Sage Journals: Discover world-class research

Abstract

Introduction:

Identifying the evolving needs of patients with advanced heart failure (AdHF) and triaging those at high risk of death can facilitate timely referrals to palliative care and advance patient-centered individualized care. There are limited models specific for patients with end-stage HF. We aim to identify risk factors associated with up to three-year all-cause mortality (ACM) and describe prognostic models developed or validated in AdHF populations.

Methods:

Frameworks proposed by Arksey, O’Malley, and Levac were adopted for this scoping review. We searched the Medline, EMBASE, PubMed, CINAHL, Cochrane library, Web of Science and gray literature databases for articles published between January 2010 and September 2020. Primary studies that included adults aged ≥ 18 years, diagnosed with AdHF defined as New York Heart Association class III/IV, American Heart Association/American College of Cardiology Stage D, end-stage HF, and assessed for risk factors associated with up to three-year ACM using multivariate analysis were included. Studies were appraised using the Quality of Prognostic Studies tool. Data were analyzed using a narrative synthesis approach.

Results:

We reviewed 167 risk factors that were associated with up to three-year ACM and prognostic models specific to AdHF patients across 65 articles with low-to-moderate bias. Studies were mostly based in Western and/or European cohorts (n = 60), in the acute care setting (n = 56), and derived from clinical trials (n = 40). Risk factors were grouped into six domains. Variables related to cardiovascular and overall health were frequently assessed. Ten prognostic models developed/validated on AdHF patients displayed acceptable model performance [area under the curve (AUC) range: 0.71–0.81]. Among the ten models, the model for end-stage-liver disease (MELD-XI) and acute decompensated HF with N-terminal pro b-type natriuretic peptide (ADHF/proBNP) model attained the highest discriminatory performance against short-term ACM (AUC: 0.81).

Conclusions:

To enable timely referrals to palliative care interventions, further research is required to develop or validate prognostic models that consider the evolving landscape of AdHF management.

Introduction

Heart failure (HF) is a progressive and incurable disease affecting more than 60 million individuals globally.¹ Despite the rapidly evolving advancements in medical therapy, HF remains characterized as a disease with high morbidity, economic burden, and all-cause mortality (ACM), with many patients progressing to advanced stages.^1,2 In the United States, the prevalence of HF is projected to increase by 33%, from 9.7 million individuals in 2025 to 12.9 million individuals by 2060.³

Patients with advanced HF (AdHF) experience severe and persistent symptoms resulting in functional constraints defined as New York Heart Association Class (NYHA) class III (i.e., marked limitation of physical activity) or class IV (i.e., have HF symptoms even at rest), severe cardiac dysfunction, unplanned hospitalizations from cardiac-related exacerbations, and severe impairment of exercise capacity despite optimal guideline-directed treatment.^4,5 They suffer a multitude of physical and psychological symptoms.⁶ Prior studies have highlighted that the end-of-life stage can begin three years before death and can vary by disease and individual.^7,8 Early integration of palliative care has been recommended by the American Heart Association (AHA)⁴ and the European Society of Cardiology⁵ as it provides an opportunity to address the unmet needs of patients and their caregivers by providing timely and optimal symptom management, including exercise-based rehabilitation, and psychological and spiritual support. In optimizing scarce palliative care resources for those with the greatest palliative needs, some studies have focused solely on AdHF when deriving its referral criteria.⁹

While there is consensus to integrate palliative care earlier, there is no systematic way to identify patients who are in the advanced stages of HF.¹⁰ Despite experiencing symptoms comparable with patients with advanced cancer,^11,12 patients with AdHF have suboptimal referral rates to palliative care, which occurs <2 weeks in the disease trajectory.^13,14 In the United States, among 2,647 hospitalized patients with AdHF, only 6% were referred for palliative care services.¹⁵ In Singapore, AdHF patients had the highest mean number of emergency department visits, inpatient hospitalizations, and substantial length of stay in the last one-year of life compared with patients with advanced cancer and other noncancer diagnoses.¹⁶

Establishing the advanced stage for HF patients has been a challenging pursuit for clinicians and health care systems due to prognostic uncertainty. Complicating matters further are the patients’ poor understanding of their prognosis and the terminal nature of this disease. Varying attitudes and beliefs toward palliative care by health care providers eventually lead to low referral rates to palliative care.^17,18 Consequently, the current management of HF patients has been biased toward life-prolonging therapies even at the end of life. Instead, the focus should be shifted toward improvement in quality of life through the reduction of cardinal symptoms such as dyspnea and fatigue, which can be managed with strategies, including palliative care and rehabilitation.^16,19–21

International guidelines recommend assessing a patient’s risk of mortality to guide discussions on prognosis.⁴Accurately prognosticating AdHF patients can redirect health care resources to high-risk patients who may benefit from early palliative interventions. HF patients are at higher risk of death for multiple reasons beyond cardiac deaths alone. Predicting the risk of ACM captures the overall risk of death and provides a more holistic view of the patient’s prognosis. Prior systematic reviews have evaluated risk prediction models for HF patients of varying disease severity in ambulatory and out-of-hospital care settings.^22–24 However, there is limited uptake of these models to augment clinical decision making and no one model can be endorsed. Several limitations have been highlighted that have hindered its adoption into clinical practice.^22–25 These include the heterogeneity of the study population, inclusion of patients from clinical trials that lack representation of the wider population, methodological issues, and lack of data to assess for generalizability. Studies evaluating HF prognostic models, including the Seattle Heart Failure Model (SHFM) and Meta-Analysis of Global Group in Chronic Heart Failure (MAGGIC), found an underestimation of one-year mortality risks among ambulatory HF patients²⁶ and those with AdHF.²⁷

It is challenging for prognostic models to keep pace with the constantly evolving care management, device therapies, and pharmacological treatments for HF, and hence important to reassess the predictive utility of variables included in these models in contemporary cohorts of patients. It is unlikely that one model will be able to predict all types of HF and severity. Focusing on a narrower group of AdHF patients may yield models with better performance. While several models are available for HF, there are far fewer validated models specific to the AdHF population.

Given the importance of risk prediction in the AdHF population, we aim to perform a scoping review to identify risk factors associated with up to three-year ACM, and describe prognostic models developed or validated in AdHF populations. Insights from this review can provide a more nuanced understanding of risk factors and models to predict ACM in AdHF. Stratifying AdHF patients based on prognosis can facilitate referrals to appropriate care settings, closer monitoring of high-risk AdHF patients by specialist palliative care teams, meaningful conversations about serious illness, and patient-centered future care plans.

Methodology

We adopted the five-stage scoping review framework proposed by Arksey and O’Malley²⁸ and Levac et al.²⁹ The formulation of the research question was guided by the Population, Index prognostic factors, Comparator prognostic factors, Timing, and Setting framework.³⁰ The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline was used to summarize the identification of studies.

Study eligibility

Primary studies published worldwide that included adults who were aged ≥18 years old, diagnosed with AdHF, assessed for risk factors that were associated with up to three-year ACM using multivariable analysis and written in English were considered for this review. AdHF was defined as HF patients with NYHA Class III or IV, or AHA/American College of Cardiology (ACC) Stage D, or patients identified to be in the advanced stages of the disease (Table 1).

Table 1.

