Revista Portuguesa de Cardiologia 42 (2023) 89---95

Revista Portuguesa de
Cardiologia
Portuguese Journal of Cardiology
www.revportcardiol.org

ORIGINAL ARTICLE

Clinical outcomes in heart failure with reduced left

ventricular ejection fraction and good functional
capacity: The illusion of stability
Sérgio Maltês ∗ , Catarína Brízido, Bruno M.L. Rocha, Gonçalo J.L. Cunha,
Christopher Strong, Pedro Freitas, Anaí Durazzo, António Tralhão,
António Ventosa, Carlos Aguiar, Miguel Mendes

Cardiology Department, Hospital de Santa Cruz, Centro Hospitalar Lisboa Ocidental, Lisbon, Portugal

Received 27 May 2021; accepted 25 October 2021

Available online 10 October 2022

KEYWORDS Abstract
Cardiopulmonary Background: Heart failure (HF) remains a prevalent syndrome with significant morbidity and
exercise testing; mortality. Optimal drug and device therapies are crucial to reduce the risk of death or HF
Heart failure; admission. Yet, less symptomatic patients with good functional capacity are often perceived as
Guideline-directed having a low risk of adverse events and their attending physicians may suffer from prescription
medical therapy; inertia or refrain from performing therapy optimization. Maximum or peak oxygen consumption
Peak oxygen uptake; (pVO2 ) assessed during cardiopulmonary exercise testing (CPET) is often used as a prognosis
Risk stratification indicator and surrogate marker for functional capacity. Our goal was to assess clinical outcomes
in a seemingly low risk HF population in Weber class A (pVO2 >20 mL/kg/min) with reduced left
ventricular ejection fraction (LVEF).
Methods: Single-center retrospective observational study enrolling consecutive HF patients
with LVEF<40% (HFrEF) performing CPET between 2003 and 2018. Those with pVO2 >20
mL/kg/min were included. The primary endpoint was a composite of all-cause death or HF
hospitalizations at two years after CPET. We also assessed the rates of N-terminal pro b-type
natriuretic peptide (NT-proBNP) elevations at baseline.
Results: Seventy-two patients were included (mean age of 53 ± 10 years; 86% male; 90% NYHA
I-II; median LVEF 32%; median pVO2 24 mL/kg/min). At baseline, 93% had an NT-proBNP level
>125 pg/mL (median NT-proBNP 388 [201---684] pg/mL). Overall, seven patients (10%) met the
primary endpoint: three died (4%) and five (7%) had at least one HF admission. Among those
who died, only one patient had an HF admission during follow up.

∗ Corresponding author.
E-mail address: [email protected] (S. Maltês).

https://doi.org/10.1016/j.repc.2021.10.012
0870-2551/© 2022 Sociedade Portuguesa de Cardiologia. Published by Elsevier España, S.L.U. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 S. Maltês, C. Brízido, B.M.L. Rocha et al.

Conclusion: In a clinically stable HFrEF population with good functional capacity, persistent
neurohormonal activation was present in the majority, and one in ten patients died or had a HF
admission at two years’ follow-up. These findings support the urgent need to motivate clinicians
to pursue optimal drug uptitration even in less symptomatic patients.
© 2022 Sociedade Portuguesa de Cardiologia. Published by Elsevier España, S.L.U. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

