Download Heart Failure with a Preserved Ejection Fraction

HEART FAILURE
|
REVIEW
Heart Failure with a Preserved Ejection Fraction:
From Pathophysiology to Biomarkers … and Beyond!
Gilles De Keulenaer, MD, PhD
Center for Heart Failure and Cardiac Rehabilitation, AZ Middelheim – University of Antwerp
Received 14/2/2011, Reviewed 19/2/2011, Accepted 21/2/2011
DOI: 10.5083/ejcm.20424884.37
ABSTRACT
CORRESPONDENCE
Diagnosing and managing heart failure according to the left ventricle’s ejection fraction (LVEF) has
become part of evidence-based medicine. Not surprisingly, LVEF - a powerful prognostic factor
in heart failure - has caused a marked heterogeneity in the clinical benefit of various therapeutic
interventions. From a pathophysiological point of view, however, many disease characteristics are
shared among the entire heart failure spectrum (from low to high LVEF). The many functional and
anatomical differences within the spectrum are merely quantitative, with an extensive overlap
between the extremes of the spectrum and belonging to the same linear relation when plotted
against LVEF. Therefore, although counter-intuitive from a clinical point of view, from a pathophysiological point of view heart failure seems to progress along a common disease trajectory
independently of LVEF. In this review, we will scrutinise this apparent paradox, estimate how it
relates to the recent biomarker-oriented (as opposed to a classic LVEF-oriented) approach to heart
failure and discuss to what extent it may affect conceptual progress in chronic heart failure.
Gilles De Keulenaer, MD, PhD
Center for Heart Failure and
Cardiac Rehabilitation
AZ Middelheim –
University of Antwerp
Lindendreef 1,
2020 Antwerp, Belgium
INTRODUCTION
Pathophysiology
More than thirty years ago, it was recognised that
signs and symptoms of heart failure may occur in
patients in whom the ejection fraction of the left
ventricle (LV) is within normal limits. Although originally studied in patients with genuine hypertrophic
cardiomyopathy, heart failure with “preserved” LV
ejection fraction (HFPEF) has now been recognised
in a more common population, which is characterised by a high prevalence of female gender, arterial
hypertension, advanced age, obesity and diabetes
mellitus.
The pathophysiology of HFPEF consists of two
parts. First, it explains clinical symptoms of HFPEF. Second, it clarifies its irreversible pathophysiological progression. Although these
pathophysiological aspects may be related, lessons from the past should remind us that a therapeutic intervention that reduces symptoms in
heart failure do not per se halts the progression
of the disease, and vice versa.
Interestingly, community surveys on chronic heart
failure in general have shown that the overall distribution of left ventricular ejection fraction (LVEF)
among patients with heart failure is bell-shaped1
and that the fraction of patients with a preserved
LVEF within this distribution is much larger than
anticipated, currently being about 50% of patients2.
Strikingly, in contrast to patients with a reduced
LVEF (HFREF), no improvements in outcomes have
been documented over the past twenty years.
Symptoms of HFPEF are characterised by exercise intolerance and breathlessness on exertion.
These chronic symptoms may be alternated by
acute episodes of lung edema or global volume
overload, usually precipitated by severe hypertension, non-compliance to therapy, (pulmonary) infection, inappropriate tachycardia (usually atrial fibrillation), or a combination of two or
more of these factors3.
In this paper, we will review pathophysiological
insights underlying both symptoms and disease
progression in HFPEF and contemplate in which direction research should develop in the next decade.
90
Tel: +32 3 265 23 38
Fax: +32 3 265 24 12
E-mail:
[email protected]
Pathophysiology of HFPEF symptoms
In many patients with suspected HFPEF an alternative non-cardiac explanation for symptoms
can be identified (like obesity, muscular deconditioning, pulmonary disease or angina), which
may lead to misdiagnosis4. This is a relevant
problem in clinical cardiology, which complicates the design of clinical trials in HFPEF and
urges defining guidelines for diagnosis.
ISSN 2042-4884
EUROPEAN JOURNAL OF CARDIOVASCULAR MEDICINE
VOL I ISSUE III
THE HEART FAILURE SPECTRUM
Nevertheless, it is now well established that patients with HFPEF
also suffer from severe cardiovascular abnormalities that exacerbate exercise intolerance, which firmly increase patient’s risk for
episodes of acute heart failure. These abnormalities predominantly
manifest during active LV relaxation and diastolic filling, leading
to so-called “diastolic dysfunction”. The entire pathophysiological
picture of HFPEF as a syndrome is, however, more complex than
an isolated problem of LV diastolic function. Abnormalities of LV
relaxation and filling consist of impaired active LV recoil and suction, blunted LV lusitropic response to adrenergic stimulation and
a steep diastolic LV pressure-volume relation5. When occurring in
combination and often exacerbated by deranged ventriculo-vascular coupling, these abnormalities explain the low LV preload reserve
and the low pulmonary capillary wedge pressure (PCWP)/work ratio of HFPEF patients. By increasing PCWP, they may also explain the
high prevalence of pulmonary arterial hypertension in HFPEF, unless other yet to be identified pathophysiological processes would
lead to pre-capillary pulmonary hypertension6.