Inclusion and Exclusion Criteria for Assessing Studies’ Eligibility

Study characteristic	Inclusion	Exclusion
Unit of analysis	• Patients	• County, hospitals, country
Type of publication	• Peer-reviewed articles, doctoral theses, gray literature, meta-analysis with individual-level data	• Editorials, letters, commentaries, abstracts, systematic reviews, protocols, book chapters
Study population	• AdHF was defined as NYHA III/IV, AHA/ACC Stage D or in advanced stages of heart failure • Any heart failure etiology • ≥18 Years old	• Pediatrics, non-AdHF, or received heart transplant • Heart failure disease staging was unclear or included all stages of heart failure (NYHA I-IV or AHA/ACC Stage A-D). • NYHA reported as mean or median values only
Study design	• Primary studies (retrospective or prospective)	• Qualitative studies
	• Observational (case–control, cohort, and cross-sectional) study designs • Randomized control trials • Studies that performed external validation and reported variables	• Cost–effectiveness studies
Intervention	• Had pacemakers, valve replacements, coronary artery bypass surgery, or none of the above	• Heart transplant • Left ventricular assisted device patients as they have a significantly different trajectory than AdHF patients
Analysis	• Multivariable analysis was conducted	• Only univariable analysis was conducted
		• Proportion or rates reported
		• Competing risk analysis was conducted
Outcome	• ACM within three years of AdHF diagnosis	• Composite outcomes
	• (1) Mean follow-up with three standard deviations, (2) median follow-up with two interquartile ranges, or (3) range of follow-up should be within three years.	• Nonmortality outcomes • Cardiac-specific mortality • Undefined mortality risk period, mortality risk period that was >3 years or standard deviation, interquartile or range of follow-up was not reported

Bibliographic databases and search strategy

We searched in PubMed, OVID, Excerpta Medical database, Cumulative Index to Nursing and Allied Health Literature, Web of Science, and the Cochrane library databases for articles published between January 2010 and September 2020 to ensure that contemporaneous studies were included. We included gray literature and hand-searched articles (Supplementary Appendix S1). All searches were conducted with the assistance of a medical librarian, with input from clinicians.

The search strategy applied medical subject headings’ terms and keywords that were tailored to each database. To identify our study population, we included keywords such as “end-stage heart failure,” “advanced heart failure,” “New York Heart Association III and/or IV,” “American College of Cardiology Stage C and/or D”; keywords for study design included “randomized controlled trial,” “observational,” “longitudinal,” “cross-sectional,” “case-control,” “cohort”; keywords for type of analysis included “prognostic model,” “prognosis,” “multivariable model,” “multivariable analysis,” and keywords to identify our outcome included “all-cause mortality,” and “death rate.” Additional details are presented in Supplementary Appendix S1.

Appraisal of included studies

All included articles were appraised for risk of bias using the Quality of Prognostic Studies (QUIPS) tool where six domains were assessed—namely, study participant, study attrition, prognostic factor measurement, adjustment for other prognostic factors, and statistical analysis and reporting.³¹

Screening and data extraction

Title and abstract screening, full-text screening, data extraction, and risk-of-bias assessment were performed using the COVIDENCE Systematic Review Software (Veritas Health Innovation, Melbourne, Australia). Title and abstract screening were performed by P.K., P.P.G., S.N.H.X., Y.W.F., E.C.S.S., R.Y.T., J.T., J.C., L.Z.H., L.T.H., V.H., T.C.K., A.C.H., T.W.S., and A.H., where each article was assessed by two reviewers. We tested the eligibility criteria to ensure interscreener agreement. Full text of articles that met the inclusion criteria were retrieved and screened to ensure relevance. Bibliographic information, study population, study design, statistical methods, risk factors examined, and mortality risk periods were extracted from each study. Full-text screening and data extraction were performed by P.K., P.P.G., S.H.N.X., R.T., and J.T. Disagreements were discussed and resolved through consensus within the team.

Analysis

We summarized key characteristics of the included studies and provided an overview of the risk factors investigated across the literature. These risk factors were described and categorized into broader domains, which may influence mortality. The emergent domains and subdomains were discussed with the research team before reaching consensus. For each risk factor, we counted the number of (1) unique articles it was assessed in, (2) models that reported the effect size of risk factors, (3) models that reported the risk factor to be significant, (4) models that found the risk factor to be associated with a reduced risk of up to three-year ACM, and (5) models that found the risk factor to be associated with an increased risk of up to three-year ACM.

Results

After removing duplicates and selecting the relevant articles that met our inclusion criteria, 65 articles were included in this scoping review, contributing to a total of 119 multivariable models (Fig. 1).

FIG. 1.

PRISMA flowchart of articles that were assessed.

Baseline characteristics of included studies

Most of the included studies were based in Western and/or European cohorts (n = 60 articles), in acute care settings (n = 56), with a median age of 65 years across study cohorts (Table 2).

Table 2.

Characteristics of the 65 Included Studies

Characteristics	n = 65
Publication year (n, %)
2010 to 2014	36 (55.4)
2015 to 2019	28 (43.1)
2020	1 (1.5)
Region of publication^a (n, %)
Asia	6 (9.2)
Europe	14 (21.5)
Middle East	3 (4.6)
Multinational	35 (53.8)
South America	2 (3.1)
United States	5 (7.7)
Heart failure severity (n, %)^b
NYHA III or IV	43 (66.2)
NYHA IV only	17 (26.2)
Population was defined as end-stage	5 (7.7)
LVEF cutoff used (n, %)^b
≤30%	19 (29.2)
≤35%	5 (7.7)
≤40%	23 (35.4)
≤45%	3 (4.6)
≤50%	2 (3.1)
≤55%	1 (1.5)
Did not use a cutoff/used multiple cutoffs	12 (18.5)
Average age of population (years)
Median (IQR)^c	65 (58.1 to 67.4)
Range (min–max)	54 to 86
Study design
Randomized controlled trial or controlled trial	3 (4.6)
Retrospective cohort study	55 (84.6)
Prospective cohort study	7 (10.8)
Sample size analyzed
Median (IQR)	433 (324 to 2,061)
Range (min–max)	54 to 4,131
Average study follow-up (months)
Median (IQR)^c	9.9 (6 to 12)
Range (min–max)	1 to 8 years
Setting
Acute hospital	56 (86.2)
Hospice/outpatient center/primary care/acute hospital and outpatient center	4 (6.1)
Unclear	5 (7.7)
Data source
Secondary analysis from clinical trial data	40 (61.5)
Electronic medical records/registry	12 (18.5)
Electronic medical records and primary data collection	10 (15.4)
Primary data collection	3 (4.6)

Asia (China, India, Japan); Europe (Austria, Germany, Greece, Italy, Poland, Spain, Sweden); Middle East (Iran, Turkey); Multinational (multiple trial sites in North and South America and Europe); South America (Brazil).

Based on individual study’s inclusion criteria.

Median is derived from the study’s reported mean or median estimate.

IQR, interquartile range; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; AdHF, advanced heart failure.

Forty studies were secondary analyses derived from HF clinical trials, primarily from the Efficacy of Vasopressin Antagonism in Heart Failure Outcomes with Tolvaptan (EVEREST) (n = 19 articles, 43 models) and the Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) (n = 14 studies, 22 models) trials. The Cox proportional hazard model was primarily used (n = 104 models), followed by logistic regression (n = 15 models).

Across the 119 models, 7 models (4 studies) assessed for in-hospital mortality (proportion died range: 4%−10.5%). A total of 81 models assessed for <1 year mortality (42 studies) (proportion died range: 3%−45%, median: 25%) and 31 models assessed for 1–3 years of mortality (31 studies) (proportion died range: 4%−72%, median: 31%).

Risk-of-bias assessment

Most articles were rated as having low-to-moderate bias based on QUIPS (Fig. 2). The domain “study attrition” was not appropriate for 12 articles as they were retrospective database studies and therefore assessed as ineligible. The domain “statistical methods and reporting” addresses appropriateness of the study’s statistical methods and completeness of reporting. Four studies were assessed as high bias as approximately 50% of the patients were excluded due to missing data.

FIG. 2.

Risk of bias using the Quality of Prognostic Studies tool.

Domains and subdomains identified

We identified 167 risk factors that were identified and categorized into the following six domains, that is, individual factors, acute conditions, cardiovascular status, overall health status, interventions, and external factors (Table 3). Due to the heterogeneity of risk factors within each domain, we further categorized these variables into 29 subdomains (Table 3). A detailed list of all 167 variables can be found in Supplementary Appendix S2.

Table 3.

Overview of Risk Factors Associated with Mortality Among Advanced Heart Failure Patients

Domains	Definition	Subdomains
1. Individual factors	Variables that measure individual characteristics and preferences	• Demographics • Advance directives • Acute conditions
2. Acute conditions	Variables that measure exacerbations and infections
3. Cardiovascular status	Variables that measure the patient’s heart failure condition, including etiology, severity, function, and other related conditions or tests	• Cardiac-related medical parameters • Cardiac biomarkers • Cardiac imaging • Heart failure etiology • Electrocardiograms • Cardiac-related symptoms	• Heart failure severity • Cardiac hemodynamics • Cardiac-related radiological features
4. Overall health status	Variables that measure the patients’ overall health status, including comorbid burden, function, nutrition, and other related condition or tests	• Renal function test • Full-blood count • Other biomarkers • Liver function test • Inflammatory markers • Lipid panel test • Glucose level	• Comorbidity • Functional status • Indexes/scores • Nutritional status • Lifestyle risk factors
5. Interventions	Variables that measure any medical or pharmaceutical interventions	• Cardiac interventions • Hemodynamic support • Medications
6. External factors	Variables that measure nonpatient factors	• Geographical region • Health care system

For most of the articles included, the tendency was to focus and report the effects of one or two key risk factors of interest while adjusting for other variables. We have reported variables that were commonly studied across the articles (Table 4). Overall, variables related to the domains of cardiovascular health and overall health status were frequently included in the models.