PALAVRAS-CHAVE Eventos clínicos na insuficiência cardíaca com fração de ejeção reduzida e boa
Prova de esforço capacidade funcional: a ilusão de estabilidade
cardiorrespiratória;
Resumo
Insuficiência
Introdução: A insuficiência cardíaca (IC) com fração de ejeção reduzida é uma síndrome preva-
Cardíaca;
lente com morbimortalidade significativa. O aprimoramento da terapêutica é crucial para
Terapêutica médica
reduzir o risco de morte ou admissão por IC. No entanto, doentes paucissintomáticos com boa
otimizada;
capacidade funcional são frequentemente considerados como tendo um baixo risco de eventos
Consumo máximo de
adversos, não são, por isso, alvo de aprimoramento terapêutico. O consumo máximo de oxigénio
oxigénio;
(pVO2 ), avaliado durante provas de esforço cardiorrespiratórias (PECR), é utilizado como um
Estratificação de risco
marcador prognóstico e da capacidade funcional do doente. O nosso objetivo foi avaliar o
número e tipo de eventos clínicos numa população de IC com aparente baixo risco em classe
Weber A (pVO2 > 20 mL/kg/min) e fração de ejeção ventricular esquerda (FEVE) reduzida.
Métodos: Estudo observacional, retrospetivo, unicêntrico, incluindo doentes consecutivos com
IC com FEVE <40% (ICFEr) que realizaram PECR entre 2003-2018. Doentes com pVO2 >20
mL/kg/min foram incluídos. O endpoint primário foi um composto de morte por todas as causas
ou hospitalização por IC nos dois anos após a PECR. Também avaliámos a percentagem de
doentes com elevação basal de NT-proBNP.
Resultados: Foram incluídos 72 doentes (idade 53 ± 10 anos; 86% homens; 90% classe NYHA I-II;
FEVE média 32%; pVO2 mediana 24 mL/kg/min), 93% apresentavamm um nível de NT-proBNP
>125 pg/mL (NT-proBNP mediano 388 [201-684] pg/mL). Globalmente, sete doentes (10%) atingi-
ram o endpoint primário: três por óbito (4%) e cinco (7%) tiveram pelo menos uma hospitalização
por IC. Entre os doentes que faleceram, apenas um teve uma hospitalização por IC, pelo que
contabilizado pelo primeiro evento (hospitalização).
Conclusão: Numa população clinicamente estável de doentes com ICFEr e boa capacidade fun-
cional, a maioria apresentou ativação neuro-hormonal persistente e um em cada dez atingiu
o endpoint primário (óbito ou internamento por IC) durante um follow-up de dois anos após a
PECR. Esses achados fortalecem a necessidade urgente de promover a prescrição otimizada de
terapêuticas modificadoras de doença.
© 2022 Sociedade Portuguesa de Cardiologia. Publicado por Elsevier España, S.L.U. Este é um
artigo Open Access sob uma licença CC BY-NC-ND (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

Introduction intermediate term mortality, the latter are perceived to

have a low risk of adverse outcomes.3 Consequently, physi-
Heart failure (HF) is a chronic syndrome with an cians will often refrain from optimizing drugs and doses,
ever-increasing prevalence and high risk of mortality, or implanting devices in HF with reduced ejection fraction
hospitalizations and disability, particularly for patients (HFrEF) patients that are deemed clinically stable and in
not receiving optimal guideline-directed medical therapy NYHA class II. This decision is taken to avoid the risks of
(GDMT).1,2 Accurate risk stratification is a key step in HF side effects and the increased burden of multiple healthcare
assessment, and misperception of risk can lead to subopti- encounters.4
mal care.3 Physicians managing HF patients are more easily Yet, HFrEF should be viewed as a progressive dis-
prompted to modify the therapy in those with New York ease in patients with a significant mass of at-risk viable
Heart Association (NYHA) class III or IV symptoms than in myocardium, in whom myocardial dysfunction is amenable
those who are asymptomatic or have a no more than mild to improvement with optimal GDMT.5 Neurohormonal acti-
limitation of functional capacity.3 Whereas the former are vation, which is initially compensatory, never switches off
expected to have a progressively worsening condition, cha- in HF.5 These systems are constantly on in an attempt to
racterized by recurrent hospitalizations and high short to compensate for the failing heart’s inability to maintain

90
 Revista Portuguesa de Cardiologia 42 (2023) 89---95

normal cardiovascular homeostasis. The chronic presence Statistical analysis

of these circulating neurohormones exacerbates hemody-
namic abnormalities, which encourages further remodeling, Continuous variables were expressed as mean ± standard
neurohormone release and additional hemodynamic deteri- deviation (SD) (normal distribution) or median and
oration. interquartile range (IQR) (non-parametric). Time-to-event
To illustrate the dire prognosis displayed in HFrEF analysis was performed by applying Kaplan-Meier curves.
patients, even if only mildly symptomatic, we assessed Categorical variables were expressed as counts (percent-
the rates of N-terminal pro b-type natriuretic peptide age). Statistical analysis was performed using Statistical
(NTproBNP) elevation and major adverse outcomes in an Package for the Social Sciences Statistics v.15.0 (IBM, New
HFrEF population with good functional capacity, defined York, USA).
by a maximum oxygen consumption (pVO2 ) higher than 20
mL/kg/min. Results

