Vous êtes sur la page 1sur 12

Archives of Cardiovascular Disease (2008) 101, 361—372

Disponible en ligne sur www.sciencedirect.com

REVIEW

Treatment of heart failure with preserved systolic


function
Traitement de l’insuffisance cardiaque à fonction systolique préservée

P. de Groote a,∗, D. Herpin b, F. Diévart c, O. Hanon d,


J.N. Trochu e, J.Y. Artigou f, M. Galinier g, G. Habib h,
Y. Juillière i, Y. Neuder j, R. Roudaut k, M. Komajda l

a
Pôle cardiologie et médicine vasculaire, hôpital cardiologique, CHRU de Lille, boulevard du
Professeur-J.-Leclercq, 59037 Lille cedex, France
b
Service cardiologie, faculté de medicine et de pharmacie, CHU Poitiers, université Poitiers,
86021 Poitiers, France
c
Clinique Villette, 18, rue Parmentier, 59240 Dunkerque, France
d
Département de gériatrie, hôpital Broca, AP—HP, université Paris-Descartes—Paris-5,
54, rue Pascal, 75013 Paris, France
e
Clinique cardiologique et des maladies vasculaires, institut du thorax, CHU de Nantes,
boulevard Jacques-Monod, 44093 Nantes, France
f
Service de cardiologie, hôpital Avicenne, 125, route Stalingrad, 93009 Bobigny cedex, France
g
Pôle cardiovasculaire et métabolique, CHU Rangueil, TSA 50032, 31409 Toulouse cedex 9,
France
h
Service de cardiologie, hôpital la Timone, boulevard Jean-Moulin, 13005 Marseille, France
i
Département de cardiologie, CHU Nancy-Brabois, rue de Morvan, 54500
Vandoeuvre-les-Nancy, France
j
Service de cardiologie, CHU de Grenoble, B.P. 217, 38043 Grenoble cedex 9, France
k
Hôpital cardiologique Haut-Leveque, avenue Magellan, 33604 Pessac, France
l
Département de cardiologie, hôpital Pitié Salpetrière, AP—HP, université
Pierre-et-Marie-Curie—Paris-6, 47-83, boulevard de l’Hôpital, 75013 Paris, France

Received 10 December 2007; accepted 25 April 2008


Available online 7 July 2008

KEYWORDS Summary Heart failure is a major public health problem. Heart failure with preserved systolic
Heart failure; function (HF-PSF) is a common form, which is difficult to diagnose. Results of recent studies show
Diastolic heart that HF-PSF has a poor prognosis, with an annual survival rate similar to that of heart failure
failure; with left ventricular systolic dysfunction. Despite these findings, the therapeutic management
Diastolic dysfunction; of HF-PSF is not clearly defined. We will discuss in this review of the literature the current

∗ Corresponding author. Fax: +33 3 20 44 48 81.


E-mail address: pdgroote@chru-lille.fr (P. de Groote).

1875-2136/$ — see front matter © 2008 Published by Elsevier Masson SAS.


doi:10.1016/j.acvd.2008.04.008
362 P. de Groote et al.

therapeutic management of HF-PSF, including the role of precipitating factors such as hyper-
Preserved ejection tension, myocardial ischaemia and supraventricular arrhythmias, and the main results of
fraction; epidemiological registries and randomized controlled clinical trials in this disease. Only four
Treatment large therapeutic trials have assessed the impact of different classes of drugs (digoxin,
angiotensin II converting enzyme inhibitors, angiotensin II receptors type I blockers and beta-
blockers) on morbidity and mortality in HF-PSF. Results of these trials are disappointing. Apart
from the beta-blockers, the other three classes of drugs did not show benefit on the outcome
of the disease. Moreover, the results of the beta-blocker trial are controversial as a mixed
population of heart failure with and without preserved systolic function was studied. Finally,
the current therapeutic management of patients with HF-PSF is still based on our pathophys-
iological knowledge: education, low salt diet, diuretics, slowing heart rate and controlling
triggering factors. Other large randomized controlled multicenter trials, which may help us in
the understanding of HF-PSP and its therapeutic management, are ongoing.
© 2008 Published by Elsevier Masson SAS.

MOTS CLÉS Résumé L’insuffisance cardiaque est un problème majeur de santé publique. L’insuffisance
Insuffisance cardiaque à fonction systolique préservée (IC-FSP) est une forme fréquente dont le diagnostic
cardiaque ; est difficile. Les résultats des études récentes montrent que le pronostic de l’IC-FSP est mauvais
Insuffisance avec une survie annuelle proche de celle de l’insuffisance cardiaque par dysfonction systolique
cardiaque du ventricule gauche. Malgré ces données, la prise en charge thérapeutique de l’IC-FSP n’est
diastolique ; pas bien codifiée. Dans cet article, nous proposons une revue de la littérature sur cette prise en
Dysfonction charge thérapeutique. Nous nous sommes d’abord intéressés aux facteurs précipitant jouant un
diastolique ; rôle fréquent dans les épisodes de décompensation cardiaque : hypertension artérielle, ischémie
Fraction éjectée coronaire et troubles du rythme supraventriculaire. Puis, nous avons résumé la prise en charge
préservée ; thérapeutique des patients atteints d’IC-FSP inclus dans les registres épidémiologiques. Enfin,
Traitement les données issues des études thérapeutiques randomisées et contrôlées, peu nombreuses dans
cette pathologie, sont analysées. Seules quatre études d’envergure ont évalué l’impact cli-
nique de certaines classes médicamenteuses : la digoxine, les antagonistes des récepteurs de
l’angiotensine II, les inhibiteurs de l’enzyme de conversion de l’angiotensine II et les bêtablo-
quants. Les résultats de ces études sont décevants. Hormis, les bêtabloquants, les trois autres
classes médicamenteuses n’ont pas démontré d’effets bénéfiques sur l’évolution de la patholo-
gie. Pour les bêtabloquants, les résultats sont controversés, car la seule grande étude a inclus
une population mixte d’insuffisant cardiaque avec et sans fonction systolique préservée. Finale-
ment, en 2008, la prise en charge thérapeutique des patients ayant une IC-FSP reste basée sur
nos connaissances physiopathologiques : éducation, régime pauvre en sel, diurétiques, ralen-
tissement de la fréquence cardiaque et contrôle des facteurs déclenchant. D’autres études
multicentriques randomisées et contrôlées sont en cours qui nous permettront peut-être de
mieux comprendre l’IC-FSP et de mieux la prendre en charge.
© 2008 Published by Elsevier Masson SAS.

