guidelines for the diagnosis and treatment of chronic heart failure

8/9/2019 Guidelines for the Diagnosis and Treatment of Chronic Heart Failure

1/34


2/34

To order bulk copies of this article, please contact - Jenni WelchTel: +44 (0) 20 7424 4389, Fax: +44 (0) 20 7424 4433 Email: [email protected]

200

1Th

eEur

opeanS

ociety

ofCardi

ology

without symptoms of heart failure and who constituteapproximately a similar prevalence[8,9]. The prognosis ofheart failure is uniformly poor if the underlying problem

cannot be rectified. Half of patients carrying a diagnosisof heart failure will die within 4 years and in patientswith severe heart failure more than 50% will die within 1year[5,7]. Recent studies have confirmed the poor long-term prognosis[10,11] in patients with asymptomatic myo-cardial dysfunction[12]. No temporal improvement overtime has been found in community reports from theFramingham study[6] or Rochester project[11]. In con-trast, a Scottish report provides survival rates afterhospital discharge from 1986 to 1995 suggestingimproved prognosis over time[13].

Recent studies show that the accuracy of diagnosis byclinical means alone is often inadequate[14,15] particu-

larly in women, the elderly and the obese. In order tostudy properly the epidemiology and prognosis and tooptimize the treatment of heart failure the uncertaintyrelating to the diagnosis must be minimized or avoided.

Descriptive terms in heart failure

Acute vs chronic heart failure

Chronic heart failure, often punctuated by acuteexacerbations, is the most common form of heart failure.A definition of chronic heart failure is given below.

The term acute heart failure is often used, exclusively,to mean acute (cardiogenic) dyspnoea characterized by

signs of pulmonary congestion including pulmonaryoedema. However, acute heart failure could also applyto cardiogenic shock, a syndrome characterized by a lowarterial pressure, oliguria and a cool periphery, thatneeds to be distinguished from pulmonary oedema. It isadvisable not to use the term acute heart failure but themore precise terms acute pulmonary oedema or, whereapplicable, cardiogenic shock.

Systolic vs diastolic heart failure

As ischaemic heart disease is the commonest cause ofheart failure in industrialized societies most heart failureis associated with evidence of left ventricular systolicdysfunction, although diastolic impairment at rest is acommon if not universal accompaniment. Diastolicheart failure is often presumed to be present whensymptoms and signs of heart failure occur in the pres-ence of a preserved left ventricular systolic function(normal ejection fraction/normal end-diastolic volume)at rest. Predominant diastolic dysfunction is relativelyuncommon in younger patients, but increases in impor-tance in the elderly, in whom systolic hypertension,myocardial hypertrophy are contributors to cardiac dys-function. Most patients with heart failure and impair-ment of diastolic function also have impaired systolicfunction.

Other descriptive terms in heart failureRight and left heart failure refer to syndromes present-ing predominantly with congestion of the systemic or

pulmonary veins, respectively. The terms do not neces-sarily indicate which ventricle is most severly damaged.High and low-output, forward and backward, overt,

treated, congestive and undulating are other descriptiveterms still in occasional use; the clinical utility of theseterms have yet to be determined.

Mild, moderate or severe heart failure is used as aclinical symptomatic description where mild is used forpatients who can move around with no importantlimitations, severe for patients who are markedlysymptomatic and need frequent medical attention andmoderate for the remaining patient cohort.

Clinical syndromes are caused by an abnormalityof the heart and recognized by a characteristic pattern ofcardiac and extra-cardiac responses, including those ofhaemodynamic, renal, neural and hormonal nature.

Definition of chronic heart failureMany definitions of chronic heart failure exist[1619] buthighlight only selective features of this complex syn-drome. None is entirely satisfactory and one commonlyused definition is: heart failure is a pathophysiologicalstate in which an abnormality of cardiac function isresponsible for the failure of the heart to pump bloodat a rate commensurate with the requirements of themetabolizing tissues.

A simple objective definition of chronic heart failure iscurrently impossible as there is no cut-off valve orcardiac or ventricular dysfunction or change in flow,pressure, dimension or volume that can be used reliablyto identify patients with heart failure. The diag-nosis of heart failure relies on clinical judgement basedon a history, physical examination and appropriateinvestigations.