Population, materials and methods Study population

Study population and data collection Overall, 407 HF patients performed CPET between 2003 and
2018, of which 72 were included (70 patients excluded due
This is a single-center retrospective cohort study enrolling to LVEF >40%; 265 patients excluded due to pVO2 equal
consecutive HFrEF patients undergoing cardiopulmonary to or lower than 20 mL/kg/min). Baseline demographics
exercise testing (CPET) in our center between January 2003 are depicted in Table 1. Mean age was 52 ± 10 years, 83%
and December 2018. Those who had a peak O2 uptake higher patients were male, most patients (90%) had NYHA func-
than 20 mL/kg/min were included. HFrEF was defined as tional class I---II symptoms, and 46% had ischemic HF. Median
per European Society of Cardiology 2016 guidelines.1 All LVEF was 32% (IQR: 27---35%), mean systolic blood pressure
patients underwent left ventricle ejection fraction (LVEF) (SBP) was 110 ± 17 mmHg (36% with systolic blood pres-
evaluation via transthoracic echocardiography using biplane sure <110 mmHg) and mean heart rate (HR) was 70 ± 10
Simpson method at our center within a six-month window bpm. With respect to GDMT, 94% patients were on renin-
before CPET. angiotensin-system inhibitors (RAASi), 90% on beta-blockers,
Maximal symptom-limited cardiopulmonary exercise 42% on mineralocorticoid receptor antagonists (MRA), and
testing (CPET) was performed and several parameters 50% were prescribed loop diuretics. A total of 32 patients
were assessed, including pVO2 , percent of predicted pVO2 (44%) had an implantable cardioverter-defibrillator and/or
(determined by the Wasserman equation6 ), minute ven- a cardiac resynchronization therapy device.
tilation/carbon dioxide output (VE/VCO2 ) slope, exercise Cardiopulmonary exercise testing results are illustrated
oscillatory ventilation (EOV) and respiratory exchange ratio in Table 1. Patients had a median pVO2 of 24 mL/kg/min
(RER), defined as the oxygen consumption/carbon dioxide (IQR 21---25 mL/kg/min), a median percentage of predicted
consumption ratio. Subjects were encouraged to exercise pVO2 of 85% (interquartile range (IQR) 74---91%) and a median
until RER was ≥1.10, as per site protocol. Twelve-lead VE/VCO2 slope of 32 (IQR 27---39). Overall, 75% had a RER
electrocardiogram, blood pressure and pulse oximetry were >1.10 and 23% had EOV.
monitored during CPET. Gas exchange was continuously The majority of patients had an NT-proBNP >125 pg/mL
assessed by face mask and determined by MedicalGraphics® (93%), with median NT-proBNP levels of 388 pg/mL (IQR:
and Carefusion® gas analyzer. Peak O2 consumption was 201---684 pg/mL) (Figure 1).
taken as the highest 20-second averaged VO2 during the
exercise period of CPET. Twenty-second averaged minute
Clinical outcomes
ventilation and carbon dioxide output, from the initiation
to the end of exercise, were used to calculate the VE/VCO2
slope. EOV was defined as an oscillatory pattern persisting During the two years of follow up after CPET, seven patients
for ≥60% of the test at an amplitude of ≥15% of the average (10%) met the primary outcome. Overall, three died (4%)
resting value.7 and five (7%) had at least one HF hospitalization (Table 2).
Demographic and clinical data, as well as laboratory Death was preceded by a HF hospitalization in only one of the
assessments were obtained from the electronic medical three patients. An event-free survival curve analysis for the
records. Laboratory evaluation (including NT-proBNP) was primary endpoint is shown in Figure 2. Sensitivity analysis
performed within seven days before CPET, as per center using a cut-off of predicted pVO2 ≥60% resulted in a similar
protocol. rate of events.

