF (perhaps reflecting inadequate decongestion prior to discharge).12Sud M. Yu B. Wijeysundera H.C. et al.Association between short or long length of stay and 30-day readmission and mortality in hospitalized patients with heart failure.JACC Heart Failure. 2017; 5: 578-588Crossref PubMed Scopus (77) Google Scholar Electrocardiographic (ECG) changes suggesting myocardial ischemia are also associated with an increase in short-term adverse events in patients with HF.13Greig D. Austin P.C. Zhou L. et al.Ischemic electrocardiographic abnormalities and prognosis in decompensated heart failure.Circ Heart Failure. 2014; 7: 986-993Crossref PubMed Scopus (18) Google Scholar Although offering promise, many of these approaches have potential challenges, and the search for a clinically reliable predictive approach remains ongoing. A potential limitation of such approaches may be that increases in ventricular filling pressures, although preceding the onset of clinical symptoms, are nonetheless consequences and not causes of underlying progression and destabilization of HF. It may be that in patients in whom ventricular pressures have already started to rise, preventive interventions may act too late. What is missing from current approaches is a mechanistic insight into why some patients remain stable over time, whereas others decompensate. Moving further upstream in the trajectory of destabilization to identify causes (and not consequences) of disease destabilization may offer greater potential to identify predictors— and perhaps mediators—of HF stability in an individual patient. With parallel revolutions in high-dimensional molecular and clinical diagnostics; detailed remote patient monitoring; and advanced predictive analytics, the promise of more individualized, cause-oriented approaches to risk prediction that identify more “upstream” markers and mediators of HF stability may be closer at hand. High-dimensional biomarker approaches using peripheral blood samples may support such efforts and can involve profiling the complement of genes (genomics), messenger RNA (transcriptomics), proteins (proteomics), and metabolites (metabolomics) in concert.14Leopold J.A. Loscalzo J. Emerging role of precision medicine in cardiovascular disease.Circ Res. 2018; 122: 1302-1315Crossref PubMed Scopus (142) Google Scholar The dimensionality of such data can grow rapidly. For example, beyond quantifying a particular protein in circulation, different states of protein activity and localization can be determined by quantifying numerous post-translational modifications, allowing a deeper understanding of biology and potentially more personalized prediction of risk.15Lawler P.R. Glycomic and cardiovascular disease: advancing down the path towards precision.Circ Res. 2018; 122: 1488-1490Google Scholar, 16Sharma P. Cosme J. Gramolini A.O. Recent advanced in cardiovascular proteomics.J Proteomics. 2013; 81: 3-14Crossref PubMed Scopus (28) Google Scholar This high dimensionality may require more advanced predictive statistical or other analytical approaches to handle its complexity. In this issue of the Canadian Journal of Cardiology, Singh and colleagues17Singh A. Dai D.L.Y. Ioannou K. et al.Ensembling electrical and proteogenomics biomarkers for improved prediction of cardiac-related 3-month hospitalizations: a pilot study.Can J Cardiol. 2019; 35: 471-479Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar attempt to improve on current approaches to predicting the need for HF hospitalization by integrating high-dimensional clinical and molecular data using advanced analytic approaches to build an accurate predictive algorithm. They performed transcriptomic and proteomic profiling in peripheral blood in 58 patients with HF and followed for the need for hospitalization by 3 months. They also examined continuous ECG data from 48-hour Holter monitors as well as routine clinical data and biomarkers. In comparison to patients with clinical stability, those patients requiring hospitalization had lower ambulatory blood pressures, higher brain natriuretic peptide, higher creatinine, and a longer duration of HF diagnosis. In this pilot study, a predictive model based on clinical variables had an area under the receiver operating characteristic curve of 0.76. When markers from the proteogenomic and sensory (Holter) data were introduced, the predictive accuracy of the model increased to 0.88. Although the small sample size of this pilot study requires expansion and eventual replication in much larger and external patient cohorts, these promising results offer optimism for next-generation predictive analytics to be developed in patients with HF. Of note, patients who required hospitalization had transcriptional upregulation of several inflammatory pathways including the interleukin-1 signalling pathway. Although inflammation plays important roles in HF pathogenesis, its effective therapeutic targeting in HF has been challenging.18Dick S.A. Epelman S. Chronic heart failure and inflammation: what do we really know?.Circ Res. 2016; 119: 159-176Crossref PubMed Scopus (360) Google Scholar It may be that between-patient variability in the degree of chronic inflammatory activation contributed to