Table 4.

Subset of Risk Factors Commonly Assessed

Domain	Subdomain	Variable	No. of unique studies	No. of models that included the RF	No. of models that reported effect size of RF²	No. of models with RF as significant	RF associated with ↓ mortality risk	RF associated with ↑ mortality risk
Individual factors	Demographics	1. Increasing age	45 ^{32–64,65,66–74,75,76}	84	13	6	0	6
Individual factors	Demographics	2. Male gender	32^{32,34–36,39–48,50–54,56–59},61 –64,77,66,68,69,76	67	8	4	0	4
Individual factors	Demographics	3. Race	6 ^{32,50,65,77,69,78}	16	4	4	3	1
Individual factors	Advance care plans	4. Advance directive	1 ⁵⁰	1	1	1	0	1
Acute conditions	Acute conditions	5. Pneumonia	1 ⁴⁵	1	0	0	0	1
Acute conditions	Acute conditions	6. Acute decompensated heart failure	1 ⁴⁸	1	1	1	0	0
Acute conditions	Acute conditions	7. Late onset of acute kidney injury	1 ⁶⁰	1	1	1	0	1
Cardiovascular status	Cardiac-related medical parameters	8. Low systolic blood pressure	33 ^{33–35,39–42,46,47,51–55,57–63,65–68,77,73,78,75,79,80,81,76}	68	16	11	2	9
Cardiovascular status	Cardiac-related medical parameters	9. Higher heart rate	7 ^{42,43,48,52,62,66,78}	19	6	3	0	3
Cardiovascular status	Cardiac-related medical parameters	10. Low diastolic blood pressure	4 ^46,58,78,79	7	3	1	0	1
Cardiovascular status	Cardiac biomarkers	11. Higher BNP, NT-proBNP, or BNP/NT-proBNP	31 ^{33–36,38–43,47–49,51–54,56,57,59,60,62–64,65,66,69,71,72,82,83}	69	10	6	0	6
Cardiovascular status	Cardiac imaging	12. Higher ejection fraction	29 ^{33–42,45–48,51–55,57,58,61,63,64,65,67,79,82,84}	52	9	5	5	0
Cardiovascular status	HF etiology (ischemic)	13. Ischemic heart failure etiology	15 ^{39,41,42,53,58,62–64,67,73,74,75,85,81}	25	3	2	0	2
Cardiovascular status	Electrocardiogram	14. Atrial fibrillation or flutter	17 ^{34,35,39,41,44,45,51,52,54,56–58,62–64,66,82}	35	7	4	0	4
Overall health status	Renal function test	15. Low serum sodium	34 ^{33–35,39–42,44,45,47,51–55,57,61–64,65,66,70–75,78,86,83,81,87,76}	71	15	5	0	5
Overall health status	Renal function test	16. High serum BUN	27 ^{33–35,39–42,47,51–54,57,59,60,62–64,65,67,68,72,73,84,83,80,88}	50	11	9	0	9
Overall health status	Renal function test	17. Low estimated glomerular filtration rate	13 ^{39,44,45,56,61,66,69,73,74,78,79,81,76}	31	10	6	0	6
Overall health status	Renal function test	18. High serum creatinine	12 ^{36,37,39,48,55,59,60,67,68,70,71,86}	15	9	6	0	6
Overall health status	Full blood count	19. Hemoglobin	14 ^{36,37,44,45,53,55,59,61,66,67,77,72,73,80}	25	12	6	5	1
Overall health status	Comorbidity	20. Diabetes	25 ^{35,37,39,41,43–45,47,51–58,61–64,66,69,78,89}	51	3	1	0	1
Overall health status	Comorbidity	21. Hypertension	21 ^{35,37,39,41–44,47,51–54,57,58,61–64,69,73,75}	42	0	0	0	0
Overall health status	Comorbidity	22. Renal insufficiency	10 ^{35,39,41–43,47,51–54}	22	0	0	0	0
Overall health status	Comorbidity	23. Chronic obstructive pulmonary disease	9 ^{39,42,44,45,53,54,62–64}	19	2	0	0	0
Overall health status	Comorbidity	24. Chronic kidney disease	4 ^57,63,64,89	5	1	0	0	0
Overall health status	Comorbidity	25. Anemia	3 ^53,67,76	8	8	2	0	2
Overall health status	Comorbidity	26. Stroke/cerebrovascular accident	3 ^44,57,66	7	0	0	0	0
Overall health status	Comorbidity	27. Cancer	2 ^80,76	5	5	1	0	1
Overall health status	Comorbidity	28. Chronic respiratory failure	1 ⁶⁶	4	0	0	0	0
Overall health status	Comorbidity	29. Dementia and/or depression	1 ⁶⁶	4	0	0	0	0
Overall health status	Comorbidity	30. Peripheral artery disease	1 ⁴⁴	1	0	0	0	0
Overall health status	Functional status	31. NYHA	18 ^{34,35,39–43,56,58,61–64,65,77,67,71}	35	4	3	0	3
Interventions	Medications	32. ACEI/ARB use	27 ^{35,39–42,47,51–55,57,61–64,65,69,73–75,78,86,85,90,88,76}	55	8	5	5	0
Interventions	Medications	33. Beta-blocker use	25 ^{33,35,39–42,47,51–55,57,58,62–64,65,73,75,78,86,85,81,88}	46	4	2	2	0
External factors	Health care system	34. Payer source	1 ⁵⁰	1	1	0	0	0
External factors	Health care system	35. Referral source	1 ⁵⁰	1	1	1	1	0

ACEI: Angiotensin-Converting Enzyme Inhibitors (ACEI); ARB: Angiotensin II Receptor Blockers; BUN: Blood urea nitrogen; RF: Risk Factor.

Individual factors

Individual factors, such as demographics and advanced directives, were included in models across 48 articles (Table 4).^32–78 Older age (6 models) and male gender (4 models) were significantly associated with high risk of ACM. Different ethnic groups, specifically AdHF patients of Caucasian, African American, and Hispanic ethnicity, had a differential risk for ACM. Only one article assessed the impact of advance directives. Patients with advance directives were noted to confound the risk of ACM.

Acute conditions

Short-term exacerbations were included across three studies^45,48,60 (3 models) (Table 4). The presence of pneumonia (1 model) and having a late onset of acute kidney injury (1 model) were significantly associated with increased risk of ACM.

Cardiovascular status

A total of 57 articles^{32–49,51–86,89,91,92} included cardiovascular-related variables in the models (Table 4). Risk factors from cardiac-related medical parameters (36 articles), cardiac biomarkers (32 articles), cardiac imaging (31 articles), AdHF etiology (25 articles), and electrocardiograms (21 articles) were commonly assessed as prognostic factors.

Among the cardiac-related medical parameters, a lower systolic blood pressure (9 models) and more rapid heart rates (3 models) were associated with increased risk of death.

Of 10 models that reported the effect sizes of the cardiac biomarker, brain natriuretic peptide (BNP)/N-terminal pro b-type natriuretic peptide (NT-proBNP), six models found an association between increasing values of BNP/NT-proBNP levels with ACM risk.

Within the cardiac imaging subdomain, higher levels of ejection fraction were negatively associated with risk of ACM (5 models).

Ischemic and nonischemic AdHF etiology was observed to be commonly included (25 articles, 70 models). Although effect sizes were not reported in most of these models, ischemic AdHF etiology overall presented a higher risk of ACM (2 models).

In the electrocardiogram subdomain, the presence of atrial fibrillation or flutter conferred an increased risk of ACM (4 models).