Discussion
Outcomes
In a real-world single-center HFrEF cohort, one in ten
The primary clinical endpoint was a composite of all-cause patients with good functional capacity diagnosed by CPET
death or HF hospitalization, whichever occurred first, at two had major clinical events over a two-year period of follow
years follow up after CPET. HF hospitalization was defined as up. More than 90% had persistent NT-proBNP elevations.
an admission for acute HF management.8 Sensitivity analysis Heart failure remains a high risk disease with a grim
included defining a good functional capacity using a pre- short-term prognosis. Our study cohort consisted of patients
dicted pVO2 ≥60% according to previous recommendations.9 receiving appropriate drug therapy (>90% BB and RAASi),

91
 S. Maltês, C. Brízido, B.M.L. Rocha et al.

Table 1 Patients’ baseline demographics.

Baseline demographics Overall cohort (N=72)
Age --- years, mean ± SD 51.6 ± 9.8
Male sex --- n (%) 60 (83.3)
BMI --- kg/m2 , median (IQR) 26.5 (25.0---28.7)
HF etiology --- n (%)
Ischemic 33 (45.8)
Valvular 3 (4.2)
Other 36 (50.0)
LVEF --- percentage, median (IQR) 32.0 (27.0---35.0)
NYHA classification --- n (%)
I 18 (25)
II 47 (65.3)
III 7 (9.7)
Mean systolic blood pressure --- mmHg, mean ± SD 110 ± 17
Systolic blood pressure<90 mmHg 4 (6)
Systolic blood pressure<100 mmHg 14 (20)
Systolic blood pressure<110 mmHg 24 (36)
Mean heart rate --- beats per minute, mean ± SD 70 ± 10
Laboratory data
Hemoglobin --- g/dL, mean ± SD 14.2 ± 1.4
Serum sodium --- mmol/L, mean ± SD 140.0 ± 4.4
eGFR (MDRD) --- mL/min/1.73 m2 , mean ± SD 92.7 ± 23.3
NT-proBNP --- pg/mL, median (IQR) 388 (201---684)
NT-proBNP≥125 pg/mL --- n (%) 67 (93.1)
NT-proBNP≥600 pg/mL --- n (%) 31 (43.1)
Comorbidities --- n (%)
Hypertension 27 (37.5)
Diabetes mellitus 9 (12.5)
AF 8 (11.1)
CKD (eGFR<60 mL/min/1.73 m2 by MDRD equation) 32 (45.1)
Medications --- n (%)
Beta-blockers 65 (90.3)
RAASi* 68 (94.4)
MRA 30 (41.7)
Loop diuretics 36 (50.0)
Devices --- n (%)
ICD 18 (25.0)
CRT-D 14 (19.4)
CPET variable
pVO2 --- mL/kg/min, median (IQR) 23.5 (21.3---25.4)
Percent of predicted pVO2 --- percentage, median (IQR) 84.7 (74.4---91.4)
VE/VCO2 slope --- L/min, median (IQR) 31.5 (27.1---38.9)
EOV ---n (%) 16 (22.9)
RER 1.10 (1.08---1.19)
>1.10, n (%) 54 (75)
>1.05, n (%) 68 (94)
AF: atrial fibrillation; CKD: chronic kidney disease; CPET: cardiopulmonary exercise testing; CRT: cardiac resynchronization therapy;
eGFR: estimated glomerular filtration rate; EOV: exercise oscillatory ventilation; HF: heart failure; HFSS: Heart Failure Survival Score;
ICD: implantable cardioverter defibrillator; IQR: interquartile range; LVEF: left ventricle ejection fraction; MECKI: Metabolic Exercise test
data combined with Cardiac and Kidney Indexes; MRA: mineralocorticoid receptor antagonist; NYHA: New York Heart Association; pVO2:
peak O2 consumption; RER: respiratory exchange ratio; SD: standard deviation; VE/VCO2 : minute ventilation/carbon dioxide production.
* Includes angiotensin-converting enzyme inhibitors, angiotensin receptor blockers and angiotensin receptors/neprilysin inhibitors.