Heart failure with preserved systolic function (HF-PSF) is a based on our pathophysiological knowledge. The major
common, complex and severe disease. Its diagnosis is not factor in this disease is the increase in left ventricu-
straightforward as it often affects elderly patients suffering lar end-diastolic pressure. Aim of the treatment is to
from multiple diseases. In order to better understand HF- reduce the end-diastolic pressure and to avoid its sud-
PSF, a French working group (the TACTIC group), proposed den rise in certain circumstances. Kitzman et al. showed
a diagnostic algorithm based on the patient’s clinical data that ventricular filling pressures increase significantly with
and on the results of further investigations (Fig. 1) [1]. A exercise in patients with isolated diastolic dysfunction
working group of the European Society of Cardiology pub- [3].
lished a similar process leading to the same conclusions [2]. Together with the specific treatment of HF-PSF, pre-
In the present article, the TACTIC group has tried to review cipitating factors or concomitant (or causative) diseases
from the literature the current therapeutic management of involved in HF-PSF must be corrected. These are hyper-
HF-PSF. For this purpose, our conclusions were based on arti- tension, myocardial ischaemia and supraventricular arrhyth-
cles published from 1985 to date and selected from medical mias. Isolated left ventricular hypertrophy (LVH) or LVH
databases using the following keywords: heart failure, dias- complicating hypertension predispose to the clinical expres-
tolic heart failure, diastolic dysfunction, preserved ejection sion of HF-PSF. Finally, HF-PSF generally occurs in the elderly
fraction, treatment. patient, who often has several comorbidities such as renal
In the absence of significant results from random- failure, which contribute to the symptoms and the deterio-
ized trials, the current treatment of HF-PSF is still ration of the disease.
Treatment of heart failure with preserved systolic function 363

Pathophysiological mechanisms of LVH in the hyper-


tensive patient
Several factors contribute to myocyte hypertrophy and
fibroblast proliferation.
Mechanical factors. Concentric LVH generally reflects
a process of myocyte adaptation to elevated blood pressure
(Laplace’s law), which becomes inappropriate and con-
tributes to the deterioration in left ventricular function.
In experimental models, genetically modified mice unable
to develop LVH when ventricular wall stress is increased by
applying an aortic ring, exhibit less degree of heart failure
than control animals, which develop hypertrophy [13]. The
role of arterial compliance and of central blood pressure in
the genesis of hypertensive LVH has also been clearly shown
by some authors as an important aetiological factor of HF-
PSF and the main pathophysiological factor for episodes of
decompensation [14].
Hormonal factors. The renin-angiotensin-aldesterone
system plays a central role in the development of LVH
[15]. Many experimental studies have demonstrated the
antitrophic role of angiotensin converting enzyme inhibitors
(ACE) or angiotensin II receptor blockers (ARBs) [16—18].
Other hormonal factors are involved: catecholamines,
insulin and growth hormone [19]. The antihypertensive
effect of some drugs in the spontaneously hypertensive rat
is associated with a reduction in heart weight only if the
haemodynamic effect is combined with a reduction in cat-
echolergic activity (which was not true with direct acting
vasodilators) [20].

Results of clinical trials


Impact of antihypertensive agents on LVH
Figure 1. Diagnostic algorithm for heart failure with preserved Results of the first studies on regression of LVH in
systolic function proposed by the TACTIC group (from reference [1]). hypertension were difficult to interpret because of some
methodological limitations [21,22]. The results of recent
comparative trials with good methodological quality [23]
partly contradict those of meta-analyses, which mostly con-
Management of precipitating or causative cluded that ACE and ARBs were superior to other drugs
[24]. The Prospective Randomized Enalapril Study Evaluat-
factors ing Regression of Ventricular Enlargement (PRESERVE) trial
showed no difference between enalapril and nifedipine [25].
Role of hypertension and left ventricular
The Left Ventricular hypertrophy regression Indapamide
hypertrophy versus Enalapril (LIVE) trial [26], which used a rigorous
methodology, with a sufficiently large number of patients
Pathophysiological review (200 patients per group), showed indapamide to be more
Factors predisposing to LVH in the hypertensive effective than enalapril on reducing left ventricular mass at
patient one year, with a comparable antihypertensive effect. The
The prevalence of electrical LVH depends on several param- largest LVH regression trial is the Losartan Intervention For
eters: Endpoint reduction in hypertension (LIFE) trial [27], which
• blood pressure level, particularly systolic [4,5]; showed that, for a similar blood pressure control, losartan
• duration of hypertension and its control by antihyperten- (together with a diuretic in 90% of cases) produced a greater
sive treatment [6]; LVH regression than atenolol (combined with a diuretic in the
• age and obesity. same proportion of patients).
Consequences of LVH regression on prognosis
A sodium rich diet [7] and alcohol consumption [8] The incidence of cardiovascular complications appears to
also predispose to the development of LVH in hypertensive be lower in patients whose LVH had regressed on treatment
patients. LVH occurs more commonly in hypertensive men; than in those in whom LVH persists [28—31], but these tri-
however, the direct influence of sex disappears when the als included small numbers of patients and do not establish
severity of hypertension and lifestyle are taken into account whether there was a parallel reduction in left ventricu-
[9,10]. Finally, genetic and ethnic factors are also involved lar mass and in cardiovascular risk. The lower incidence of
[11,12]. stroke with losartan in the LIFE trial was associated with a
364 P. de Groote et al.

greater LVH regression [32,33], although this difference may coronary disease is associated with an increase in mortality
be largely explained by a lower decrease in central blood in patients with HF-PSF: findings from the coronary artery
pressure by atenolol, as clearly shown in the Conduit Artery surgery study (CASS) showed that in patients with HF-PSF,
Function Evaluation study (CAFE) trial [34]. the six-year mortality rate was 8% in the absence of coro-
nary artery disease, compared to 17% in the presence of a
Role of bilateral renal artery stenosis significant single or double vessel disease and 32% in case of
triple vessel disease [44].
Secondary hyperaldosteronism and volume overload, There are therefore arguments indicating that myocar-
related to a decrease of natriuresis, in case of significant dial ischaemia can contribute to diastolic dysfunction and
bilateral renal artery stenosis could explain the clinical may be the cause of heart failure in a patient with diastolic
presentation of HF-PSF and particularly the recurrent dysfunction.
‘‘flash’’ acute pulmonary oedema. Revascularization in
this situation can prevent the recurrence of heart failure Treatment of ischaemia and prevention of HF-PSF
decompensation [35,36]. This justifies to perform a renal
and episodes of heart failure
artery Doppler study, particularly if predisposing factors
for renal stenosis are present: severe systolic—diastolic Two questions must be answered to examine the relationship
hypertension resistant to antihypertensive treatments, between anti-ischaemic treatment and prognosis of HF-PSF:
age greater than 70 years old, multiple vascular lesions, • does anti-ischaemic treatment (pharmacological or revas-
renal insufficiency, significant increase in plasma creatinine cularization) in a patient without cardiac dysfunction
after ACE or ARBs introduction. At the opposite, the role reduces the risk of HF-PSF?
of a unilateral renal stenosis in HF-PSF decompensation is • does anti-ischaemic treatment (pharmacological or by
still discussed and angioplasty has not been shown to be revascularization) prevent heart failure decompensation
beneficial in this situation. in a patient with HF-PSF?
To the best of our knowledge, no clinical trials have
Role of myocardial ischaemia answered to these two questions. We just have substud-
ies of different clinical trials and/or meta-analyses without
Physiological relationship between ischaemia and any information about the type of heart failure assessed.
HF-PSF While the treatment of coronary risk factors, particularly
In order to examine the relationship between left ventric- the treatment of hypertension, reduces the risk of heart fail-
ular diastolic dysfunction and myocardial ischaemia, two ure, anti-ischaemic treatment does not appear to reduce the
questions need to be answered: incidence of heart failure [45,46]. In the meta-analysis of
• is repeated or chronic myocardial ischaemia a factor the Blood Pressure Lowering Treatment Trialists’ collabora-
which initiates or predisposes to diastolic dysfunction? tion (BPLTT) [46], beta-blockers were not superior to ACE in
• can an acute ischaemic episode be a factor that initiates preventing heart failure. However, calcium channel block-
or predisposes to heart failure in a patient with diastolic ers appeared to be less effective than ACE, but also less
dysfunction? effective than diuretics and beta-blockers in preventing the
development of heart failure.
In answer to the first question, we have to keep in mind In the Occluded Artery Trial (OAT), desocclusion of a coro-
that left ventricular relaxation requires the active trans- nary artery by angioplasty did not reduce the risk of heart
port of calcium into the sarcoplasmic reticulum, enabling failure [47]. Furthermore, angioplasty of an occluded artery
the dissociation of actin—myosin bridges. As a result of had no effect on prognosis, whether or not the patients had
hypoxia, which inhibits this dissociation process, ischaemia previous heart failure and whether or not they had LVEF less
can impaired active relaxation [37,38]. In addition, by than 50%.
predisposing to cell necrosis and the development of inter- A study published in 2000 examined the findings from
stitial fibrosis, ischaemia contributes to the deterioration in 46 patients hospitalized for rapidly resolving acute pul-
left ventricular structure and diastolic function. Ischaemia monary oedema (‘‘flash’’ oedema) [45]. Twenty-seven of
can therefore reduce passive left ventricular compliance. these patients (58%) had LVEF greater than 40%, 38 (82%)
Myocardial ischemia could also be related to microcircula- underwent coronary angiography and among these, 33 (86%)
tory abnormalities, as found in hypertension and in diabetes had a significant coronary artery disease. Nineteen of these
mellitus [39]. 33 patients underwent bypass surgery and eight a coronary
In order to answer the second question, many years ago, angioplasty. During the three-year of the follow-up period,
acute myocardial ischaemia was shown to be involved in pulmonary oedema recurred in half of the patients, and
the development of heart failure [40,41], and this is still in nine patients (47%) of the 19 who were revascularized.
a current issue [42,43]. In 1987, a study demonstrated in 32 Despite its major limitations (small number of patients and
patients that all the echocardiographic diastolic parameters lack of randomization), this study suggests that coronary
were impaired in the first 15 s after coronary artery occlusion revascularization has no preventive effect on the devel-
[41]. In 2004, in a small study including 18 patients hospital- opment of pulmonary oedema in patients with preserved
ized for heart failure with left ventricular ejection fraction systolic function and a past history of pulmonary oedema.
(LVEF) greater or equal to 50%, a significant coronary artery Finally, the ACC/AHA 2001 guidelines for the manage-
disease was found by coronary angiography in 39% of patients ment of HF-PSF state that ‘‘coronary revascularization is
and in 25% of cases, heart failure was associated with an recommended in patients with coronary artery disease in
episode of myocardial ischaemia [43]. Finally, the extend of whom symptomatic or demonstrable myocardial ischaemia
Treatment of heart failure with preserved systolic function 365