Overall health status

A total of 57 articles^{33–37,39–45,47–81,83–92} included variables related to overall health status in the multivariable analysis. The most common subdomains studied were renal function (48 articles), full blood count (19 articles), comorbidity (30 articles), functional status (22 articles), and lifestyle risk factors (10 articles) (Table 4).

Renal function reflected by low estimated glomerular filtration rate (eGFR), low levels of serum sodium, and high levels of serum blood urea nitrogen was associated with increased risk of ACM. Five out of six models reported lower risk of death among those with higher hemoglobin levels. In contrast, one of the six models reported higher hemoglobin levels at admission being associated with higher risk of death.

While comorbidity burden was frequently included in the models, only 19 models reported effect sizes, of which comorbidity burden was not found to have a significant impact on mortality in 15 models.

With regard to function using the NYHA status, a higher NYHA status (IV vs. III) was associated with higher risk of death (18 articles; 3 models).

Interventions

A total of 37 studies^{32,33,35,39–42,44,45,47,51–55,57,58,61–65,69,73–78,81,82,85,86,88,93,94} assessed interventions such as medications, hemodynamic support, and cardiac-related interventions (Table 4). The use of angiotensin-converting enzyme inhibitors and/or angiotensin receptor blockers, and beta-blockers conferred a lower risk of ACM. Hemodynamic support and history of cardiac interventions such as coronary artery bypass grafts and implantable cardioverter defibrillators had no statistically significant influence on ACM (Supplementary Appendix S2).

External factors

A total of 17 studies accounted for nonpatient factors such as geographical region and health care payment sources (Table 4).^{32,34,35,39–42,47,50–54,57,63–65} Only one study found a differential effect of different payment sources (i.e., Medicare vs. Medicaid) on mortality.

Prognostic models developed or were validated in the AdHF population

One model predicted 6-month ACM based on the ESCAPE population, which had reasonable model performance (area under the curve; AUC: 0.78).⁸³ There were no data on external validation.

Szczurek et al. (2018) validated three models among the AdHF population in 2018.⁹⁵ The discriminatory performance of the model for end-stage liver disease (MELD-XI) was the highest (AUC: 0.812; 95% confidence interval [CI]: 0.769–0.856), followed by the HF survival score (AUC: 0.781; 95% CI: 0.732–0.829) and MAGGIC (AUC: 0.771; 95% CI: 0.720–0.823).

Scrutinio et al. (2015) validated six models among advanced acute decompensated HF patients for 90-day mortality (AUC range: 0.714–0.810) and in-hospital mortality (AUC range: 0.749–0.915).⁹⁶ The acute decompensated HF updated with the NT-proBNP (ADHF-NT-proBNP) model had the highest discriminatory performance in 90-day mortality (AUC: 0.810; 95% CI: 0.769–0.852) and in-hospital mortality (AUC: 0.815; 95% CI: 0.761–0.868). Prognostic models developed or validated on the AdHF population are summarized in Supplementary Appendix S3.

Discussion

Key findings of this scoping review

We reviewed 167 risk factors that were associated with up to three-year ACM and prognostic models specific to AdHF patients, across 65 articles that were rated with low-to-moderate bias. Risk factors were categorized into 6 domains and 29 subdomains, mostly reflecting cardiovascular health and overall health status. Several risk factors have emerged consistently across the articles. Lower levels of systolic blood pressure, ejection fraction, serum sodium, eGFR, hemoglobin levels, and higher levels of BNP/NT-proBNP and serum blood urea nitrogen were associated with increased risks of death. Generally, prognostic models identified in this review exhibited moderate model performance in the AdHF population, except for the MELD-XI and ADHF-NT-proBNP models that had the highest discriminatory performance against short-term ACM.

Variables identified in this review echo findings from previous systematic reviews,^22–25 which included HF patients in either inpatient or outpatient settings. However, most of these HF prognostic models that utilized these variables, including those reviewed for AdHF in this study, have demonstrated poor-to-moderate discrimination (AUC range: 0.66–0.81). This average model performance can be attributed to population, variables, and methodological issues in model development.

Prognostic models for patients diagnosed with HF

Prognostic models for chronic HF such as SHFM (1992–1994),⁹⁷ HFSS (1986–1991),⁹⁸ and MAGGIC (1995–2008),⁹⁹ as well as for acute decompensated HF such as the ADHF-NT-ProBNP (2005–2011),¹⁰⁰ were developed on cohorts from past decades. As medical therapy and care management have evolved, the significance of predictors in these models may evolve over time. These risk models were developed before the introduction of angiotensin receptor/neprilysin inhibitor and sodium/glucose cotransporter 2 inhibitors to the AHA guidelines, which significantly reduces mortality risks among HF patients.⁴ Prognostic models should be reassessed as new therapies are introduced to this population.

Current models may have included diverse HF populations, which lack variables to accurately discriminate survival probabilities or were derived from clinical trials that included less complex HF patients, limiting the generalizability to AdHF patients. It is unlikely that one prognostic model will be suited for all HF patients as underlying disease mechanisms and potential predictive variables differ by ejection fraction, etiology, or severity of HF. Risk stratification based on these factors may improve model accuracy.^101,102

The ESCAPE and HFSS model were developed in AdHF populations. For the ESCAPE model, having had cardiopulmonary resuscitation/mechanical ventilation and BNP > 1300 pg/mmol had the greatest influence on risk of ACM. The presence of interventricular conduction defect and presence of ischemic cardiomyopathy had the highest weights in the HFSS model. Creatinine, sodium, hemoglobin, systolic blood pressure, left ventricular ejection fraction, and blood urea nitrogen had the highest weights across these models, indicating greater influence on short-term ACM.

To provide a multidimensional understanding of a patient’s risk, functional parameters such as 6-minute walking distance, frailty, or other patient-reported outcomes such as dyspnea, fatigue, and the use of repeated measurements to capture dynamic changes of risk factors instead of static measurements can be considered to improve model performance.¹⁰³

There should be concerted efforts to report prognostic models in accordance with the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis framework. Methodological issues such as handling of missing data or appropriate reporting of model performance measures should be addressed to avoid biased estimates.^97,98 Discrimination and calibration measures provide a comprehensive understanding of how the model can reliably predict a patient’s probability of death. Discrimination, for example, AUC, is frequently reported in the isolation of other metrics. Calibration is the agreement between predicted risk probabilities produced by the model and observed mortality risk. It provides valuable insights on how well the model performs across the different levels of risk probabilities, indicating whether it tends to over- or underestimate a patient’s risk.¹⁰⁴ Nutter et al. evaluated six prognostic models and found that despite attaining moderate discrimination, mortality risks were underestimated in an elderly cohort of AdHF patients.²⁷ The generalizability of most models is unknown. Model performance derived from the development cohort tends to be optimistic, but external validation can provide a better understanding of how the model performs in a new population of HF patients.¹⁰⁵

Implementation to clinical practice

Referrals to palliative care should be based on needs. However, given that resources are limited, it may be challenging for health care providers to recognize all needs and disease-based criteria listed in Chang et al. (2020).^9,14 Therein lies the advantage of using risk scores. To improve the uptake of prognostic models, they can be embedded into electronic health records so that risk scores can be generated automatically to alert clinicians of the risk of deterioration. In addition, interventions can be implemented if patients are identified as high risk. This includes developing individualized patient-centered care models that integrate early palliative care and cardiac rehabilitation that offer exercise-based training and other psychosocial interventions, which significantly improves patients’ quality of life and reduces hospitalizations.²¹ Before deployment of the model, the probability threshold used to classify a patient as high risk must be considered in parallel with other factors such as acceptable rate of error, implications on treatment, care plans, local health care practice, health care resources, and funding mechanisms.¹⁰⁶ If models such as MEDL-XI and HFSS with higher sensitivity or a lower threshold are preferred, more patients will be identified as high-risk patients and would be referred to palliative care.

Accurate prognostication remains essential for directing goals of care and deciding the need for specialist or hospice care. The latter is often guided by prognosis due to resources and funding considerations. These risk models highlight to the health care community that certain patients are at a higher risk of deterioration despite the use of curative treatments. Palliative care should be initiated alongside these treatments by cardiologists, with early advanced care planning discussions for goal setting. Specialist palliative care can be initiated when symptoms become more complex, with a shift in resource utilization to community support. Therefore, these risk models are essential in initiating a process of collaborative care involving cardiologists, palliative care physicians, and the wider health care community.