92
 Revista Portuguesa de Cardiologia 42 (2023) 89---95

Figure 1 NT-proBNP boxplot.

Figure 2 CPET: cardiopulmonary exercise testing; HF: heart failure.

who seemed clinically well (good functional capacity and event was CV death with no preceding HF hospitalization
half not requiring a loop diuretic). Nevertheless, 10% died or in 51% of these most clinically stable patients, therefore
had at least one HF hospitalization within two years. Similar precluding patients and physicians from an alert that could
observations were previously reported in the PARADIGM-HF have prompted treatment intensification and optimization
(Angiotensin---Neprilysin Inhibition versus Enalapril in Heart to avoid this dismal outcome.11 Overall, 60% of those CV
Failure) trial. Among the least symptomatic patients in NYHA deaths were sudden cardiac deaths.11 In our study, death
class I or II, 22% either died from a cardiovascular cause or was preceded by a HF hospitalization in only one of three
had at least one HF hospitalization during a median follow up patients.
of 27 months.10 Furthermore, among the more clinically sta- In HFrEF, evidence of subclinical disease progression can
ble patients without any history of HF hospitalization, 20% be found even in patients who seem to be doing well and to
either died from a cardiovascular cause or had at least one be receiving appropriate treatment. Neurohormonal activa-
HF hospitalization during the course of the trial. The primary tion persisted in the majority of our study patients, although