is judged to have an adverse effect on diastolic function’’ reduce the risk of dehydration. In HP-PSP, the adaptive possi-
although with a level of evidence of C [48]. This proposal bility of the heart is decreased, related to a steeper diastolic
is based on pathophysiological reasoning but not on clinical pressure—volume curve. Minor changes in volume are asso-
evidence. In 2005, the European and American guidelines ciated with large changes in pressure (a moderate increase
for the management of heart failure both stated that there in volume increases the risk of decompensation, and in con-
was a strong evidence to perform coronary revascularization trast, a small reduction in ventricular volume predisposes to
only to patients with angina and not to patients with asymp- hypotension).
tomatic coronary disease [49,50]. They did not recommend Heart rate plays an important role in the increase of
coronary revascularization in order to improve ventricular diastolic pressures. Alongside diuretics, nitrates and beta-
function or to reduce symptoms of heart failure [51]. blockers, calcium channel blockers slowing heart rate can
be used. Finally, theoretically, renin-angiotensin system
Role of supraventricular arrhythmias inhibitors are useful in the management of these patients:
they control blood pressure, reduce left ventricular mass
Supraventricular arrhythmias are considered to have an and may improve diastolic function [53].
adverse effect in HF-PSF. Tachycardia, which reduces the
duration of diastole, and the loss of atrial systole con- Epidemiological studies
siderably decrease cardiac output. LVH increases these
abnormalities. In experimental studies in dogs, sinus tachy- Several observational epidemiological studies have exam-
cardia in the presence of LVH rapidly increases left atrial ined the numbers of prescriptions of different classes of
pressure [52]. The theoretical treatment of choice to control drugs. In some of these studies, the authors examine the
heart rate is beta-blockers, which also reduce myocardial impact of various drugs on morbidity and mortality. These
ischaemia and blood pressure. Non-dihydropyridine calcium studies are not randomized trials and all of these analyses
blockers and/or digoxin, particularly in case of atrial fibrilla- were conducted retrospectively. Nevertheless, they provide
tion, are an alternative to beta-blockers. The ideal situation useful information on the therapeutic management of HF-
is to obtain a heart rate close to 70 per minute and in PSF.
any case lower than 90 per minute. Atrial ventricular node An important initial finding from these registries is the
ablation with pacemaker implantation is rarely required. relatively limited use of LVEF measurement in patients
Atrial fibrillation ablation may be useful in case of recurrent hospitalized for acute heart failure. In American studies
heart failure decompensations. Finally, the role of ivabra- [54—57], LVEF measurement was performed in approx-
dine in controlling heart rate in patients with sinus rhythm imately 50—60% of the patients, at the exception of
is unknown in HF-PSF. one study, the Organized Program to Initiate Lifesav-
ing Treatment in Hospitalized Patients With Heart Failure
Other triggering factors (OPTIMIZE-HF) trial [58] where LVEF was measured in 85% of
the patients. In Europe, LVEF is measured in almost 80% of
Heart failure decompensation may be promoted by extrac- patients [59,60].
ardiac factors such as non compliance to the low sodium
diet or to medical treatment, intake of non-recommended Prescription rates for the different classes of drugs
drugs (particularly non-steroidal anti-inflammatory drugs,
rich sodium drugs, etc.), pulmonary infection or any other Prescription rates of the different classes of drugs from
infection and anemia. These extracardiac factors are not some epidemiological studies are summarized in Table 1.
specific to HF-PSF but also affect heart failure with systolic Most of these studies involved patients admitted in hospitals
dysfunction. for heart failure decompensation. Shamagian et al. divided
their Spanish population into three groups depending on the
year of hospitalization [60]. The study, from the Cardiovas-
cular Health Study, examined the general population over
Data from registers and controlled trials 65 years old with ambulatory HF-PSF. In the Berry study,
on the pharmacological management of diuretic and ACE prescription rates should be interpreted
HF-PSF with caution as these drugs were one of the inclusion criteria
[59].
Specific features of HF-PSF In some studies, the authors compared the type of man-
agement according to the severity of LV dysfunction. In the
As described in the introduction, the treatment of HF-PSF is Management to Improve Survival in Congestive Heart Fail-
currently based on our pathophysiological knowledge. The ure (MISCHF) register, the 1291 patients were divided into
main goal of the treatment is to reduce the left ventricular three subgroups depending on LVEF function: less or equal
end-diastolic pressure. to 39%, 40 to 50%, and greater than 50% [55]. In the major-
By avoiding salt and water retention, diuretics are the ity of cases, ACE were being prescribed more commonly in
treatment of choice for this disease. However, as in heart patients with systolic dysfunction, both on admission and
failure with systolic dysfunction, no study has assessed their at discharge from hospital. The doses of ACE were also
benefit in HF-PSF. With nitrates, diuretics are effective in the higher in systolic dysfunction [55]. Similar results were found
management of acute episodes. In chronic stable patients, in the EuroHeart Failure survey [62,63]. The same finding
the minimal doses of diuretics will be prescribed, partic- also applies to beta-blockers in the most recent analyses
ularly in the elderly, in order to control symptoms and to [57,63].
366
Table 1 Prescription rates of different classes of drugs in epidemiological studies.
n Date LVEF (%) Diur (%) ACE (%) ARBs (%) BB (%) Spir (%) Ca-B (%) Digo (%)
CHS 170 1989—1993 ≥ 55 59 25 17 31 41
REP 91 59 1991 ≥ 50 78 31 19 24 27
Alabama 238 1994 ≥ 40 30 31 47
MISCHF 312 1995—1997 ≥ 50 59 29 23 39 30
76 45 19 43 36
Smith et al. 200 1996—1998 ≥ 40 79 34 19 50 30
Berry 130 2000 ≥ 40 95 65 38 12 24 23
Spanish 147 1991—1996 ≥ 50 64 42 0 12 10 18 39
Study 109 1997—1999 69 55 5 12 5 30 22
186 2000—2001 64 49 6 28 12 36 16
EuroHeart 3,148 2000—2001 ≥ 40 85 58 4 39 17 28 31
Failure
Dobre 443 2000—2005 ≥ 40 88 66 12 51 41 15 22
FOCALE 245 ? ≥ 45 67 53 16 34 16 12 19
REP 03 306 2003—2005 ≥ 50 24 17 21
71 51 63
ADHERE 26,322 2004 ≥ 40 65 36 13 46 5 19
80 47 13 52 11 21
LVEF: left ventricular ejection fraction; Diur: diuretics; ACE: angiotensin II converting enzyme inhibitors; ARBs: angiotensin II receptor blockers; BB: beta-blockers; Spir: spironolactone:
CaB: calcium channel blockers; Digo: digoxin.
For the trials by Smith et al., Berry et al., Dobre et al. and the Spanish trial, treatment was at discharge from hospital. The second line in the MISCHG, REP 03 and ADHERE studies is
treatment at discharge from hospital. The FOCALE trial randomized ambulatory patients over 65 years old with an ECG and echocardiogram at inclusion.
CHS, [61]; REP91, [54]; Alabama, [65]; MISCH, [55]; Smith et al., [82]; Berry et al., [59]; Spanish trial, [60]; EuroHeart Failure Survey, [63]; Dobre et al., [66]; FOCALE, [81]; REP 03, [64];
ADHERE, [57].