Strength and limitations

This review enhances our current understanding of risk factors that can influence ACM, aiding the development or the updating of AdHF prognostic models. However, findings should be interpreted with caution. We were unable to account for the magnitude, significance, and direction of all risk factors across the 119 models as some studies only reported the effect size of the key risk factor. An incomplete model presentation hinders the ability of other researchers to reproduce the study findings or validate the model in external populations. The final model should be provided in the form of the original regression equation, including intercept or baseline hazard, to ensure adequate transparency in reporting and accurate interpretation of the results. Most of the studies included used data derived from clinical trials, which may limit the generalizability of these findings to the wider AdHF population. In addition, there were multiple studies stemming from the same HF trials. As each of these studies investigated different risk factors, the concern for publication may be reduced. Most of the studies included in this review were based in acute care settings. It is essential to recognize potential variability in prognostic factors in different care settings.¹⁰⁷ AdHF patients are not exclusively confined to the acute care setting. They are also encountered in ambulatory care settings and there exists a gap in the comprehension of risk factors and models for patients seen in such settings. Therefore, as an area for future research, understanding differences in prognostic factors for AdHF seen across different care settings may offer some pertinent insights in improving palliative care in AdHF. As the models account for ACM, the competing risk of dying from a noncardiac cause cannot be addressed.

Due to the substantial number of articles identified, and with the evolving landscape of care management in this population, we limited the search strategy to 2010–2020. We acknowledge that the search strategy could have been extended till 2023. We posit that the nature of risk factors may not have undergone substantial changes. We performed a rapid search of the literature in PubMed for relevant articles published between October 2020 and December 2023 and two relevant articles were identified.^10,108 The variables identified in these articles have already been accounted for in the scoping review (Table 4 and Supplementary Appendix S3).

Conclusion

To enable the early identification of high-risk patients who would benefit from timely palliative care interventions, further research is required to develop new models or validate the existing ones that consider the evolving landscape of AdHF management.

Footnotes

Acknowledgment

The authors thank Yasmeen Munro, medical librarian from the Lee Kong Chian School of Medicine, for her support in generating the research strategy.

Authors’ Contributions

A.H. and T.W.S. obtained funding for the study. The study was conceptualized and the research question was refined by P.K., T.W.S., and A.H. P.K., P.P.G., S.N.H.X., Y.W.F., J.C., L.Z.H., L.T.H., V.H., T.C.K., L.T., A.C.H., T.W.S., and A.H. provided inputs for the search strategy. P.K., P.P.G., S.N.H.X., Y.W.F., J.C., L.Z.H., L.T.H., V.H., T.C.K., A.C.H., T.W.S., and A.H. performed title and abstract screenings. Full-text screening and data extraction were performed by P.K., P.P.G., S.H.N.X., R.T., and J.T. Evidence was synthesized, and the article was written by P.K. All the authors provided inputs to the article.

Availability of Data and Materials

Data extracted from the included articles can be found in the supplementary information.

Author Disclosure Statement

The authors declare that they have no competing interests.

Funding Information

This study is funded by the National Medical Research Council Health Services Research Grant—NMRC HSG (HSRGEoL18may-0003). Grant: National Medical Research Council Health Services Research Grant (HSRGEoL18may-0003). Funding period: December 15, 2019, to January 14, 2024.

Supplementary Material

References

Lippi

, Sanchis-Gomar

. Global epidemiology and future trends of heart failure. AME Med J, 2020; 5:15–15; doi: 10.21037/amj.2020.03.03

Savarese

, Becher

, Lund

, et al. Global burden of heart failure: A comprehensive and updated review of epidemiology. Cardiovasc Res, 2023; 118(17):3272–3287; doi: 10.1093/cvr/cvac013

Mohebi

, Chen

, Ibrahim

, et al. Cardiovascular disease projections in the United States Based on the 2020 Census Estimates. J Am Coll Cardiol, 2022; 80(6):565–578; doi: 10.1016/j.jacc.2022.05.033

Heidenreich

, Bozkurt

, Aguilar

, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2022; 145(18):e895–e1032; doi: 10.1161/CIR.0000000000001063

McDonagh

, Metra

, Adamo

, et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J, 2021; 42(36):3599–3726; doi: 10.1093/eurheartj/ehab368

Whellan

, Goodlin

, Dickinson

, et al. End-of-life care in patients with heart failure. J Card Fail, 2014; 20(2):121–134; doi: 10.1016/j.cardfail.2013.12.003

Cohen-Mansfield

, Cohen

, Skornick-Bouchbinder

, et al. What is the end of life period? Trajectories and characterization based on primary caregiver reports. J Gerontol A Biol Sci Med Sci, 2018; 73(5):695–701; doi: 10.1093/gerona/glx195

Cohen-Mansfield

, Skornick-Bouchbinder

, Brill

. Trajectories of end of life: A systematic review. J Gerontol B Psychol Sci Soc Sci, 2018; 73(4):564–572; doi: 10.1093/geronb/gbx093

Chang

, Allen

, McClung

, et al. Criteria for referral of patients with advanced heart failure for specialized palliative care. J Am Coll Cardiol, 2022; 80(4):332–344; doi: 10.1016/j.jacc.2022.04.057

10.

Subramaniam

, Weston

, Killian

, et al. Development of advanced heart failure: A Population-Based Study. Circ Heart Fail, 2022; 15(5):e009218; doi: 10.1161/CIRCHEARTFAILURE.121.009218

11.

Bekelman

, Rumsfeld

, Havranek

, et al. Symptom burden, depression, and spiritual well-being: A comparison of heart failure and advanced cancer patients. J Gen Intern Med, 2009; 24(5):592–598; doi: 10.1007/s11606-009-0931-y

12.

O’Leary

, Murphy

, O’Loughlin

, et al. A comparative study of the palliative care needs of heart failure and cancer patients. Eur J Heart Fail, 2009; 11(4):406–412; doi: 10.1093/eurjhf/hfp007

13.

Beernaert

, Cohen

, Deliens

, et al. Referral to palliative care in COPD and other chronic diseases: A population-based study. Respir Med, 2013; 107(11):1731–1739; doi: 10.1016/j.rmed.2013.06.003

14.

Chang

, Kaplan

, Geng

, et al. Referral criteria to palliative care for patients with heart failure: A Systematic Review. Circ Heart Fail, 2020; 13(9):e006881; doi: 10.1161/CIRCHEARTFAILURE.120.006881

15.

Greener

, Quill

, Amir

, et al. Palliative care referral among patients hospitalized with advanced heart failure. J Palliat Med, 2014; 17(10):1115–1120; doi: 10.1089/jpm.2013.0658

16.

Kaur

, Wu

, Hum

, et al. Medical cost of advanced illnesses in the last-year of life-retrospective database study. Age Ageing, 2021; 51(1); doi: 10.1093/ageing/afab212

17.

Schallmo

, Dudley-Brown

, Davidson

. Healthcare providers’ perceived communication barriers to offering palliative care to patients with heart failure: An integrative review. J Cardiovasc Nurs, 2019; 34(2):E9–E18; doi: 10.1097/JCN.0000000000000556

18.

Klindtworth

, Oster

, Hager

, et al. Living with and dying from advanced heart failure: Understanding the needs of older patients at the end of life. BMC Geriatr, 2015; 15:125; doi: 10.1186/s12877-015-0124-y

19.

Hutchinson

, Gutheil

, Wessler

, et al. What is quality end-of-life care for patients with heart failure? A qualitative study with physicians. J Am Heart Assoc, 2020; 9(18):e016505; doi: 10.1161/JAHA.120.016505

20.

Erbs

, Hollriegel

, Linke

, et al. Exercise training in patients with advanced chronic heart failure (NYHA IIIb) promotes restoration of peripheral vasomotor function, induction of endogenous regeneration, and improvement of left ventricular function. Circ Heart Fail, 2010; 3(4):486–494; doi: 10.1161/CIRCHEARTFAILURE.109.868992

21.

Taylor

, Dalal

, Zwisler

. Cardiac rehabilitation for heart failure: ‘Cinderella’ or evidence-based pillar of care? Eur Heart J, 2023; 44(17):1511–1518; doi: 10.1093/eurheartj/ehad118

22.