93
 S. Maltês, C. Brízido, B.M.L. Rocha et al.

analyses of PARADIGM-HF and HEAAL (effects of high-dose

Table 2 Primary and secondary outcomes.
versus low-dose losartan on clinical outcomes in patients
Primary outcome with heart failure) trials.18,20
Combined endpoint of all-cause death 7 (9.7) Our findings strengthen previous recommendations:
or HF admission at two-years, n (%) GDMT initiation and uptitration is essential, since even
clinically stable, ‘low risk’ patients, with good functional
Secondary outcomes
capacity have persistent neurohormonal activation and a sig-
All cause death at two-years, n (%) 3 (4.2)
nificant rate of events. Accordingly, viewing chronic HF as a
HF admission at two-years, n (%) 5 (6.9)
stable disease may be a misnomer and, therefore, must be
Heart transplantation, n (%) 0 (0)
avoided.
HF: heart failure. This study has some limitations. First, it is a single-center
retrospective study with a limited sample size. Second, we
exercise tolerance, as judged by NHYA class, was normal or did not assess other important adverse events related to
only mildly impaired in >90%. Additional evidence of this HF, namely cardiovascular death, malignant arrythmias or
illusion of clinical stability derives from studies using tro- implantable cardioverter defibrillator shocks. Third, we did
ponin measurements where the majority of stable HFrEF not assess whether GDMT was being taken at maximum rec-
patients in NYHA class II have chronic cardiac troponin ele- ommended dose. However, in order to estimate if our cohort
vations, corroborating the subclinical continuous myocardial was close to (or distant from) optimal GDMT we assessed
lesion.12 Several mechanisms can lead to troponin elevation patients’ SBP and HR as surrogates --- assuming that those
in HF including supply-demand mismatch, which may occur presenting a low SBP/HR would not tolerate further up-
in the absence of coronary obstruction due to increased oxy- titration of renin-angiotensin-aldosterone inhibitors and/or
gen demand related to increased wall tension, anemia, or beta-blockers. Only 36% of patients had a systolic blood pres-
other factors provoking subendocardial injury. Non-coronary sure <110 mmHg. Although it may be possible that some
triggers include cellular necrosis or apoptosis in the con- patients with higher SBP and HR were still in the process of
text of wall stress, as well as the toxic effects of circulating up-titration, already on maximal doses of disease-modifying
neurohormones, toxins, inflammation, and infiltrative pro- drugs or unable to reach these because of intolerance or
cesses. Regardless of the mechanism, troponin elevation in other limitations (e.g., kidney failure or hyperkalemia), it
HF is associated with greater tendency toward LV remod- seems reasonable to consider that a significant proportion of
eling, and a higher risk of death or hospitalization.12,13 the patients in our cohort would be indeed amenable to fur-
Recent evidence indicates that changes in NT-proBNP lev- ther drug up-titration. Fourth, included patients performed
els are predictive of similar directed changes in troponin an echocardiographic assessment within six months of CPET.
levels, which in turn are associated with left ventricular for- We cannot exclude a possible worsening or improvement of
ward remodeling.14 Nevertheless, in our population, we had LGEF within this time frame.
only access to baseline NT-proBNP levels, and serial assess-
ments of cardiac biomarkers were beyond the scope of this
study. Conclusion
On the basis of the above-mentioned observations, GDMT
uptitration to maximum tolerated doses or maximum tested
In a clinically stable HFrEF population displaying preserved
doses is paramount and currently recommended in European
functional capacity, persistent neurohormonal activation
Society of Cardiology HF guidelines.1 However, the afore-
was present in the majority and one in ten patients died
mentioned illusion of stability may be hazardous and lead
or had one HF admission at two years post-CPET. These
to treatment inertia, particularly when dealing with chronic
findings rightfully challenge the perception that treatment
individuals not overwhelmingly symptomatic. Adherence to
optimization may be less important for low-symptomatic
GDMT recommendations by physicians has traditionally been
patients. Uptitration of pharmacological therapy to maxi-
shown to be suboptimal and clinical stability may be one
mal tolerated or tested doses is mandatory to slow disease
of the culprit reasons.15,16 Indeed, less than one in five
progression as effectively as possible, even when the clinical
patients admitted for acute HF and discharged on beta-
scenario seems (deceptively) favorable. Therefore, strate-
blockers had these uptitrated in the following month.16
gies are urgently needed to motivate clinicians to pursue
These observations account for a reduced number of HFrEF
GDMT even in such patients.
patients achieving GDMT-target doses as demonstrated in
the IMPROVE-HF registry.5
Such inertia is particularly important in stable, pauci-
symptomatic patients, where misperception of risk and Authors’ contributions
concerns regarding side effects may deter clinicians and
patients from appropriate treatment intensification.3 Yet, SM wrote the first draft of this article; CB, GJL, BR, and
evidence shows that even less symptomatic, NYHA class II FG incorporated feedback in subsequent drafts and revi-
patients benefit from appropriate GDMT. Indeed, such cases sions; CS, PF, AD, AT, AV, CA and MM contributed to revisions
often account for a significant number (if not the majority) and reviewed the final draft; SM composed the figures and
of patients included in landmark HF trials.17---20 Moreover, submitted the final version of this article, on behalf of all
the benefits of GDMT in less symptomatic patients are the authors. All named authors complied with the Interna-
at least similar in magnitude to those observed in more tional Committee of Medical Journal Editors guidelines for
symptomatic patients, as shown in pre-specified subgroup authorship.