P. de Groote et al.
Treatment of heart failure with preserved systolic function 367

In other studies, prescription rates of different drugs The open study by Aronow et al. assessed the effect of
before hospitalization and at discharge have been exam- a beta-blocker, propanolol (90 mg/d), on prognosis in 158
ined [55,57,64]. Diuretics were the most widely prescribed patients, aged 81 ± 8 years old, in class II-III NYHA with
drugs in these studies, followed by the ACE, with an increase LVEF greater or equal to 40% and a history of myocardial
in their prescription rates during the hospitalization. How- infarction [67]. All the patients were receiving diuretics
ever, we do not know whether this approach was guided by and ACE for at least two months. Total mortality after 32-
blood pressure level. The prescribing rate of beta-blockers months of follow-up was reduced by 35% with propanolol
also appears to increase in the studies conducted after 2000 (44 deaths compared to 60, p = 0.007) and total mortality
whereas digoxin prescription decreased, from 40% in the and non-fatal myocardial infarction were reduced by 37%
1990s to 20% after 2000. Prescription rate of calcium chan- (p = 0.002).
nel blockers remains relatively stable with no information The Survival of Myocardial Infarction Long-Term Evalu-
available about the type of calcium blocker (vasodilators or ation study (SMILE-ISCHEMIA study) examined zofenopril in
bradycardiac agents). 303 patients admitted with myocardial infarction and LVEF
greater than 40% [68]. The primary objective of the study
Relationship between treatments and was to examine the effect of zofenopril on the extent of
morbidity—mortality rates myocardial ischemic burden (studied by a 24 h Holter moni-
toring and by an exercise test) after six months of treatment.
Because of some contradictory results of different retro- The study demonstrated a significant reduction of the pri-
spective studies from registries, it is difficult to draw any mary objective with zofenopril. There were only two deaths
conclusions. during the follow-up period and heart failure developed or
An analysis of a population from the American MEDICARE worsened in 12 patients: all except for one of these were on
system including 1091 patients over 65 years old, hospital- placebo.
ized in 1994 for heart failure did not show any relationship A retrospective analysis of the European Trial on Reduc-
between the prescription of ACE and four-year survival tion of Cardiac Events with perindopril in stable coronary
in the 238 patients with LVEF greater and equal to 40% artery disease study (EUROPA) was conducted in patients
[65]. with LVEF greater than 40% [69]. The EUROPA study was
In the MISCHF register, ACE prescription was associated a double-blind, randomized, placebo controlled trial of an
with a shorter length of hospitalization and a 60% reduc- ACE, perindopril (8 mg/d), in 12,218 patients with coronary
tion in the six months mortality rate in patients with LVEF artery disease but without clinical evidence of heart failure.
between 40 and 49%. Conversely, prescription of ACE in the LVEF was measured in 58% of the population, of whom 6,878
subgroup of patients with LVEF greater or equal to 50% was patients (56%) had LVEF greater than 40%. Perindopril sig-
associated with a 30% non-significant decrease in morbid- nificantly reduced the incidence of the primary end point
ity and mortality [55]. The power of this analysis is limited (cardiovascular mortality, non-fatal myocardial infarction
by the small number of patients per subgroups (less than and resuscitated sudden death) from 9.8 to 8.3% (RR = 0.84
200). [0.77—0.98]).
In the Spanish study, authors compared mortality of their
patients admitted for heart failure decompensation with
a separate analysis over three different inclusion periods Large-scale trials
[60]. Annual mortality was similar during the three periods, Four randomized, placebo-controlled, multicenter trials
between 9 to 12%. By multivariate analysis, the prescription have examined the effects of different pharmacological
of ACE or of an ARBs was associated with a greater survival treatments in HF-PSF: the DIG, CHARM-preserved, PEP-CHF
rate (36% mortality reduction). On the other hand, digoxin and SENIORS trials.
was associated with a 40% significant increase in mortality. The major therapeutic characteristics of these trials are
In the Dutch study [66], prescription of beta-blockers was summarized in Table 2. Diuretics were still the most com-
associated with a significant reduction in mortality, with an monly prescribed class of drugs, followed by beta- blockers
adjusted relative risk of 0.57 [0.37—0.88, p = 0.01], with a in the two most recent studies. ACE were widely pre-
dose effect. It is interesting to note that in this study, digoxin scribed in the DIG trial and digoxin prescription was variable,
prescription was also associated with an excess mortality between 12 and 40%, in the three other trials.
(adjusted RR = 1.58 [1.006—2.47, p = 0.05]).
The ACE and beta-blockers were also found to have a
Digitalis Intervention Group (DIG trial)
beneficial effect in the EuroHeart Failure Study [63]. After
The DIG trial can be considered to be the first random-
12-week of follow-up, prescription of ACE and of beta-
ized multicenter trial in HF-PSF. A subgroup of 988 North
blockers were associated with a 45% (RR = 0.55 [0.43—0.71])
American patients with a LVEF greater than 45% was ran-
and with a 39% (RR = 0.61 [0.48—0.77]) reduction in total
domized to receive digoxin (492 patients) or placebo (496
mortality, respectively.
patients). Digoxin had no impact on total or cardiovascular
mortality, mortality from heart failure or on the majority
Clinical trials of the combined end-points. Hospitalizations for heart fail-
ure were significantly reduced with digoxin but only after
Small studies and retrospective analyses two-year of follow-up and this difference was not significant
A few small studies (with less than 400 patients) have at the end of the trial. There was a trend towards more
assessed the effect of different classes of treatment in HF- hospitalization for unstable angina in patients on digoxin
PSF. [70].
368 P. de Groote et al.