Rahimi

, Bennett

, Conrad

, et al. Risk prediction in patients with heart failure: A systematic review and analysis. JACC Heart Fail, 2014; 2(5):440–446; doi: 10.1016/j.jchf.2014.04.008

23.

Alba

, Agoritsas

, Jankowski

, et al. Risk prediction models for mortality in ambulatory patients with heart failure: A systematic review. Circ Heart Fail, 2013; 6(5):881–889; doi: 10.1161/CIRCHEARTFAILURE.112.000043

24.

Di Tanna

, Wirtz

, Burrows

, et al. Evaluating risk prediction models for adults with heart failure: A systematic literature review. PLoS One, 2020; 15(1):e0224135; doi: 10.1371/journal.pone.0224135

25.

Ouwerkerk

, Voors

, Zwinderman

. Factors influencing the predictive power of models for predicting mortality and/or heart failure hospitalization in patients with heart failure. JACC Heart Fail, 2014; 2(5):429–436; doi: 10.1016/j.jchf.2014.04.006

26.

Allen

, Matlock

, Shetterly

, et al. Use of risk models to predict death in the next year among individual ambulatory patients with heart failure. JAMA Cardiol, 2017; 2(4):435–441; doi: 10.1001/jamacardio.2016.5036

27.

Nutter

, Tanawuttiwat

, Silver

. Evaluation of 6 prognostic models used to calculate mortality rates in elderly heart failure patients with a fatal heart failure admission. Congest Heart Fail, 2010; 16(5):196–201; doi: 10.1111/j.1751-7133.2010.00180.x

28.

Arksey

, O’Malley

. Scoping studies: Towards a methodological framework. International Journal of Social Research Methodology, 2005; 8(1):19–32; doi: 10.1080/1364557032000119616

29.

Levac

, Colquhoun

, O’Brien

. Scoping studies: Advancing the methodology. Implement Sci, 2010; 5:69; doi: 10.1186/1748-5908-5-69

30.

Riley

, Moons

KGM

, Snell

KIE

, et al. A guide to systematic review and meta-analysis of prognostic factor studies. Bmj, 2019; 364:k4597; doi: 10.1136/bmj.k4597

31.

Hayden

, van der Windt

, Cartwright

, et al. Assessing bias in studies of prognostic factors. Ann Intern Med, 2013; 158(4):280–286; doi: 10.7326/0003-4819-158-4-201302190-00009

32.

Ambrosy

, Khan

, Udelson

, et al. Changes in dyspnea status during hospitalization and postdischarge health-related quality of life in patients hospitalized for heart failure: Findings From The Everest Trial. Circ Heart Fail, 2016; 9(5); doi: 10.1161/CIRCHEARTFAILURE.115.002458

33.

Ambrosy

, Pang

, Khan

, et al. Clinical course and predictive value of congestion during hospitalization in patients admitted for worsening signs and symptoms of heart failure with reduced ejection fraction: Findings from the EVEREST trial. Eur Heart J, 2013; 34(11):835–843; doi: 10.1093/eurheartj/ehs444

34.

Ambrosy

, Vaduganathan

, Huffman

, et al. Clinical course and predictive value of liver function tests in patients hospitalized for worsening heart failure with reduced ejection fraction: An analysis of the EVEREST trial. Eur J Heart Fail, 2012; 14(3):302–311; doi: 10.1093/eurjhf/hfs007

35.

Ambrosy

, Vaduganathan

, Mentz

, et al. Clinical profile and prognostic value of low systolic blood pressure in patients hospitalized for heart failure with reduced ejection fraction: Insights from the Efficacy of Vasopressin Antagonism in Heart Failure: Outcome Study with Tolvaptan (EVEREST) trial. Am Heart J, 2013; 165(2):216–225; doi: 10.1016/j.ahj.2012.11.004

36.

Behnes

, Brueckmann

, Lang

, et al. Diagnostic and prognostic value of osteopontin in patients with acute congestive heart failure. Eur J Heart Fail, 2013; 15(12):1390–1400; doi: 10.1093/eurjhf/hft112

37.

Bozorgi

, Mehrabi Nasab

, Khoshnevis

, et al. Red cell distribution width and severe left ventricular dysfunction in ischemic heart failure. Crit Pathw Cardiol, 2016; 15(4):174–178; doi: 10.1097/HPC.0000000000000094

38.

Demir

, Akpinar

, Akkus

, et al. The prognostic value of arterial stiffness in systolic heart failure. Cardiol J, 2013; 20(6):665–671; doi: 10.5603/CJ.2013.0168

39.

Greene

, Gheorghiade

, Vaduganathan

, et al. Haemoconcentration, renal function, and post-discharge outcomes among patients hospitalized for heart failure with reduced ejection fraction: Insights from the EVEREST trial. Eur J Heart Fail, 2013; 15(12):1401–1411; doi: 10.1093/eurjhf/hft110

40.

Greene

, Harinstein

, Vaduganathan

, et al. EVEREST Trial Investigators. Prognostic value of monocyte count in patients hospitalized for heart failure with reduced ejection fraction (from the EVEREST Trial). Am J Cardiol, 2012; 110(11):1657–1662; doi: 10.1016/j.amjcard.2012.07.035

41.

Greene

, Vaduganathan

, Lupi

, et al. Prognostic significance of serum total cholesterol and triglyceride levels in patients hospitalized for heart failure with reduced ejection fraction (from the EVEREST Trial). Am J Cardiol, 2013; 111(4):574–581; doi: 10.1016/j.amjcard.2012.10.042

42.

Greene

, Vaduganathan

, Wilcox

, et al. The prognostic significance of heart rate in patients hospitalized for heart failure with reduced ejection fraction in sinus rhythm: Insights from the EVEREST (Efficacy of Vasopressin Antagonism in Heart Failure: Outcome Study With Tolvaptan) trial. JACC Heart Fail, 2013; 1(6):488–496; doi: 10.1016/j.jchf.2013.08.005

43.

Hohensinner

, Rychli

, Zorn

, et al. Macrophage-modulating cytokines predict adverse outcome in heart failure. Thromb Haemost, 2010; 103(2):435–441; doi: 10.1160/TH09-06-0399

44.

Jobs

, Schwind

, Katalinic

, et al. Prognostic significance of atrial fibrillation in acute decompensated heart failure with reduced versus preserved ejection fraction. Clin Res Cardiol, 2019; 108(1):74–82; doi: 10.1007/s00392-018-1321-4

45.

Jobs

, Simon

, de Waha

, et al. Pneumonia and inflammation in acute decompensated heart failure: A registry-based analysis of 1939 patients. Eur Heart J Acute Cardiovasc Care, 2018; 7(4):362–370; doi: 10.1177/2048872617700874

46.

, Bagarhatta

, Rathore

, et al. Is ambulatory blood pressure measurement a new indicator for survival among advanced heart failure cases. Indian Heart J, 2018; 70 Suppl 1(Suppl 1):S73–S78; doi: 10.1016/j.ihj.2017.08.028

47.

Lanfear

, Sabbah

, Goldsmith

, et al. Association of arginine vasopressin levels with outcomes and the effect of V2 blockade in patients hospitalized for heart failure with reduced ejection fraction: Insights from the EVEREST trial. Circ Heart Fail, 2013; 6(1):47–52; doi: 10.1161/CIRCHEARTFAILURE.112.970012

48.

, Lu

, Wu

, et al. Admission hypoxia-inducible factor 1alpha levels and in-hospital mortality in patients with acute decompensated heart failure. BMC Cardiovasc Disord, 2015; 15:79; doi: 10.1186/s12872-015-0073-6

49.

, Haugen

, Ikemoto

, et al. S100A8/A9 complex as a new biomarker in prediction of mortality in elderly patients with severe heart failure. Int J Cardiol, 2012; 155(1):26–32; doi: 10.1016/j.ijcard.2011.01.082

50.

Masterson Creber

, Russell

, Dooley

, et al. Use of the palliative performance scale to estimate survival among home hospice patients with heart failure. ESC Heart Fail, 2019; 6(2):371–378; doi: 10.1002/ehf2.12398

51.