The Task Force considers the essential components ofheart failure to be a syndrome where the patients shouldhave the following features; symptoms of heart failure,typically breathlessness or fatigue, either at rest orduring exertion, or ankle swelling and objective evidenceof cardiac dysfunction at rest (Table 1). A clinicalresponse to treatment directed at heart failure alone isnot sufficient for diagnosis, although the patient shouldgenerally demonstrate some improvement in symptomsand/or signs in response to those treatments where a

relatively fast symptomatic improvement could beanticipated e.g. diuretic or nitrate administration. Itshould also be recognized that treatment may obscure

Table 1 Definition of heart failure. Criteria 1 and 2

should be fulfilled in all cases

1. Symptoms of heart failure(at rest or during exercise)

and2. Objective evidence of cardiac dysfunction

(at rest)and

(in cases where the diagnosis is in doubt)3. Response to treatment directed towards heart failure

1528 Task Force Report

Eur Heart J, Vol. 22, issue 17, September 2001


3/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

a diagnosis of heart failure by relieving the patientssymptoms. Therapy should not usually be initiated untila diagnosis of chronic heart failure has been establishedwith reasonable certainty.

The distinctions between cardiac dysfunction, persist-ent heart failure, heart failure that has been renderedasymptomatic by therapy and transient heart failure areoutlined in Fig. 1. It is important to note that exercise-induced ventricular dysfunction, usually due to myo-cardial ischaemia, may cause a rise in ventricularfilling pressure and a fall in cardiac output and inducesymptoms of heart failure such as breathlessness. How-ever, as both the underlying pathophysiology and thetreatment of this condition is generally different fromthat of heart failure secondary to chronic ventriculardysfunction, such patients should not be diagnosed ashaving chronic heart failure.

Aetiology of heart failure in Europe

Heart failure should never be the final diagnosis.

The aetiology of heart failure and the presence ofexacerbating factors or other diseases that may have animportant influence on management should be carefullyconsidered in all cases. The extent to which the cause ofheart failure should be pursued by further investigationwill depend on the resources available and the likelihoodthat diagnosis will influence management.

Chronic heart failure may be due to myocardialdysfunction, arrhythmias, valve abnormalities, peri-cardial disease or induced by rhythm disturbances.Anaemia, renal or thyroid dysfunction and cardio-depressant drugs may exacerbate, or more rarely cause,heart failure. Acute pulmonary oedema and cardiogenicshock have an aetiological spectrum similar to chronic

heart failure, though pulmonary oedema is rarely due topericardial disease. Standard cardiology textbooksshould be consulted for a more extensive list of the

causes of heart failure. In Europe, myocardial dys-function secondary to coronary artery disease, usually asa consequence of myocardial infarction, is the mostcommon cause of heart failure among patients under75 years of age[20] and clear abnormalities in systolicfunction are usually present. Among elderly patients,who are often less intensively investigated, an accuratediagnosis of the presence and aetiology of heart failure ismore difficult and obscured by multiple other diagnoses.Systolic hypertension and cardiac hypertrophy, cell lossand fibrosis may be more important causes of heartfailure in the elderly and may be more likely to manifestpredominantly as abnormalities of diastolic function.The aetiology of heart failure will also depend on ethnicorigin, socioeconomic status and geographic location.Several background factors can also induce heart failuree.g. hypertension, coronary artery disease and valvularabnormalities.

Importance of identifying potentially reversibleexacerbating factors

Symptoms of chronic heart failure, pulmonary oedema

and shock may be caused by tachy- and bradyarrhyth-mias or myocardial ischaemia even in patients withoutmajor, permanent cardiac dysfunction. Myocardial is-chaemia, changes in valvular regurgitation, pulmonaryembolism, infection, arrhythmia or renal dysfunction,side effects of drug therapy and excessive fluid, sodiumor alchohol intake may all cause or exacerbate symp-toms and/or signs of heart failure in patients withpre-existing cardiac dysfunction. It is important toidentify any reversible factors in order to treat heartfailure optimally.

Importance of the holistic approach to the diagnosis of

heart failureA proper diagnostic formulation must extend be-yond the cardiac problem, particularly in the elderly

CARDIAC DYSFUNCTIONCORRECTED OR RESOLVED

SYMPTOMS

THERAPY

HEARTFAILURE

Therapy CANNOTbe withdrawn

without recurrenceof symptoms

Symptomsrelieved

Symptomspersist

Systolicdysfunction

CARDIACDYSFUNCTION

TransientHeart

Failure

Nosymptoms

Asymptomaticcardiac

dysfunction

Therapy CANbe withdrawn

without recurrenceof symptoms

NORMAL

Figure 1 Relationship between cardiac dysfunction, heart failure and heart failurerendered asymptomatic.