Table 2 Prescription rates of different classes of drugs in the 4 randomized trials in HF-PSF.
n Date LVEF (%) Diur (%) ACE (%) ARBs (%) BB (%) Spir (%) CaB (%) Digo (%)
DIG 988 1991—1993 ≥ 45 76 86 8
CHARM 3,023 1999—2000 ≥ 40 75 19 56 12 32 28
SENIORS 2,135 2000—2002 86 83 7 28 11 40
PEP-CHF 850 2000—2003 ≥ 40 46 55 10 33 12
55a
LVEF: left ventricular ejection fraction; Diur: diuretics; ACE: angiotensin II converting enzyme inhibitors; ARBs: angiotensin II receptor
blockers; BB: beta-blockers; Spir: spironolactone; CaB: calcium channel blockers; Digo: digoxin.
a Including thiazides.

Candesartan in Heart Failure Assessment of Reduction Perindopril in Elderly People — Congestive Heart Fail-
in Mortality and Morbidity-Preserved study (CHARM- ure trial (PEP-CHF)
preserved study) The PEP-CHF trial examined the effect of an ACE, perindo-
This is to date the largest trial in terms of number of pril 4 mg/d, on morbidity and mortality in elderly patients
patients, which examine the effects of an ARBs, the can- followed up for HF-PSF [73]. The inclusion criteria were
desartan in HF-PSF [71]. CHARM-preserved trial included age greater or equal to 70 years, heart failure treated with
3023 patients older than 18 years, with NYHA class II-IV diuretics, with a LVEF greater than 40%. In addition, three
clinical heart failure who had been hospitalized at least out of nine clinical criteria for heart failure and two out
once for a cardiac cause and had an LVEF greater than 40%. of four echocardiographic criteria for diastolic dysfunction
Patients were randomized double-blind against placebo. The were required. The average age was 75 years and 55% of
primary objective was to examine the effect of candesartan the patients were women, 79% were hypertensive and 26%
on cardiovascular mortality and hospitalizations for heart had a past history of myocardial infarction. The primary
failure. After 36.6-months of follow-up, candesartan had not end-point was total mortality and hospitalizations for heart
reduced the incidence of the primary end point (22% versus failure at one year. The trial showed a non-significant impact
24% with a 11% non-significant reduction on candesartan). of perindopril on the primary end point. The annual event
Candesartan had no impact on the different secondary end rate was 12.2% in the perindopril subgroup and 13.2% in the
points (hospitalization for heart failure, myocardial infarc- placebo subgroup. At the end of the trial, event rates were
tion, stroke or coronary revascularization). The average 23.6 and 25.1%, respectively. Similar results were found
dose of candesartan at six months was 25 mg/d. There were with all the secondary end points: total or cardiovascular
significantly more adverse effects on candesartan than on mortality, hospitalizations for heart failure, duration of hos-
placebo: more hypotension (2.4% versus 1.1%), more hyper- pitalizations, increase in treatment.
kalaemia (1.5% versus 0.6%) and a greater rise in serum These results are difficult to interpret because of numer-
creatinine (4.8% versus 2.4%). ous problems. The inclusion period, as well as the follow-up
Despite this statistically negative result, the CHARM trial period, were extended in order to obtain a sufficient number
shows that a large scale randomized trial can be conducted of events. The drop-out rate was relatively high, 40% in the
in this complex disease. This study opened the way for new perindopril group and 36% in the placebo group. Open label
trials in HF-PSF. ACE was prescribed in 35% of the patients in the perindopril
arm and in 37% in the placebo arm.
Study of the Effect of Nebivolol Intervention on Out- In summary, these trials did not show any significant
comes and Rehospitalization in Seniors With Heart effect of different classes of drugs studies, digoxin, ACE or
Failure trial (SENIORS trial) ARBs, except for a beta-blocker (nebivolol) in a non-specific
The SENIORS trial cannot be considered to be a specific HF-PSF study.
trial of HF-PSF because HF-PSF patients only represented a Two other randomized trials are ongoing: one, the Irbe-
subgroup of the total population and because the LVEF cut- sartan in Heart Failure With Preserved Systolic Function trial
off value defining HF-PSF was unusually low (LVEF greater (I-PRESERVE) [74] is assessing irbesartan and the other, the
than 35%) [72]. The aim of the trial was to examine the trial of aldosterone antagonist therapy in adults with pre-
effect of a beta-blocker, nebivolol, in 2135 patients over served ejection fraction congestive heart failure (TOPCAT
70 years, who had either been hospitalized within the pre- trial) is assessing an aldosterone receptor blocker, spirono-
vious year for heart failure or in whom an LVEF of less or lactone. The results of these two trials will perhaps enable
equal to 35% had been found within the previous six months. us to improve the management of patients suffering from
The major end-point was total mortality and hospitaliza- HF-PSF.
tions for a cardiovascular cause. After a follow-up period
of 21 months, nebivolol produced a significant 14% reduc-
tion in the incidence of the primary end-point (31.1% versus Special features of elderly patients over 80
35.3%, p = 0.04). Median LVEF was 34%, although only 20% of years old
the population had an LVEF greater than 40%. The effect of
Nebivolol on the primary end-point was similar whether the The prevalence of HF-PSF increases with age. The diagno-
LVEF was over or under 35%. sis in octogenarians is often difficult because of atypical
Treatment of heart failure with preserved systolic function 369