previous challenges in translation. This observation by Singh and colleagues highlights both the potential for individualized, “precision medicine” prognostication—and eventually treatment—in patients with HF as well as, more broadly, the ability of predictive approaches to provide insight into underlying disease biology. Such potentially causal pathways may lie upstream of increases in ventricular filling pressure and may hence represent important markers of progression and destabilization of underlying disease, identifying risk, and potential targets for therapy. Advanced bioanalytic approaches will be required to support the emergence of increasingly large and complex integrated clinical and molecular data. Employing modelling approaches that are more nimble than traditional regression models could enable the creation of more accurate predictive models. For example, in the study by Singh and colleagues, the authors employed advanced statistical tools to fine-tune their models and optimize variable selection. This included applying elastic net and least absolute shrinkage and selection operator (LASSO) penalties, which incorporate a zeroing/regularization step (involving parameter shrinkage and variable selection). Such methods drive low and correlated model β-coefficients to zero and hence create more parsimonious prediction models that better account for high degrees of correlation inherent in these types of data. Such approaches to statistical variable selection can complement systems biology-based approaches—such as network analysis, pathway enrichment, and other tools—to build accurate prediction models as well as to identify potential key disease mediators. The application of machine learning (ML) methods may offer further opportunities to build accurate, integrative, high-dimensional clinical-molecular risk prediction tools. ML approaches seem well suited to handle increasingly large and complex patient-level datasets generated from molecular and clinical data, as they offer flexible tuning parameters, often handle correlation better than penalized regression models, and expand the scope of predictors beyond what is input into the model by creating “hidden layers” of data interactions.19Deo R.C. Machine learning in medicine.Circulation. 2015; 132: 1920-1930Crossref PubMed Scopus (1251) Google Scholar Deep learning extends this by using stacked layers of increasingly higher order representations of objects. These layers could represent clinical measurements such as continuous vital signs, Holter monitor-based ECG data, and high-dimensional biomarker data for risk stratification and prediction. There is growing evidence that applying ML methods to high-dimensional, patient-level data may improve upon traditional predictive analytics. For example, applying an ML approach, called neural networks, to cardiopulmonary exercise test data from patients with HF allowed a complete breath-by-breath analysis, which was of greater discriminative value than the information provided by summary results alone.20Hearn J. Ross H.J. Mueller B. et al.Neural networks for prognostication of patients with heart failure.Circ Heart Failure. 2018; 11e005193Google Scholar ML may one day provide the potential for combining data from multiple sources, including remote “early warning” systems such as telemonitoring,21Ware P. Ross H.J. Cafazzo J.A. Laporte A. Gordon K. Seto E. Evaluating the implementation of a mobile phone-based telemonitoring program: longitudinal study guided by the consolidated framework for implementation research.JMIR. 2018; 6e10768Google Scholar along with various clinical risk-prediction tools such as the EHMRG30-ST risk score,22Lee D.S. Lee J.S. Schull M.J. Grimshaw J.M. Austin P.C. Tu J.V. Design and rationale for the Acute Congestive Heart Failure Urgent Care Evaluation: the ACUTE Study.Am Heart J. 2016; 181: 60-65Crossref PubMed Scopus (9) Google Scholar, 23Lee D.S. Lee J.S. Schull M.J. et al.Prospective validation of the emergency heart failure mortality risk grad for acute heart failure: the Acute Congestive Heart Failure Urgent Care Evaluation (ACUTE) study.Circulation. 2019; 139: 1146-1156Crossref PubMed Scopus (49) Google Scholar with high-dimensional biomarker data, to predict adverse events, including the need for hospitalization, in patients with HF.24Topol E.J. High-performance medicine: the convergence of human and artificial intelligence.Nat Med. 2019; 25: 44-56Crossref PubMed Scopus (1731) Google Scholar Predicting the need for HF hospitalization remains a high public health priority. Current approaches have important limitations, including a focus on consequences rather than causes of disease progression and destabilization, low-dimensional clinical and biomarker data, and reliance on limited statistical modelling approaches. Next-generation risk-prediction models that incorporate high-dimensional clinical and molecular data—perhaps with the support of ML approaches—may one day offer opportunities for personalized, cause-oriented, systems biology-based risk prediction in patients with HF." @default.
• W2913887066 created "2019-02-21" @default.
• W2913887066 creator A5021437066 @default.
• W2913887066 creator A5034441524 @default.