Task Force Report 1529



4/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

population in whom multiple rather than single diseasesare common. Disease of the peripheral vasculature[21]

and other organs including the kidney and lungs mayhave an important influence on diagnosis and the choiceof treatment. For instance, in patients with prostatichypertrophy a vigorous diuresis may precipitate acuteurinary retention.

Aspects of the pathophysiology of thesymptoms of heart failure relevant todiagnosis

The origin of the symptoms of heart failure are not fullyunderstood. Increased pulmonary capillary pressure isundoubtedly responsible for pulmonary oedema in part,but studies conducted during exercise in patients withchronic heart failure demonstrate no simple relationshipbetween capillary pressure and exercise perform-ance[22,23]. This suggests either that raised pulmonarycapillary pressure is not the only factor responsible forexertional breathlessness or that current techniques tomeasure true pulmonary capillary pressure may not beadequate. In this context variation in the degreeof mitral regurgitation will influence breathlessness.Abnormalities of pulmonary diffusion, peripheral orrespiratory skeletal muscle[24], general cardiovasculardeconditioning or overweight[25,26] may also contributeimportantly to the sensation of breathlessness. Fatigue isanother essential symptom in heart failure. The originsof fatigue are even more obscure and compounded bydifficulties in quantifying this symptom[27]. Peripheraloedema is poorly related to right heart pressures;capillary permeability for fluid and small proteins andreduced physical activity being important additionalfactors. Extracardiac causes of oedema not related to

heart failure are common.Although impairment of cardiac function is central to

the development of heart failure, altered peripheral

blood flow, especially to the kidney and skeletal muscle,is typical and probably of major pathophysiologicalimportance[28]. Similarly, activation of a number ofneuroendocrine systems is characteristic of heart fail-ure[29,30]. Baroreceptor dysfunction is an important linkbetween vasomotor and neuroendocrine dysfunction[31].Understanding chronic heart failure has moved from ahaemodynamic concept into accepting the importance ofneuroendocrine pathophysiological changes in the pro-gression as well as for the treatment of heart failure[32].Activation of various cytokines may also contributeto cardiac dysfunction and to the clinical syndrome,particularly in more advanced stages[33].

Possible methods for the diagnosis of heartfailure in clinical practice

Symptoms and signs in the diagnosis of heart failure Symptoms and signs are important as they alert the

observer to the possibility that heart failure exists.The clinical suspicion of heart failure must be con-

firmed by more objective tests, particularly aimed atassessing cardiac function (Fig. 2).

Breathlessness, ankle swelling and fatigue are the char-acteristic symptoms of heart failure, but may be difficultto interpret, particularly among elderly patients, theobese and in women. Inter-observer agreement on thepresence or absence of symptoms of heart failure may below[34] notably in the days following a myocardialinfarction. There is no standard questionnaire availablefor the diagnosis of heart failure. In the context ofclinical or epidemiological studies, several scoring sys-tems are available that await proper validation andcannot be recommended for clinical practice at

present[35].Peripheral oedema, a raised venous pressure and

hepatomegaly are the characteristic signs of congestion

Suspected Heart Failurebecause of symptoms and signs

Assess presence of cardiac disease by ECG, X-Ray orNatriuretic peptides (where available)

Imaging by Echocardiography(Nuclear angiography or

MRI where available)

Tests abnormal

Tests abnormal

Assess etiology, degree, precipitatingfactors and type of cardiac dysfunction

Choose therapy

Additional diagnostic testswhere appropriate

(e.g. coronary angiography)

NormalHeart Failure

unlikely

NormalHeart Failureunlikely

Figure 2 Algorithm for the diagnosis of heart failure.




5/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

of systemic veins[36,37]. Clinical signs of heart failureshould be assessed in a careful clinical examinationincluding observing, palpating and auscultating the

patient. Unfortunately, clinical examination is oftenreplaced by laboratory investigations which reduce theexperience in bedside medicine among physicians.Peripheral oedema and hepatomegaly have low positivepredictive value, while determination of the jugularvenous pressure may be difficult. Peripheral oedema isusually absent in well-treated heart failure and primarilyleft ventricular dysfunction, even if severe[37]. Althoughcardiologists attain high agreement on the presence ofelevated jugular venous pressure under study conditions,the reproducibility is much lower among non-specialists[36]. Moreover, many patients, even with welldocumented heart failure, even if severe, do not have an

elevated jugular venous pressure

[37]

. Tachycardia is non-specific and may be absent even in severe heart failure,particularly in the presence of beta-blocker therapy[37].Other signs of heart failure require considerable exper-tise for their detection. A third heart sound is usuallyconsidered to be present in patients with severe heartfailure[37], but is not specific to heart failure[38]. Althoughcardiology specialists may attain high agreement for thepresence of a third heart sound under study condi-tions[36] the inter-observer agreement is less than 50%among non-specialists[39] and probably even lower inclinical practice. Pulmonary crepitations also have lowpositive predictive value and inter-observer differencesin eliciting this sign are high[40]. When cardiac murmurs

are present their origin and role in the symptomatologyshould be identified.