symptoms [75]. In addition, some non-specific ‘‘geriatric’’ ciated with therapeutic education compared to therapeutic
signs (asthenia, confusion, behavioral disorders, fall, loss education alone showed significant improvement in walk-
of independence) can also be the only clinical feature of ing tolerance, quality of life and depression indices in the
heart failure, which is responsible for a delay in diagnosis intervention group [80].
and treatment. The clinical presentation is even more atyp- Therapeutic education and multidisciplinary manage-
ical and the prognosis even poorer if the heart failure occurs ment in patients with HF-PSF is probably more useful in
in frail patients with several comorbidities. Epidemiological controlling etiological and precipitating factors than in
data from a register of 86,094 institutionalized 85-year-old terms of preventing decompensation and reducing mortal-
patients with heart failure [76] indicate that one third of ity. Therefore, the influence of therapeutic education is
the patients had at least six comorbidities in addition to undoubtedly involved in the beginning of the disease. Deteri-
heart failure, with a higher prevalence of cognitive disor- oration in diastolic function often precedes the development
ders (57% of cases). In the Euro Heart Failure Survey I [77], of the initial symptoms by several years and early diagnosis
infection was present in 42% of patients over 80 years old, and treatment of these abnormalities are essential in order
confusion in 23%, anemia in 23% and severe renal failure to prevent irreversible damages. Primary prevention of HF-
(creatinine clearance less than 30 ml/min) in 15%. The pres- PSF requires smoking cessation and drastic control of blood
ence of these comorbidities implies geriatric management in pressure, hypercholesterolaemia and all other risk factors
order to identify markers of ‘‘fragility’’ in the elderly (cogni- for coronary artery disease.
tive problems, walking disorders, depression, malnutrition, As for systolic dysfunction, changes in lifestyle are essen-
social isolation, loss of independence etc.) [78]. tial and must be combined with an appropriate diet with
Cardiovascular disease is more serious in the elderly. In low sodium and alcohol intake and physical activity must
the Euro Heart Failure Survey I registry, total mortality at be maintained. Weight monitoring and control of conges-
12 weeks in patients with HF-PSF was significantly higher tive signs are important but often difficult as exacerbations
in octogenarians (average age 85 years old) compared to of left heart failure are usually sudden and more rarely
younger patients (average age 69 years old) (17% versus 7%, preceded by warning signs (reduced walking tolerance,
p < 0.0001) [77]. The excess mortality was due to both car- arrhythmia, increase of blood pressure, infection etc.).
diovascular and non-cardiovascular causes. Because of this, Knowledge of these warning signs however remains essential
it may be useful to consider the best end-points to be used in and patients must be educated about them.
this population in future clinical trials: mortality, cardiovas-
cular events, rehospitalisation, quality of life? In practice,
the therapeutic management for each patient must be based Conclusions
not on actual age but on an individual assessment taking
into account the ‘‘geriatric evaluation’’ of the elderly (life The therapeutic management of patients with HF-PSF is dif-
expectancy, quality of life, comorbidities, frailty, iatrogenic ficult. The lack of clear results from the few randomized
risk, etc.) [78]. trials in this disease does not help us in its management. This
management is based on our pathophysiology knowledge
Educational program of HF-PSF which is multifactorial, involving hypertension,
myocardial ischaemia, intrinsic myocardial damage and con-
Few studies have assessed the effect of multidisciplinary comitant diseases (diabetes, renal failure). In the absence
management in patients with HF-PSF. Interpretation of the of significant results of large therapeutic trials, treatment
results of theses studies are therefore difficult but the effect of HF-PSF is therefore based on patient education and con-
of multidisciplinary management appears to be less effec- trol of salt and water retention, blood pressure, heart rate
tive than in patients with systolic dysfunction. and episodes of myocardial ischaemia.
It is possible that studies demonstrating a benefit of
multidisciplinary management in terms of morbidity and
mortality have included patients with HP-FSP, although no
analysis by type of dysfunction was available until the pub-
References
lication by Galbreath et al. [79]. These authors randomized
[1] Juilliere Y, Trochu JN, de Groote P, et al. Heart failure with pre-
1069 patients with heart failure and compared multidis- served systolic function: a diagnostic algorithm for a pragmatic
ciplinary follow-up with telephone calls and education to definition. Arch Mal Coeur Vaiss 2006;99:279—86.
conventional follow up for 18 months. Seventy per cent of [2] Paulus WJ, Tschope C, Sanderson JE, et al. How to diagnose
patients had systolic dysfunction and 30% had HF-PSF. This diastolic heart failure: a consensus statement on the diagnosis
study demonstrated a significant reduction in mortality and of heart failure with normal left ventricular ejection frac-
cardiovascular events in the group of patients with systolic tion by the Heart Failure and Echocardiography Associations
dysfunction whereas no significant effect was found in the of the European Society of Cardiology. Eur Heart J 2007;28:
subgroup of patients with HF-PSF. Multidisciplinary manage- 2539—50.
ment in this study had little impact on blood pressure and [3] Kitzman DW, Higginbotham MB, Cobb FR, Sheikh KH, Sulli-
van MJ. Exercise intolerance in patients with heart failure
this subgroup of patients appeared to be at lower risk.
and preserved left ventricular systolic function: failure of the
In terms of effect on quality of life, this may be more Frank-Starling mechanism. J Am Coll Cardiol 1991;17:1065—72.
significant although again there are insufficient studies [4] Mosterd A, D’Agostino RB, Silbershatz H, et al. Trends in the
available. One study conducted in 32 women with NYHA prevalence of hypertension, antihypertensive therapy, and left
class II or III and LVEF greater than 45%, randomized to 12 ventricular hypertrophy from 1950 to 1989. N Engl J Med
weeks of mild to moderate regular physical exercise asso- 1999;340:1221—7.
370 P. de Groote et al.