• W2913887066 creator A5045991393 @default.
• W2913887066 creator A5050949626 @default.
• W2913887066 creator A5054561346 @default.
• W2913887066 creator A5084851726 @default.
• W2913887066 date "2019-04-01" @default.
• W2913887066 modified "2023-09-25" @default.
• W2913887066 title "Next-Generation Approaches to Predicting the Need for Heart Failure Hospitalization" @default.
• W2913887066 cites W1418063743 @default.
• W2913887066 cites W2024071559 @default.
• W2913887066 cites W2098325126 @default.
• W2913887066 cites W2102814494 @default.
• W2913887066 cites W2106742300 @default.
• W2913887066 cites W2134071356 @default.
• W2913887066 cites W2153889487 @default.
• W2913887066 cites W2164008533 @default.
• W2913887066 cites W2177870565 @default.
• W2913887066 cites W2297309830 @default.
• W2913887066 cites W2400898762 @default.
• W2913887066 cites W2463195288 @default.
• W2913887066 cites W2483497709 @default.
• W2913887066 cites W2579907536 @default.
• W2913887066 cites W2612176626 @default.
• W2913887066 cites W2755609443 @default.
• W2913887066 cites W2800216577 @default.
• W2913887066 cites W2804829629 @default.
• W2913887066 cites W2891676555 @default.
• W2913887066 cites W2907198067 @default.
• W2913887066 cites W2908201961 @default.
• W2913887066 cites W2908727577 @default.
• W2913887066 cites W3012582253 @default.
• W2913887066 doi "https://doi.org/10.1016/j.cjca.2019.01.008" @default.
• W2913887066 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30935628" @default.
• W2913887066 hasPublicationYear "2019" @default.
• W2913887066 type Work @default.
• W2913887066 sameAs 2913887066 @default.
• W2913887066 citedByCount "2" @default.
• W2913887066 countsByYear W29138870662019 @default.
• W2913887066 countsByYear W29138870662020 @default.
• W2913887066 crossrefType "journal-article" @default.
• W2913887066 hasAuthorship W2913887066A5021437066 @default.
• W2913887066 hasAuthorship W2913887066A5034441524 @default.
• W2913887066 hasAuthorship W2913887066A5045991393 @default.
• W2913887066 hasAuthorship W2913887066A5050949626 @default.
• W2913887066 hasAuthorship W2913887066A5054561346 @default.
• W2913887066 hasAuthorship W2913887066A5084851726 @default.
• W2913887066 hasBestOaLocation W29138870661 @default.
• W2913887066 hasConcept C126322002 @default.
• W2913887066 hasConcept C164705383 @default.
• W2913887066 hasConcept C177713679 @default.
• W2913887066 hasConcept C194828623 @default.
• W2913887066 hasConcept C2778198053 @default.
• W2913887066 hasConcept C71924100 @default.
• W2913887066 hasConceptScore W2913887066C126322002 @default.
• W2913887066 hasConceptScore W2913887066C164705383 @default.
• W2913887066 hasConceptScore W2913887066C177713679 @default.
• W2913887066 hasConceptScore W2913887066C194828623 @default.
• W2913887066 hasConceptScore W2913887066C2778198053 @default.
• W2913887066 hasConceptScore W2913887066C71924100 @default.
• W2913887066 hasIssue "4" @default.
• W2913887066 hasLocation W29138870661 @default.
• W2913887066 hasLocation W29138870662 @default.
• W2913887066 hasOpenAccess W2913887066 @default.
• W2913887066 hasPrimaryLocation W29138870661 @default.
• W2913887066 hasRelatedWork W1491055865 @default.
• W2913887066 hasRelatedWork W2012241433 @default.
• W2913887066 hasRelatedWork W2039995053 @default.
• W2913887066 hasRelatedWork W2051712573 @default.
• W2913887066 hasRelatedWork W2076843464 @default.
• W2913887066 hasRelatedWork W2077083067 @default.
• W2913887066 hasRelatedWork W2314128040 @default.
• W2913887066 hasRelatedWork W2965474825 @default.
• W2913887066 hasRelatedWork W4247718175 @default.
• W2913887066 hasRelatedWork W4315784592 @default.
• W2913887066 hasVolume "35" @default.
• W2913887066 isParatext "false" @default.
• W2913887066 isRetracted "false" @default.
• W2913887066 magId "2913887066" @default.
• W2913887066 workType "article" @default.