Accordingly, the pathophysiology of clinical symptoms of patients
with HFPEF is predominated by diastolic LV dysfunction. The subcellular processes leading to diastolic LV dysfunction are beyond
scope of this review but include an abnormal inactivation of the
excitation-contraction process, passive myocyte stiffness, extracellular matrix abnormalities and impaired paracrine endothelial signalling.
Figure 1: Clinical versus pathophysiological interpretation of the
heterogeneity heart failure.
Although the LVEF-based approach of clinical trial design seems to constitute
a platform to dichotomise heart failure in two separate disease entities, from
a pathophysiological point of view, heart failure phenotypes are strongly
related, and form a spectrum of overlapping phenotypes.
Pathophysiology of HFPEF progression
Apart from the fact that HFPEF is associated with diastolic LV dysfunction, it is also usually associated with a progressive concentric
LV remodelling. Similarly, HFREF is associated with systolic dysfunction, but also with progressive eccentric LV remodelling. These associations between function and morphology of the left ventricle,
together with the LVEF-based design of clinical trials, have created
a tendency to dichotomise heart failure in two separate pathophysiological entities, apparently progressing along two divergent
pathophysiological trajectories (Figure 1).
However, diastolic dysfunction is not unique for HFPEF; it even is
the strongest determinant of symptoms in HFREF7. Also, systolic
dysfunction is not unique for HFREF8. In addition, concentric and
eccentric LV remodelling often occur in combination, independently of LVEF. These observations suggest that HFPEF and HFREF are
related, overlapping entities and may progress along a common
pathophysiological trajectory (Figure 1).
Consistently, there currently is not a single pathognomonic feature,
at any level of biological complexity (gene, protein, cell, organ, or
organ system) that distinguishes HFPEF from HFREF. Instead, HFPEF
and HFREF share many pathophysiological features, such as neurohormonal abnormalities, upregulation of growth factors, volume
overload, ventricular collagen turnover, titin isoform switching and
titin phosphorylation deficits, endothelial dysfunction, atrial dysfunction, arterial stiffening, etc.) . The genuine differences between
HFPEF and HFREF, as described in literature (e.g. longitudinal contractile function9, serum brain natriuretic peptide10, LV end diastolic volume11 and cardiomyocyte diameter12) are merely quantitative
differences, usually presented as statistically significant different
mean values, but with an extensive overlap of the individual data
points. Hence, LVEF is a rather weak predictor of cardiomyocyte diameter and even of LV volumes and thus fails to capture the type of
LV remodelling in a reliable fashion13. When individual data points
of HFPEF and HFREF are plotted against LVEF, they belong to same
(sometimes weak) linear relation14.
EUROPEAN JOURNAL OF CARDIOVASCULAR MEDICINE
VOL I ISSUE III
Why is it not surprising that we fail to recognise two clear-cut
distinct types of structural and functional LV remodelling? LV remodelling is the product of multiple interacting and complex signalling processes. The contribution of each of these processes is
linked to the patient’s biological profile and medical background.
For example, some of these processes are triggered by myocardial
ischemia, myocardial infection or type I diabetes; these processes
seem to promote predominant eccentric remodeling. Other signaling processes are influenced by type II diabetes, obesity, arterial
hypertension and female gender; these tend to promote concentric remodelling. The biochemistry of these signalling processes are
under intense investigation and has been related to intermediate
mediators such as leptin, oxidative stress, hyperinsulinemia, estrogen, cytokines, etc. In vivo, these mediators and their complex signalling events merge in qualitative and quantitative combinations,
specific for each patient. In a population of patients with heart failure, this creates heterogeneity, leading to a continuous spectrum
of overlapping phenotypes15. Any attempt to dichotomising this
spectrum is artificial and lacks a conceptual basis.
According to the above reasoning, HFPEF and HFREF progress
along the same pathophysiological trajectory, and only vary in the
quantitative contribution of numerous signaling events. So why
then, one may argue, have clinical trials on the inhibition of the renine-angiotensin system, been successful in HFREF and disappointing in HFPEF? Does this not contradict the premise that HFPEF and
HFREF are part of a common disease trajectory?