When multiple signs of heart failure are present,including a displaced apex beat, pitting oedema, a raisedvenous pressure and when a third heart sound is heardconfidently then, in the presence of appropriate symp-toms, a clinical diagnosis of heart failure may be madewith some confidence. Although a clinical diagnosisreached in this way may be specific it will fail to identifymany patients who might benefit from treatment. Thesubjective component of the examination and the inabil-ity to make a permanent direct record are furtherweaknesses of a diagnosis made on clinical featuresalone.

Symptoms and the severity of heart failure There is a poor relationship between symptoms and

the severity of cardiac dysfunction[15,35] and betweensymptoms and prognosis[41].

Once a diagnosis of heart failure has been establishedsymptoms may be used to classify the severity of heartfailure and should be used to monitor the effects oftherapy. The New York Heart Association classification(NYHA) is in widespread use (Table 2). The use ofexamples such as walking distance or number of stairsclimbed is recommended. In other situations, the classi-

fication of symptoms into mild, moderate or severe isused. Patients in NYHA class I would have to haveobjective evidence of cardiac dysfunction, have a past

history of heart failure symptoms and be receiving

treatment for heart failure in order to fulfil the basicdefinition of heart failure.In acute myocardial infarction, the classification

described by Killip has been used[42]. The value ofquestionnaires for the measurement of quality of life inthe context of classification of severity is still beingdebated. The most frequently used questionnaire is theMinnesota Living With Heart Failure[43]. It is importantto realize the common dissociation between symptomsand myocardial dysfunction. The severity of symptomsare highly dependent on the efficacy of therapy, patientexpectation and medical interpretation. Mild symptomsshould not be equated with minor cardiac dysfunction.

Electrocardiogram A normal ECG suggests that the diagnosis of chronic

heart failure should be carefully reviewed.

Electrocardiographic changes in patients with heartfailure are frequent. The negative predictive value of anormal ECG to exclude left ventricular systolic dysfunc-tion exceeds 90%[4447]. On the other hand, the presenceof anterior Q waves and a left bundle branch block inpatients with ischaemic heart disease are good predictorsof a decreased ejection fraction[14]. ECG signs of leftartrial overload or left ventricular hypertrophy may beassociated with systolic as well as isolated diastolicdysfunction, but they have a low predictive value. The

ECG is crucial for detecting atrial fibrillation or flutterand sometimes ventricular arrhythmia as causative orcontributing factors for heart failure. The diagnosticcontribution of ECG anomalies markedly increases ifclinical symptoms and signs of heart failure co-exist.ECG recordings do not need to be repeated in theabsence of changes of clinical status.

The chest X-ray Chest X-ray should be part of the initial diagnostic

work-up in heart failure.

A high predictive value of X-ray findings is onlyachieved by interpreting the X-ray in the context of

clinical findings and ECG anomalies[45]

. The investiga-tion is useful to detect the presence of cardiac enlarge-ment and pulmonary congestion[4851]. Cardiomegaly is

Table 2 New York Heart Association Classification of

Heart Failure

Class I. No limitation: ordinary physical exercise does notcause undue fatigue, dyspnoea or palpitations.

Class II. Slight limitation of physical activity: comfortable atrest but ordinary activity results in fatigue,palpitations or dyspnoea.

Class III. Marked limitation of physical activity: comfortableat rest but less than ordinary activity results insymptoms.

Class IV. Unable to carry out any physical activity withoutdiscomfort: symptoms of heart failure are presenteven at rest with increased discomfort with anyphysical activity.