Mentz

, Allen

, Kwasny

, et al. Influence of documented history of coronary artery disease on outcomes in patients admitted for worsening heart failure with reduced ejection fraction in the EVEREST trial. Eur J Heart Fail, 2013; 15(1):61–68; doi: 10.1093/eurjhf/hfs139

52.

Mentz

, Chung

, Gheorghiade

, et al. Atrial fibrillation or flutter on initial electrocardiogram is associated with worse outcomes in patients admitted for worsening heart failure with reduced ejection fraction: Findings from the EVEREST Trial. Am Heart J, 2012; 164(6):884–892e2; doi: 10.1016/j.ahj.2012.09.011

53.

Mentz

, Greene

, Ambrosy

, et al. Clinical profile and prognostic value of anemia at the time of admission and discharge among patients hospitalized for heart failure with reduced ejection fraction: Findings from the EVEREST trial. Circ Heart Fail, 2014; 7(3):401–408; doi: 10.1161/CIRCHEARTFAILURE.113.000840

54.

Mentz

, Schmidt

, Kwasny

, et al. The impact of chronic obstructive pulmonary disease in patients hospitalized for worsening heart failure with reduced ejection fraction: An analysis of the EVEREST Trial. J Card Fail, 2012; 18(7):515–523; doi: 10.1016/j.cardfail.2012.04.010

55.

Polat

, Aydin

, Yildiz

, et al. The prognostic significance of serum albumin in patients with acute decompensated systolic heart failure. Acta Cardiol, 2014; 69(6):648–654; doi: 10.1080/ac.69.6.1000007

56.

Rychli

, Richter

, Hohensinner

, et al. Hepatocyte growth factor is a strong predictor of mortality in patients with advanced heart failure. Heart, 2011; 97(14):1158–1163; doi: 10.1136/hrt.2010.220228

57.

Sarma

, Mentz

, Kwasny

, et al. Association between diabetes mellitus and post-discharge outcomes in patients hospitalized with heart failure: Findings from the EVEREST trial. Eur J Heart Fail, 2013; 15(2):194–202; doi: 10.1093/eurjhf/hfs153

58.

Shamoun

, De Marco

, DeMets

, et al. Impact of degree of left ventricular remodeling on clinical outcomes from cardiac resynchronization therapy. JACC Heart Fail, 2019; 7(4):281–290; doi: 10.1016/j.jchf.2018.11.004

59.

Takaya

, Yoshihara

, Yokoyama

, et al. Risk stratification of acute kidney injury using the blood urea nitrogen/creatinine ratio in patients with acute decompensated heart failure. Circ J, 2015; 79(7):1520–1525; doi: 10.1253/circj.CJ-14-1360

60.

Takaya

, Yoshihara

, Yokoyama

, et al. Impact of onset time of acute kidney injury on outcomes in patients with acute decompensated heart failure. Heart Vessels, 2016; 31(1):60–65; doi: 10.1007/s00380-014-0572-x

61.

Testani

, Hanberg

, Arroyo

, et al. Hypochloraemia is strongly and independently associated with mortality in patients with chronic heart failure. Eur J Heart Fail, 2016; 18(6):660–668; doi: 10.1002/ejhf.477

62.

Vaduganathan

, Ambrosy

, Greene

, et al. Predictive value of low relative lymphocyte count in patients hospitalized for heart failure with reduced ejection fraction: Insights from the EVEREST trial. Circ Heart Fail, 2012; 5(6):750–758; doi: 10.1161/CIRCHEARTFAILURE.112.970525

63.

Vaduganathan

, Greene

, Ambrosy

, et al. Relation of serum magnesium levels and postdischarge outcomes in patients hospitalized for heart failure (from the EVEREST Trial). Am J Cardiol, 2013; 112(11):1763–1769; doi: 10.1016/j.amjcard.2013.07.020

64.

Vaduganathan

, Marti

, Mentz

, et al. Serum osmolality and postdischarge outcomes after hospitalization for heart failure. Am J Cardiol, 2016; 117(7):1144–1150; doi: 10.1016/j.amjcard.2015.12.059

65.

Wang

, Piccini

, Konstam

, et al. Implantable cardioverter-defibrillators in patients hospitalized for heart failure with chronically reduced left ventricular ejection fraction. Am J Ther, 2010; 17(4):e78-87–e87; doi: 10.1097/MJT.0b013e3181e70a65

66.

Zairis

, Patsourakos

, Georgilas

, et al. Anemia and early mortality in patients with decompensation of chronic heart failure. Cardiology, 2011; 119(3):125–130; doi: 10.1159/000330494

67.

Damluji

, Macon

, Fox

, et al. The association between in-hospital hemoglobin changes, cardiovascular events, and mortality in acute decompensated heart failure: Results from the ESCAPE trial. Int J Cardiol, 2016; 222:531–537; doi: 10.1016/j.ijcard.2016.07.264

68.

, Paniagua

, Denktas

, et al. Usefulness of the sum of pulmonary capillary wedge pressure and right atrial pressure as a congestion index that prognosticates heart failure survival (from the evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness trial). Am J Cardiol, 2016; 118(6):854–859; doi: 10.1016/j.amjcard.2016.06.040

69.

McCallum

. Relation Between Biomarkers of Decongestion and Kidney Function with Outcomes in Acute Decompensated Heart Failure. Tufts University: 2019.

70.

Omar

, Charnigo

, Guglin

. Prognostic significance of discharge hyponatremia in heart failure patients with normal admission sodium (from the ESCAPE Trial). Am J Cardiol, 2017; 120(4):607–615; doi: 10.1016/j.amjcard.2017.05.030

71.

Omar

, Guglin

. Rise in BNP despite appropriate acute decompensated heart failure treatment: Patient characteristics and outcomes. Herz, 2017; 42(4):411–417; doi: 10.1007/s00059-016-4478-5

72.

Omar

, Guglin

. Extremely elevated BNP in acute heart failure: Patient characteristics and outcomes. Int J Cardiol, 2016; 218:120–125; doi: 10.1016/j.ijcard.2016.05.038

73.

Testani

, Brisco

, Turner

, et al. Loop diuretic efficiency: A metric of diuretic responsiveness with prognostic importance in acute decompensated heart failure. Circ Heart Fail, 2014; 7(2):261–270; doi: 10.1161/CIRCHEARTFAILURE.113.000895

74.

Testani

, Chen

, McCauley

, et al. Potential effects of aggressive decongestion during the treatment of decompensated heart failure on renal function and survival. Circulation, 2010; 122(3):265–272; doi: 10.1161/CIRCULATIONAHA.109.933275

75.

Testani

, Coca

, Shannon

, et al. Influence of renal dysfunction phenotype on mortality in the setting of cardiac dysfunction: Analysis of three randomized controlled trials. Eur J Heart Fail, 2011; 13(11):1224–1230; doi: 10.1093/eurjhf/hfr123

76.

Verdu-Rotellar

, Calero

, Abellana

, et al. Short-term mortality in end-stage heart failure patients. Aten Primaria, 2020; 52(7):477–487; doi: 10.1016/j.aprim.2019.07.019

77.

, Ghali

, Neuberg

, et al. Uric acid level and allopurinol use as risk markers of mortality and morbidity in systolic heart failure. Am Heart J, 2010; 160(5):928–933; doi: 10.1016/j.ahj.2010.08.006

78.

Testani

, Coca

, McCauley

, et al. Impact of changes in blood pressure during the treatment of acute decompensated heart failure on renal and clinical outcomes. Eur J Heart Fail, 2011; 13(8):877–884; doi: 10.1093/eurjhf/hfr070

79.

Kawase

, Kadota

, Nakamura

, et al. Low systolic blood pressure on admission predicts mortality in patients with acute decompensated heart failure due to moderate to severe aortic stenosis. Circ J, 2014; 78(10):2455–2459; doi: 10.1253/circj.cj-14-0712

80.

Omar

, Guglin

. Hypothermia is an independent predictor of short and intermediate term mortality in acute systolic heart failure: Insights from the ESCAPE trial. Int J Cardiol, 2016; 220:729–733; doi: 10.1016/j.ijcard.2016.06.166

81.

Testani

, McCauley

, Kimmel

, et al. Characteristics of patients with improvement or worsening in renal function during treatment of acute decompensated heart failure. Am J Cardiol, 2010; 106(12):1763–1769; doi: 10.1016/j.amjcard.2010.07.050

82.