No, it doesn’t. Imposing an arbitrary cut-off in any of the many
prognostic continuous variables of heart failure, be it LVEF or any of
the other currently available biomarkers, does not necessarily signify that a novel paradigm is generated or that disease taxonomy
should be introduced, even if it unveils a different clinical response
to a therapeutic intervention.
91
HEALTHCARE BULLETIN
|
HEART FAILURE
What if a clinical trial would indicate that heart failure patients with
a normal maximal oxygen consumption (VO2 max) would show
less benefit to aldosteron inhibitors as compared to patients with
a reduced VO2 max? Would this be a sufficient argument to claim
that heart failure with a normal or reduced VO2 max progress along
different disease trajectories? No, it wouldn’t. Yet, we might still
consider the result of the trial as clinically relevant. The same is true
for LVEF or any other (prognostic) variable, usually coined as “biomarker” of heart failure.
CONCLUSIONS
Is it clinically relevant to distinguish HFNEF and
HFREF during the clinical management of heart failure?
Yes, it does. Diagnosis of heart failure can be challenging when
LVEF is normal and necessitates specific diagnostic guidelines. In
addition, haemodynamic and prognostic management is different
in HFNEF and HFPEF.
“Biomarkers”: rising stars of heart failure
Does this mean that HFNEF and HFREF are different diseases?
There is a current trend to characterise and manage heart failure
patients by the measurement of one or multiple biomarkers16. New
biomarkers have emerged from large-scale quantitative analyses of
gene expression, including cDNA microarrays and proteomic analyses. This multi-marker strategy looks refreshing and may provide
new insights. At least, it surpasses patient management based on
a dogmatic cut-off of LVEF alone. It may personalise heart failure
management and help to optimise current heart failure guidelines.
On the other hand, one may seriously question where-to to this approach will eventually lead. It may be feared that adding still more
biomarkers and other disease parameters in heart failure will never
end and fail to re-launch conceptual thinking. Time has perhaps
come to envisage integrative approaches, already introduced in
other fields of life sciences, that encounter limits of reductionism
and seek understanding out of myriad validated bits of data17.
No, it doesn’t. HFNEF and HFREF share many disease characteristics, and the observed pathophysiological differences are merely
quantitative, with a large overlap between individual data points.
This indicates that HFNEF and HFREF progress along the same
pathophysiological disease trajectory, although the end result is a
heterogeneous disease consisting of a spectrum with different (but
overlapping) clinical phenotypes. This heterogeneity is caused by
patient’s biological “background”, such as age, gender, metabolism,
etc which influences the process of LV remodelling.
Systems biology aims to provide a framework for how structural
and functional components, often known as disease biomarkers
in clinical medicine, interact in self-organising modular biological
networks. Networks rather than the components themselves create
physiological behaviour and disease. Each node in a network represents a component (a gene, a transcript, an organelle, a cell, …)
and interconnecting nodes describe a typical architecture, which
is imposed by biological evolution and selection. At each level of
complexity, a network obtains new properties, which are not predicted from the properties of the network at the lower level, referring to emerging properties in a dissipative structure, introduced
by Prigogine. Within a network perspective to biology, a disease is
defined as the failure of a biological network or as the failure to
obtain a next-level emerging property.
Rather than further studying heart failure and its pathophysiology
in terms of a preserved or a reduced LVEF, one might better start to
begin thinking on the network failures in chronic heart failure. What
are the crucial biological networks (in the network of networks) of
cardiac function and heart failure, and where are the vulnerable
hubs which can destabilise networks? Can network perspective to
heart failure provide novel biological (organ-specific) fingerprints
of failure that allow predicting a patient’s risk, early disease development or disease stage? How do these fingerprints relate to the
current growing list of heart failure biomarkers? Can systems biology help to define novel surrogate endpoints for clinical trials? How
can a network, if associated with a disease, be targeted pharmacologically? How does this relate to the growing list of targets, now
emerging from reductionist sciences?
These are the future scientific questions of chronic heart failure,
independently of whether LVEF is preserved or reduced. Hence, although relevant in daily clinical cardiology, from a conceptual and
scientific perspective the value of LVEF in heart failure is becoming
irrelevant.
92
Why then do HFNEF and HFREF respond differently to
ACE-inhibitors?
It is not unusual that arbitrary cut-off values of prognostic variables
of a disease, such as LVEF > 50% in chronic heart failure, induce differences in the clinical response to therapy. This does not indicate,
however, that this prognostic variable constitutes a good platform
to introduce disease taxonomy with distinct pathophysiological
entities. There is little doubt that cut-off values of many other variables in heart failure (e.g. BNP, VO2 max, etc.) would induce a similar
variability in the response to therapeutic interventions.
If LV ejection fraction > 50% is not a good basis to introduce
disease taxonomy in heart failure, what about biomarkers?