[5] Kannel WB, Gordon T, Offutt D. Left ventricular hyper- [24] Klingbeil AU, Schneider M, Martus P, Messerli FH, Schmieder RE.
trophy by electrocardiogram. Prevalence, incidence, and A meta-analysis of the effects of treatment on left ventricular
mortality in the Framingham study. Ann Intern Med 1969;71: mass in essential hypertension. Am J Med 2003;115:41—6.
89—105. [25] Devereux RB, Palmieri V, Sharpe N, et al. Effects of once-daily
[6] Cuspidi C, Lonati L, Sampieri L, et al. Impact of blood pressure angiotensin-converting enzyme inhibition and calcium channel
control on prevalence of left ventricular hypertrophy in treated blockade-based antihypertensive treatment regimens on left
hypertensive patients. Cardiology 2000;93:149—54. ventricular hypertrophy and diastolic filling in hypertension:
[7] Schmieder RE, Messerli FH, Garavaglia GE, Nunez BD. Dietary the prospective randomized enalapril study evaluating regres-
salt intake. A determinant of cardiac involvement in essential sion of ventricular enlargement (PRESERVE) trial. Circulation
hypertension. Circulation 1988;78:951—6. 2001;104:1248—54.
[8] Manolio TA, Levy D, Garrison RJ, Castelli WP, Kannel WB. Rela- [26] Gosse P, Sheridan DJ, Zannad F, et al. Regression of left ven-
tion of alcohol intake to left ventricular mass: The Framingham tricular hypertrophy in hypertensive patients treated with
Study. J Am Coll Cardiol 1991;17:717—21. indapamide SR 1.5 mg versus enalapril 20 mg: the LIVE study. J
[9] Garavaglia GE, Messerli FH, Schmieder RE, Nunez BD, Oren S. Hypertens 2000;18:1465—75.
Sex differences in cardiac adaptation to essential hyperten- [27] Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular
sion. Eur Heart J 1989;10:1110—4. morbidity and mortality in the Losartan Intervention For End-
[10] Kupari M, Koskinen P, Virolainen J. Correlates of left ven- point reduction in hypertension study (LIFE): a randomised trial
tricular mass in a population sample aged 36 to 37 years. against atenolol. Lancet 2002;359:995—1003.
Focus on lifestyle and salt intake. Circulation 1994;89: [28] Muiesan ML, Salvetti M, Rizzoni D, Castellano M, Donato F,
1041—50. Agabiti-Rosei E. Association of change in left ventricular mass
[11] Ravogli A, Trazzi S, Villani A, et al. Early 24-hour blood pressure with prognosis during long-term antihypertensive treatment. J
elevation in normotensive subjects with parental hypertension. Hypertens 1995;13:1091—5.
Hypertension 1990;16:491—7. [29] Yurenev AP, Dyakonova HG, Novikov ID, et al. Management of
[12] Koren MJ, Mensah GA, Blake J, Laragh JH, Devereux RB. Com- essential hypertension in patients with different degrees of
parison of left ventricular mass and geometry in black and left ventricular hypertrophy. Multicenter trial. Am J Hypertens
white patients with essential hypertension. Am J Hypertens 1992;5:182S—9S.
1993;6:815—23. [30] Verdecchia P, Schillaci G, Borgioni C, et al. Prognostic signif-
[13] Esposito G, Rapacciuolo A, Naga Prasad SV, et al. Genetic icance of serial changes in left ventricular mass in essential
alterations that inhibit in vivo pressure-overload hypertrophy hypertension. Circulation 1998;97:48—54.
prevent cardiac dysfunction despite increased wall stress. Cir- [31] Cipriano C, Gosse P, Bemurat L, et al. Prognostic value of
culation 2002;105:85—92. left ventricular mass and its evolution during treatment in
[14] Haider AW, Larson MG, Franklin SS, Levy D. Systolic blood pres- the Bordeaux cohort of hypertensive patients. Am J Hypertens
sure, diastolic blood pressure, and pulse pressure as predictors 2001;14:524—9.
of risk for congestive heart failure in the Framingham Heart [32] Devereux RB, Dahlof B, Gerdts E, et al. Regression of hyper-
Study. Ann Intern Med 2003;138:10—6. tensive left ventricular hypertrophy by losartan compared with
[15] Dahlof B. Factors involved in the pathogenesis of hyperten- atenolol: the Losartan Intervention for Endpoint Reduction in
sive cardiovascular hypertrophy A review. Drugs 1988;35(Suppl Hypertension (LIFE) trial. Circulation 2004;110:1456—62.
5):6—26. [33] Devereux RB, Wachtell K, Gerdts E, et al. Prognostic signif-
[16] Kaneko K, Susic D, Nunez E, Frohlich ED. ACE inhibition reduces icance of left ventricular mass change during treatment of
left ventricular mass independent of pressure without affecting hypertension. JAMA 2004;292:2350—6.
coronary flow and flow reserve in spontaneously hypertensive [34] Williams B, Lacy PS, Thom SM, et al. Differential impact of
rats. Am J Med Sci 1997;314:21—7. blood pressure-lowering drugs on central aortic pressure and
[17] Nagano M, Higaki J, Mikami H, et al. Converting enzyme clinical outcomes: principal results of the Conduit Artery Func-
inhibitors regressed cardiac hypertrophy and reduced tissue tion Evaluation (CAFE) study. Circulation 2006;113:1213—25.
angiotensin II in spontaneously hypertensive rats. J Hypertens [35] Pickering TG, Herman L, Devereux RB, et al. Recurrent pul-
1991;9:595—9. monary oedema in hypertension due to bilateral renal artery
[18] Sabri A, Samuel JL, Marotte F, Poitevin P, Rappaport L, Levy BI. stenosis: treatment by angioplasty or surgical revascularisa-
Microvasculature in angiotensin II-dependent cardiac hypertro- tion. Lancet 1988;2:551—2.
phy in the rat. Hypertension 1998;32:371—5. [36] Bloch MJ, Trost DW, Pickering TG, Sos TA, August P. Prevention
[19] Weber KT, Sun Y, Guarda E. Structural remodeling in hyper- of recurrent pulmonary edema in patients with bilateral ren-
tensive heart disease and the role of hormones. Hypertension ovascular disease through renal artery stent placement. Am J
1994;23:869—77. Hypertens 1999;12:1—7.
[20] Sen S, Tarazi RC, Bumpus FM. Cardiac hypertrophy and antihy- [37] Nayler WG, Poole-Wilson PA, Williams A. Hypoxia and calcium.
pertensive therapy. Cardiovasc Res 1977;11:427—33. J Mol Cell Cardiol 1979;11:683—706.
[21] Gottdiener JS, Reda DJ, Massie BM, Materson BJ, Williams DW, [38] Mahmarian JJ, Pratt CM. Silent myocardial ischemia in patients
Anderson RJ, The Department of Veterans Affairs Cooperative with coronary artery disease. Possible links with diastolic left
Study Group on Antihypertensive Agents. Effect of single-drug ventricular dysfunction. Circulation 1990;81:III33—40.
therapy on reduction of left ventricular mass in mild to moder- [39] Galderisi M, Cicala S, Caso P, et al. Coronary flow reserve and
ate hypertension: comparison of six antihypertensive agents. myocardial diastolic dysfunction in arterial hypertension. Am
Circulation 1997;95:2007—14. J Cardiol 2002;90:860—4.
[22] Liebson PR, Grandits GA, Dianzumba S, et al. Comparison of [40] Clark LT, Garfein OB, Dwyer Jr EM. Acute pulmonary edema
five antihypertensive monotherapies and placebo for change in due to ischemic heart disease without accompanying myocar-
left ventricular mass in patients receiving nutritional-hygienic dial infarction. Natural history and clinical profile. Am J Med
therapy in the Treatment of Mild Hypertension Study (TOMHS). 1983;75:332—6.
Circulation 1995;91:698—706. [41] Labovitz AJ, Lewen MK, Kern M, Vandormael M, Deligonal U,
[23] Gosse P, Dubourg O, Gueret P. Regression of left ventricular Kennedy HL. Evaluation of left ventricular systolic and diastolic
hypertrophy with echocardiography: some lessons from the dysfunction during transient myocardial ischemia produced by
LIVE study. J Hypertens 2003;21:217—21. angioplasty. J Am Coll Cardiol 1987;10:748—55.
Treatment of heart failure with preserved systolic function 371