6/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

frequently absent in patients with acute heart failure andalso in cases with diastolic dysfunction. In patients withchronic heart failure, however, an increased cardiac size,as judged by a cardiothoracic ratio >050, and thepresence of pulmonary venous congestion are usefulindicators of abnormal cardiac function with adecreased ejection fraction and/or elevated left ventricu-lar filling pressure[52]. Interstitial and alveolar pulmon-

ary oedema are also reliable and important signs ofsevere left ventricular dysfunction[53]. However, in indi-vidual patients the radiographic findings alone do notallow a reliable estimation of the pulmonary capillarypressure and are therefore not suitable as the onlybasis for therapeutic decisions[54]. There may also beinter-observer variations in the interpretation of chestX-ray changes[55,56]. The relationship between radio-logical signs and haemodynamic findings may dependon the duration as well as the severity of cardiacdysfunction[57].

Haematology and biochemistry

The following laboratory investigations are recom-mended as part of a routine diagnostic evaluation ofpatients with chronic heart failure: Complete bloodcount (Hb, leukocytes, platelets), S-electrolytes,S-creatinine, S-glucose, S-hepatic enzymes andurinalysis. Additional tests to consider include:C-reactive protein (CRP), thyroid stimulating hor-mone (TSH), S-uric acid and S-urea. In acute exacer-bations it is important to exclude acute myocardialinfarction by myocardial specific enzyme analysis.

Anaemia may exacerbate pre-existing heart failure. Araised haematocrit suggests that breathlessness may bedue to pulmonary disease, cyanotic congenital heartdisease or a pulmonary arteriovenous malformation.

Elevated serum creatinine can also be caused byprimary renal disease, which may induce all the featuresof heart failure by volume overload. Heart failure andrenal dysfunction often coincide because of the underly-ing diseases, such as diabetes and hypertension, or as aconsequence of impaired kidney perfusion by reductionin cardiac output. Further, age alone can be a causeof reduced creatinine clearance. For calculation ofcreatinine clearance, see Table 3. Excessive treatmentwith diuretics and/or ACE-inhibitors, sometimestogether with potassium-sparing diuretics, are otherreasons for a high s-creatinine value. Concomitantadministration of ACE inhibition and potassium-

sparing diuretics may lead to hyperkalaemia. Untreatedheart failue is rarely associated with major electrolytedisturbances, but they are quite common in patients on

diuretics. Liver enzymes may be elevated by hepaticcongestion.

Urine analysis is useful in detecting proteinuria and

glycosuria, alerting the clinician to the possibility ofunderlying renal problems or diabetes mellitus, condi-tions that may contribute to or complicate heart failure.

Heart failure due to thyrotoxicosis is frequentlyassociated with rapid atrial fibrillation which may bethe presenting feature of thyrotoxicosis in the elderly.Hypothyroidism may also present as heart failure.

Hyponatraemia and renal dysfunction in the setting ofheart failure indicate a bad prognosis.

Echocardiography As objective evidence of cardiac dysfunction at rest is

necessary for the diagnosis of heart failure, echo-cardiography is the preferred method for this

documentation.

The access to and use of echocardiography is encour-aged for the diagnosis of heart failure. TransthoracicDoppler echocardiography is rapid, safe and widelyavailable. It allows the assessment of chamber dimen-sions, wall thicknesses and geometry, indices of regional,global, systolic and diastolic ventricular function. Themost important parameter of ventricular function foridentifying patients with cardiac systolic dysfunctionand those with preserved systolic function is the leftventricular ejection fraction. Echocardiography alsoprovides rapid and semi-quantitiative assessment ofvalvular function, especially of mitral, tricuspid and

aortic stenosis and regurgitation and grading of mitralregurgitation. The degree of secondary tricuspidregurgitation gives an estimate of pulmonary arterypressures.

Although M-mode measurements benefit from hightemporal resolution, they are inaccurate in patients withspherical ventricles and regional dysfunction. The apicalbiplane summation of discs method modified Simp-son method is validated[58] but relies on accurateendocardial definition. Although quantitative visual as-sessment has been shown to detect low left ventricularejection fraction with good sensitivity and specificity,this procedure is only reliable with experienced observ-ers. Other measurements include: fractional shortening,sphericity index, atrioventricular plane displacement[59]

myocardial performance index[60], and left ventricularwall motion index[61]. Although eyeball grading of leftventricular systolic dysfunction into mild, moderate orsevere categories is widely used in clinical practice,clearly, standardization among different observers isdifficult to obtain[62]. The interpretation of ejectionfraction shortly after an acute myocardial infarction orin the context of a mitral insufficiency is more uncertain.

Reproducibility of ejection fraction among differentobservers is poor even when the same techniques areused. Preserved left ventricular systolic function, how-ever, usually implies a resting baseline left ventricular

ejection fraction ofd4045%, normal or, in the absenceof significant left valvular regurgitation, only slightlyenlarged ventricular volumes.