Biomarkers are interesting tools indeed and can be refreshing in
heart failures sciences during the endeavour to optimise therapeutic guidelines and personalise medicine. On the other hand,
the number of biomarkers is rapidly increasing, and the biological
meaning of these biomarkers are often so obscure that a biomarker-oriented approach may soon become very confusing.
How then should we proceed? Heart failure research is
stagnating, and pharmaceutical companies are discouraged
by the negative results of many expensive trials.
Reductionistic and high-trough put sciences have yielded a massive
amount of new and exciting data in heart failure over the past ten
years. New molecular pathways, therapeutic targets and biomarkers have been defined. It is now time to seek understanding out of
myriad validated bits of data. This can happen only if we embrace
systems approaches to heart failure. These approaches will define
biological networks and dysfunction of networks in the pathophysiology of chronic heart failure. The strategy of approaching heart
failure from one arbitrary variable, like LVEF, has reached its limits.
EUROPEAN JOURNAL OF CARDIOVASCULAR MEDICINE
VOL I ISSUE III
THE HEART FAILURE SPECTRUM
REFERENCES
1
Solomon SD, Anavekar N, Skali H at al. Candesartan in Heart Failure
Reduction in Mortality (CHARM) Investigators. Influence of ejection
fraction on cardiovascular outcomes in a broad spectrum of heart
failure patients. Circulation. 2005;112:3738-44.
2
Owan TE, Redfield MM. Epidemiology of diastolic heart failure.
Prog Cardiovasc Dis. 2005;47:320-32.
3
Klapholz M, Maurer M, Lowe AM, et al. Hospitalization for heart
failure in the presence of a normal left ventricular ejection fraction:
results of the New York Heart Failure Registry. J Am Coll Cardiol.
2004;43:1432-8.
4
Caruana L, Petrie MC, Davie AP, McMurray JJ. Do patients with
suspected heart failure and preserved left ventricular systolic
function suffer from “diastolic heart failure” or from misdiagnosis?
A prospective descriptive study. BMJ. 2000;321:215-8.
5
Paulus WJ. Culprit mechanism(s) for exercise intolerance in heart
failure with normal ejection fraction. J Am Coll Cardiol. 2010;56:864-6.
6
Lam CS, Roger VL, Rodeheffer RJ, Borlaug BA, Enders FT, Redfield MM.
Pulmonary hypertension in heart failure with preserved ejection
fraction: a community-based study. J Am Coll Cardiol. 2009;53:
1119-26.
7
8
Ommen SR, Nishimura RA, Appleton CP et al. Clinical utility of
Doppler echocardiography and tissue Doppler imaging in the
estimation of left ventricular filling pressures: a comparative
simultaneous Doppler-catheterization study. Circulation.
2000;102:1788–1794.
9
Yip G, Wang M, Zhang Y et al. Left ventricular long axis function in
diastolic heart failure is reduced in both diastole and systole: time for
a redefinition? Heart. 2002;87:121–125
10
Januzzi JL, van Kimmenade R, Lainchbury J et al. NT-proBNP
testing for diagnosis and short-term prognosis in acute destabilized
heart failure: an international pooled analysis of 1256 patients:
the International Collaborative of NT-proBNP Study. Eur Heart J.
2006;27:330-7
11
He KL, Burkhoff D, Leng WX et al. Comparison of ventricular structure
and function in Chinese patients with heart failure and ejection
fractions >55% versus 40% to 55% versus <40%. Am J Cardiol.
2009;103:845-51
12
van Heerebeek L, Borbély A, Niessen HW, et al. Myocardial structure
and function differ in systolic and diastolic heart failure. Circulation.
2006;113:1966–1973
13
De Keulenaer GW, Brutsaert DL. Systolic and diastolic heart failure
are overlapping phenotypes within the heart failure spectrum.
Circulation 2011 (in press)
14
De Keulenaer GW, Brutsaert DL. The heart failure spectrum: time
for a phenotype-oriented approach. Circulation. 2009;119:3044-6
15
Shah AM, Solomon SD. A unified view of ventricular remodelling.
Eur J Heart Fail. 2010;12:779-81
16
Braunwald E. Biomarkers in heart failure. N Engl J Med.
2008;358:2148-59
17
Lander AD. The edges of understanding. BMC Biol 2010;8:40-42
Tan YT, Wenzelburger F, Lee E et al. The pathophysiology of heart
failure with normal ejection fraction: exercise echocardiography
reveals complex abnormalities of both systolic and diastolic
ventricular function involving torsion, untwist, and longitudinal
motion. J Am Coll Cardiol. 2009;54:36-46.
EUROPEAN JOURNAL OF CARDIOVASCULAR MEDICINE
VOL I ISSUE III
93