Osteresch

, Diehl

, Kuhl

, et al. Impact of right heart function on outcome in patients with functional mitral regurgitation and chronic heart failure undergoing percutaneous edge-to-edge-repair. J Interv Cardiol, 2018; 31(6):916–924; doi: 10.1111/joic.12566

83.

O’Connor

, Hasselblad

, Mehta

, et al. Triage after hospitalization with advanced heart failure: The ESCAPE (Evaluation Study of congestive heart failure and pulmonary artery catheterization effectiveness) risk model and discharge score. J Am Coll Cardiol, 2010; 55(9):872–878; doi: 10.1016/j.jacc.2009.08.083

84.

Subramanian

, Subramanian

, Deswal

, et al. New predictive models of heart failure mortality using time-series measurements and ensemble models. Circ Heart Fail, 2011; 4(4):456–462; doi: 10.1161/CIRCHEARTFAILURE.110.958496

85.

Ochiai

, Cardoso

, Vieira

, et al. Predictors of low cardiac output in decompensated severe heart failure. Clinics (Sao Paulo), 2011; 66(2):239–244; doi: 10.1590/s1807-59322011000200010

86.

Carlo

, Cardoso

, Ochia

, et al. Temporal variation in the prognosis and treatment of advanced heart failure - before and after 2000. Arq Bras Cardiol, 2014; 102(5):495–504; doi: 10.5935/abc.20140050

87.

Szygula-Jurkiewicz

, Szczurek

, Skrzypek

, et al. One-year survival of ambulatory patients with end-stage heart failure: The analysis of prognostic factors. Pol Arch Intern Med, 2017; 127(4):254–260; doi: 10.20452/pamw.3975

88.

Bilchick

, Chishinga

, Parker

, et al. Plasma volume and renal function predict six-month survival after hospitalization for acute decompensated heart failure. Cardiorenal Med, 2017; 8(1):61–70; doi: 10.1159/000481149

89.

Tzou

, Tung

, Frankel

, et al. Ventricular tachycardia ablation in severe heart failure: An International Ventricular Tachycardia Ablation Center Collaboration Analysis. Circ Arrhythm Electrophysiol, 2017; 10(1); doi: 10.1161/CIRCEP.116.004494

90.

Adejumo

, Koelling

, Hummel

. Nutritional risk index predicts mortality in hospitalized advanced heart failure patients. J Heart Lung Transplant, 2015; 34(11):1385–1389; doi: 10.1016/j.healun.2015.05.027

91.

Scrutinio

, Ammirati

, Guida

, et al. The ADHF/NT-proBNP risk score to predict 1-year mortality in hospitalized patients with advanced decompensated heart failure. J Heart Lung Transplant, 2014; 33(4):404–411; doi: 10.1016/j.healun.2013.12.005

92.

Yamaguchi

, Abe

, Arasaki

, et al. The prognostic impact of a concentric left ventricular structure evaluated by transthoracic echocardiography in patients with acute decompensated heart failure: A retrospective study. Int J Cardiol, 2019; 287:73–80; doi: 10.1016/j.ijcard.2018.07.097

93.

Lynga

, Persson

, Hagg-Martinell

, et al. Weight monitoring in patients with severe heart failure (WISH). A Randomized Controlled Trial. Eur J Heart Fail, 2012; 14(4):438–444; doi: 10.1093/eurjhf/hfs023

94.

Packer

, Pitt

, Rouleau

, et al. Long-term effects of flosequinan on the morbidity and mortality of patients with severe chronic heart failure: Primary results of the profile trial after 24 years. JACC Heart Fail, 2017; 5(6):399–407; doi: 10.1016/j.jchf.2017.03.003

95.

Szczurek

, Szyguła-Jurkiewicz

, Zakliczyński

, et al. Prognostic value of selected risk scales in patients with end-stage heart failure. Kardiol Pol, 2018; 76(9):1320–1326; doi: 10.5603/KP.a2018.0090

96.

Scrutinio

, Ammirati

, Passantino

, et al. Predicting short-term mortality in advanced decompensated heart failure—role of the updated acute decompensated heart failure/N-terminal pro-B-type natriuretic Peptide risk score. Circ J, 2015; 79(5):1076–1083; doi: 10.1253/circj.CJ-14-1219

97.

Levy

, Mozaffarian

, Linker

, et al. The seattle heart failure model: Prediction of survival in heart failure. Circulation, 2006; 113(11):1424–1433; doi: 10.1161/CIRCULATIONAHA.105.584102

98.

Aaronson

, Schwartz

, Chen

, et al. Development and prospective validation of a clinical index to predict survival in ambulatory patients referred for cardiac transplant evaluation. Circulation, 1997; 95(12):2660–2667; doi: 10.1161/01.cir.95.12.2660

99.

Pocock

, Ariti

, McMurray

, et al. Predicting survival in heart failure: A risk score based on 39 372 patients from 30 studies. Eur Heart J, 2013; 34(19):1404–1413; doi: 10.1093/eurheartj/ehs337

100.

Scrutinio

, Ammirati

, Guida

, et al. Clinical utility of N-terminal pro-B-type natriuretic peptide for risk stratification of patients with acute decompensated heart failure. Derivation and validation of the ADHF/NT-proBNP risk score. Int J Cardiol, 2013; 168(3):2120–2126; doi: 10.1016/j.ijcard.2013.01.005

101.

Vucicevic

, Honoris

, Raia

, et al. Current indications for transplantation: Stratification of severe heart failure and shared decision-making. Ann Cardiothorac Surg, 2018; 7(1):56–66; doi: 10.21037/acs.2017.12.01

102.

Siddiqi

, Ahmed

, Greene

, et al. Performance of current risk stratification models for predicting mortality in patients with heart failure: A systematic review and meta-analysis. Eur J Prev Cardiol, 2022; 29(15):2027–2048; doi: 10.1093/eurjpc/zwac148

103.

Jha

, Ha

, Hickman

, et al. Frailty in advanced heart failure: A systematic review. Heart Fail Rev, 2015; 20(5):553–560; doi: 10.1007/s10741-015-9493-8

104.

Steyerberg

, Vergouwe

. Towards better clinical prediction models: Seven steps for development and an ABCD for validation. Eur Heart J, 2014; 35(29):1925–1931; doi: 10.1093/eurheartj/ehu207

105.

Moons

, Altman

, Reitsma

, et al. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): Explanation and elaboration. Ann Intern Med, 2015; 162(1):W1–73; doi: 10.7326/M14-0698

106.

Kaur

, Kannapiran

, Ng

SHX

, et al. Predicting mortality in patients diagnosed with advanced dementia presenting at an acute care hospital: The PROgnostic Model for Advanced DEmentia (PRO-MADE). BMC Geriatr, 2023; 23(1):255; doi: 10.1186/s12877-023-03945-8

107.

Kaur

, Tan

, Gunapal

PPG

, et al. Deaths in dementia: A scoping review of prognostic variables. BMJ Support Palliat Care, 2020; 11(3):242–252; doi: 10.1136/bmjspcare-2020-002217

108.

Codina

, Dobarro

, de Juan-Baguda

, et al. Heart failure risk scores in advanced heart failure patients: Insights from the LEVO-D registry. ESC Heart Fail, 2023; 10(5):2875–2881; doi: 10.1002/ehf2.14400

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB

0.92 MB

0.03 MB

Risk Factors for All-Cause Mortality in Patients Diagnosed with Advanced Heart Failure: A Scoping Review

Abstract

Introduction:

Methods:

Results:

Conclusions:

Introduction

Methodology

Study eligibility

Bibliographic databases and search strategy

Appraisal of included studies

Screening and data extraction

Analysis

Results

Baseline characteristics of included studies

Risk-of-bias assessment

Domains and subdomains identified

Individual factors

Acute conditions

Cardiovascular status

Overall health status

Interventions

External factors

Prognostic models developed or were validated in the AdHF population

Discussion

Key findings of this scoping review

Prognostic models for patients diagnosed with HF

Implementation to clinical practice

Strength and limitations

Conclusion

Footnotes

Acknowledgment

Authors’ Contributions

Availability of Data and Materials

Author Disclosure Statement

Funding Information

Supplementary Material

References

Supplementary Material