[42] Bogaty P, Mure P, Dumesnil JG. New insights into diastolic [57] Yancy CW, Lopatin M, Stevenson LW, De Marco T, Fonarow GC.
dysfunction as the cause of acute left-sided heart failure Clinical presentation, management, and in-hospital outcomes
associated with systemic hypertension and/or coronary artery of patients admitted with acute decompensated heart fail-
disease. Am J Cardiol 2002;89:341—5. ure with preserved systolic function: a report from the Acute
[43] Arques S, Ambrosi P, Gelisse R, Roux E, Lambert M, Habib G. Decompensated Heart Failure National Registry (ADHERE)
Prevalence of angiographic coronary artery disease in patients Database. J Am Coll Cardiol 2006;47:76—84.
hospitalized for acute diastolic heart failure without clinical [58] Gheorghiade M, Abraham WT, Albert NM, et al. Systolic
and electrocardiographic evidence of myocardial ischemia on blood pressure at admission, clinical characteristics, and out-
admission. Am J Cardiol 2004;94:133—5. comes in patients hospitalized with acute heart failure. JAMA
[44] Judge KW, Pawitan Y, Caldwell J, Gersh BJ, Kennedy JW. Con- 2006;296:2217—26.
gestive heart failure symptoms in patients with preserved left [59] Berry C, Hogg K, Norrie J, Stevenson K, Brett M, McMurray J.
ventricular systolic function: analysis of the CASS registry. J Am Heart failure with preserved left ventricular systolic function:
Coll Cardiol 1991;18:377—82. a hospital cohort study. Heart 2005;91:907—13.
[45] Kramer K, Kirkman P, Kitzman D, Little WC. Flash pul- [60] Grigorian Shamagian L, Gonzalez-Juanatey JR, Roman AV,
monary edema: association with hypertension and reoc- Acuna JM, Lamela AV. The death rate among hospitalized heart
currence despite coronary revascularization. Am Heart J failure patients with normal and depressed left ventricular
2000;140:451—5. ejection fraction in the year following discharge: evolution
[46] Turnbull F, Blood Pressure Lowering Treatment Trialists’ Col- over a 10-year period. Eur Heart J 2005;26:2251—8.
laboration. Effects of different blood-pressure-lowering [61] Gottdiener JS, McClelland RL, Marshall R, et al. Outcome of
regimens on major cardiovascular events: results of congestive heart failure in elderly persons: influence of left
prospectively-designed overviews of randomised trials. ventricular systolic function. The Cardiovascular Health Study.
Lancet 2003;362:1527—35. Ann Intern Med 2002;137:631—9.
[47] Hochman JS, Lamas GA, Buller CE, et al. Coronary intervention [62] Komajda M, Follath F, Swedberg K, et al. The EuroHeart Fail-
for persistent occlusion after myocardial infarction. N Engl J ure Survey programme–a survey on the quality of care among
Med 2006;355:2395—407. patients with heart failure in Europe. Part 2: treatment. Eur
[48] Hunt SA, Baker DW, Chin MH, et al. ACC/AHA guidelines for Heart J 2003;24:464—74.
the evaluation and management of chronic heart failure in the [63] Lenzen MJ, Scholte op Reimer WJ, Boersma E, et al. Differences
adult: executive summary. A report of the American College of between patients with a preserved and a depressed left ven-
Cardiology/American Heart Association Task Force on Practice tricular function: a report from the EuroHeart Failure Survey.
Guidelines (Committee to revise the 1995 Guidelines for the Eur Heart J 2004;25:1214—20.
Evaluation and Management of Heart Failure). J Am Coll Cardiol [64] Bursi F, Weston SA, Redfield MM, et al. Systolic and diastolic
2001;38:2101—13. heart failure in the community. JAMA 2006;296:2209—16.
[49] Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005 Guide- [65] Ahmed A, Roseman JM, Duxbury AS, Allman RM, DeLong JF.
line Update for the Diagnosis and Management of Chronic Correlates and outcomes of preserved left ventricular systolic
Heart Failure in the Adult: a report of the American Col- function among older adults hospitalized with heart failure.
lege of Cardiology/American Heart Association Task Force on Am Heart J 2002;144:365—72.
Practice Guidelines (Writing Committee to Update the 2001 [66] Dobre D, van Veldhuisen DJ, DeJongste MJ, et al. Prescription of
Guidelines for the Evaluation and Management of Heart Fail- beta-blockers in patients with advanced heart failure and pre-
ure): developed in collaboration with the American College of served left ventricular ejection fraction. Clinical implications
Chest Physicians and the International Society for Heart and and survival. Eur J Heart Fail 2007;9:280—6.
Lung Transplantation: endorsed by the Heart Rhythm Society. [67] Aronow WS, Ahn C, Kronzon I. Effect of propranolol versus
Circulation 2005;112:e154—235. no propranolol on total mortality plus nonfatal myocardial
[50] Swedberg K, Cleland J, Dargie H, et al. Guidelines for the infarction in older patients with prior myocardial infarc-
diagnosis and treatment of chronic heart failure: executive tion, congestive heart failure, and left ventricular ejection
summary (update 2005): The Task Force for the Diagnosis and fraction > or = 40% treated with diuretics plus angiotensin-
Treatment of Chronic Heart Failure of the European Society of converting enzyme inhibitors. Am J Cardiol 1997;80:
Cardiology. Eur Heart J 2005;26:1115—40. 207—9.
[51] McMurray J, Swedberg K. Treatment of chronic heart fail- [68] Borghi C, Ambrosioni E. Effects of zofenopril on myocardial
ure: a comparison between the major guidelines. Eur Heart ischemia in post-myocardial infarction patients with preserved
J 2006;27:1773—7. left ventricular function: the Survival of Myocardial Infarc-
[52] Berger RD, Wolff MR, Anderson JH, Kass DA. Role of atrial con- tion Long-term Evaluation (SMILE)-ISCHEMIA study. Am Heart
traction in diastolic pressure elevation induced by rapid pacing J 2007;153(445):e7—14.
of hypertrophied canine ventricle. Circ Res 1995;77:163—73. [69] Bertrand ME, Remme WJ, Fox KM, Ferrari R, Simoons ML.
[53] Friedrich SP, Lorell BH, Rousseau MF, et al. Intracardiac Effects of perindopril on long-term clinical outcome of patients
angiotensin-converting enzyme inhibition improves diastolic with coronary artery disease and preserved left ventricular
function in patients with left ventricular hypertrophy due to function. Int J Cardiol 2007;121:57—61.
aortic stenosis. Circulation 1994;90:2761—71. [70] Ahmed A, Rich MW, Fleg JL, et al. Effects of digoxin
[54] Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive on morbidity and mortality in diastolic heart failure:
heart failure in the community: a study of all incident cases the ancillary digitalis investigation group trial. Circulation
in Olmsted County, Minnesota, in 1991. Circulation 1998;98: 2006;114:397—403.
2282—9. [71] Yusuf S, Pfeffer MA, Swedberg K, et al. Effects of can-
[55] Philbin EF, Rocco Jr TA, Lindenmuth NW, Ulrich K, Jenkins PL. desartan in patients with chronic heart failure and preserved
Systolic versus diastolic heart failure in community practice: left-ventricular ejection fraction: the CHARM-Preserved Trial.
clinical features, outcomes, and the use of angiotensin- Lancet 2003;362:777—81.
converting enzyme inhibitors. Am J Med 2000;109:605—13. [72] Flather MD, Shibata MC, Coats AJ, et al. Randomized trial to
[56] Masoudi FA, Havranek EP, Smith G, et al. Gender, age, and heart determine the effect of nebivolol on mortality and cardiovas-
failure with preserved left ventricular systolic function. J Am cular hospital admission in elderly patients with heart failure
Coll Cardiol 2003;41:217—23. (SENIORS). Eur Heart J 2005;26:215—25.
372 P. de Groote et al.

[73] Cleland JG, Tendera M, Adamus J, Freemantle N, Polon- [78] Komajda M, Forette F, Aupetit JF, et al. Recommendations for
ski L, Taylor J. The perindopril in elderly people with the diagnosis and management of cardic failure in the elderly
chronic heart failure (PEP-CHF) study. Eur Heart J 2006;27: subject. Arch Mal Coeur Vaiss 2004;97:803—22.
2338—45. [79] Galbreath AD, Krasuski RA, Smith B, et al. Long-term health-
[74] Carson P, Massie BM, McKelvie R, et al. The irbe- care and cost outcomes of disease management in a large,
sartan in heart failure with preserved systolic function randomized, community-based population with heart failure.
(I-PRESERVE) trial: rationale and design. J Card Fail 2005;11: Circulation 2004;110:3518—26.
576—85. [80] Gary RA, Sueta CA, Dougherty M, et al. Home-based exercise
[75] Tresch DD. The clinical diagnosis of heart failure in older improves functional performance and quality of life in women
patients. J Am Geriatr Soc 1997;45:1128—33. with diastolic heart failure. Heart Lung 2004;33:210—8.
[76] The SAGE Study GroupGambassi G, Forman DE, Lapane KL, et [81] Dubourg O, Gueret P, Beauchet A, Nisse-Durgeat S, Ducardon-
al. Management of heart failure among very old persons living net A. Focale: Study of systolic and diastolic heart failure in a
in long-term care: has the voice of trials spread? Am Heart J French elderly population. Int J Cardiol 2007;124:188—92.
2000;139:85—93. [82] Smith GL, Masoudi FA, Vaccarino V, Radford MJ, Krumholz HM.
[77] Komajda M, Hanon O, Hochadel M, et al. Management of octo- Outcomes in heart failure patients with preserved ejection
genarians hospitalized for heart failure in Euro Heart Failure fraction: mortality, readmission, and functional decline. J Am
Survey I. Eur Heart J 2006;28:1310—8. Coll Cardiol 2003;41:1510—8.