94
 Revista Portuguesa de Cardiologia 42 (2023) 89---95

Ethical approval in clinical trials: a report of the American College of Cardi-

ology/American Heart Association Task Force on Clinical Data
Standards (Writing Committee to Develop Cardiovascular End-
This investigation was conducted in accordance with the
points Data Standards) [published correction appears in J Am
World Medical Association Declaration of Helsinki (seventh Coll Cardiol. J Am Coll Cardiol. 2015;66:403---69 [published cor-
revision, Fortaleza, 2013) and the Declaration of Istanbul rection appears in J Am Coll Cardiol 2015;66(8):982].
(2008). This study was exempt from ethical approval. 9. Riebe D, Ehrman JK, Liguori G. ACSM guidelines for exercise
testing and prescription. tenth edition Wolters Kluwer; 2018.
Funding 10. Mogensen UM, Gong J, Jhund PS, et al. Effect of sacu-
bitril/valsartan on recurrent events in the prospective
comparison of ARNI with ACEI to determine impact on global
None declared. mortality and morbidity in Heart Failure trial (PARADIGM-HF).
Eur J Heart Fail. 2018;20:760---8.
Conflicts of interest 11. Solomon SD, Claggett B, Packer M, et al. Efficacy of sacu-
bitril/valsartan relative to a prior decompensation: the
The authors have no conflicts of interest to declare. PARADIGM-HF trial. JACC Heart Fail. 2016;4:816---22.
12. Januzzi JL Jr, Filippatos G, Nieminen M, et al. Troponin elevation
in patients with heart failure: on behalf of the third Univer-
References sal Definition of Myocardial Infarction Global Task Force: Heart
Failure Section. Eur Heart J. 2012;33:2265---71.
1. Ponikowski P, Voors AA, Anker SD, et al. ESC guidelines for 13. Kociol RD, Pang PS, Gheorghiade M, et al. Troponin elevation
the diagnosis and treatment of acute and chronic heart fail- in heart failure prevalence, mechanisms, and clinical implica-
ure: the Task Force for the diagnosis and treatment of acute tions. J Am Coll Cardiol. 2010;56:1071---8.
and chronic heart failure of the European Society of Cardi- 14. Murphy SP, Prescott MF, Maisel AS, et al. Association
ology (ESC). Developed with the special contribution of the between angiotensin receptor-neprilysin inhibition, cardiovas-
Heart Failure Association (HFA) of the ESC. Eur J Heart Fail. cular biomarkers, and cardiac remodeling in heart failure with
2016;18:891---975. reduced ejection fraction. Circ Heart Fail. 2021;14:e008410.
2. Jones NR, Roalfe AK, Adoki I, et al. Survival of patients with 15. Wirtz HS, Sheer R, Honarpour N, et al. Real-world analysis of
chronic heart failure in the community: a systematic review guideline-based therapy after hospitalization for heart failure.
and meta-analysis. Eur J Heart Fail. 2019;21:1306---25. J Am Heart Assoc. 2020;9:e015042.
3. Butler J, Gheorghiade M, Metra M. Moving away from symptoms- 16. Fiuzat M, Ezekowitz J, Alemayehu W, et al. Assessment of lim-
based heart failure treatment: misperceptions and real risks for itations to optimization of guideline-directed medical therapy
patients with heart failure. Eur J Heart Fail. 2016;18:350---2. in heart failure from the GUIDE-IT trial: a secondary analysis of
4. Kocabaş U, Kıvrak T, Yılmaz Öztekin GM, et al. Adherence to a randomized clinical trial. JAMA Cardiol. 2020;5:757---64.
guideline-directed medical and device therapy in outpatients 17. Packer M, Poole-Wilson PA, Armstrong PW, et al. Comparative
with heart failure with reduced ejection fraction: the ATA study. effects of low and high doses of the angiotensin-converting
Anatol J Cardiol. 2020;24:32---40. enzyme inhibitor, lisinopril, on morbidity and mortality in
5. Gheorghiade M, Albert NM, Curtis AB, et al. Medication dos- chronic heart failure. Circulation. 1999;100:2312---8.
ing in outpatients with heart failure after implementation of 18. Konstam MA, Neaton JD, Dickstein K, et al. Effects of high-dose
a practice-based performance improvement intervention: find- versus low-dose losartan on clinical outcomes in patients with
ings from IMPROVE HF. Congest Heart Fail. 2012;18:9---17. heart failure (HEAAL study): a randomised, double-blind trial.
6. Arena R1, Myers J, Guazzi M. Cardiopulmonary exercise testing Lancet. 2009;374:1840---8.
is a core assessment for patients with heart failure. Congest 19. Bristow MR, Gilbert EM, Abraham WT, et al. Carvedilol pro-
Heart Fail. 2011;17:115---9. duces dose-related improvements in left ventricular function
7. Guazzi M, Adams V, Conraads V, et al. Clinical recommendations and survival in subjects with chronic heart failure. MOCHA Inves-
for cardiopulmonary exercise testing data assessment in specific tigators. Circulation. 1996;94:2807---16.
patient populations. Circulation. 2012;126:2261---74. 20. McMurray JJ, Packer M, Desai AS, et al. Angiotensin-neprilysin
8. Hicks KA, Tcheng JE, Bozkurt B, et al. 2014 ACC/AHA key data inhibition versus enalapril in heart failure. N Engl J Med.
elements and definitions for cardiovascular endpoint events 2014;371:993---1004.

95