Table 3 Renal function estimated by a modified

creatinine clearance according to Cockroft and Gault[191].

Values should be reduced by 15% for women

Creatinine clearance=(140age)weight (kg)*122/S-creatinine(umoll1)




7/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

Doppler measurement may give additional informa-tion on cardiac filling characteristics. Commonly mademeasurements include isovolumic relaxation time, early

to atrial left ventricular filling ratio, early left ventricularfilling deceleration time, pulmonary venous atrial flowvelocity duration and ratio of pulmonary vein systolicand diastolic flow velocities. Age-related changes ofthese indices have been used. This helps provide evi-dence of slow left ventricular relaxation or reduced leftventricular diastolic distensibility.

Pulsed Doppler indices of left ventricular filling andpulmonary venous flow are influenced by several physio-logical variables, such as relaxation, compliance, heartrate, age and filling pressures and may be confounded bysuboptimal machine settings and arrhythmias. Morerecently, colour M-mode recordings of the left ventricu-

lar inflow and tissue Doppler diastolic myocardialvelocities have been incorporated into the differentDoppler filling patterns. All parameters are influencedby increasing age, which adds complexity to theinterpretation. In experienced hands, Doppler echo-cardiography provides haemodynamic measurements,such as cardiac output, stroke volume, pressure gradi-ents, valve area, mitral regurgitant volume and pulmon-ary artery pressures in the presence of tricuspid and/orpulmonary regurgitation.

Detailed diagnostic criteria for heart failure withdiastolic dysfunction (and preserved left ventricularfunction) have been proposed by the European StudyGroup on Diastolic Heart Failure[63]. However, there is

no universally accepted minimal criteria for the diag-nosis of diastolic dysfunction. In this context, it is alsoreasonable to include assessment of left atrial size byvolumetric data[64].

When the diagnosis of heart failure has been con-firmed by objective evidence of cardiac dysfunction,echocardiography is also helpful in determining itsaetiology. Primary valvular lesions can be identified.Regional akinesis or dyskinesis usually implies coronaryartery disease, especially in the presence of thin and/orecho-dense myocardium. Echocardiography may docu-ment constrictive pericarditis, cardiac amyloidosis orhypertrophic cardiomyopathy.

Transoesophageal echocardiography is not recom-mended routinely and can only be recommended inpatients who have an inadequate echo window, incomplicated valvular patients, in those with suspecteddysfunction of mechanical mitral valve prosthesis orwhen it is mandatory to identify or exclude a thrombusin the atrial appendage.

Repeated echocardiography can only be recom-mended in the follow-up of patients with heart failurewhen there is an important change in the clinical statussuggesting significant improvement or deterioration incardiac function.

Additional non-invasive tests to be consideredIn patients where echocardiography at rest has notprovided enough information and in patients with

coronary artery disease e.g. severe or refractory chronicheart failure and coronary artery disease, further, non-invasive imaging may include:

Exercise or pharmacological stress echocardiographymay be useful for detecting ischaemia as a cause ofreversible or persistent dysfunction and in determin-ing the viability of akinetic myocardium[65]. Gradeddobutamine infusion may be used to recruit contractilereserve[66]. Sustained contractile improvement isobserved when flow reserve is appropriate, in thepresence of stunning or non-transmural infarction. Abiphasic response indicates that flow reserve is bluntedand suggests the presence of myocardial hibernation.Although several non-controlled studies have shown

that revascularization can improve regional function,clinical status and survival in patients with a significantamount of hibernating myocardium[67,68], a systematicassessment of myocardial viability in patients withcoronary artery disease and heart failure with systolicdysfunction cannot yet be recommended.

Radionuclide angiography (RNA) provides reasonablyaccurate measurements of left, and to a lesser extent,right ventricular ejection fraction and cardiac volumes.Left ventricular filling dynamics can also be analysed.None of these measurements are reliable in the presence

of atrial fibrillation. Planar scintigraphy or singlephoton emission computed tomography (SPECT) canbe performed at rest or during stress using infusions ofdifferent agents, such as thallium201 or 99m technetiumsestamibi. The presence and extent of ischaemia can beevaluated. Although each of these imaging modalitiesmay have certain diagnostic and prognostic value, theroutine use of nuclear cardiology cannot be recom-mended. As with echocardiography, values of ejectionfraction vary with the technique used. Thus, analysisusing a single region of interest give values significantlylower than when two regions are used. However, repro-ducibility is considerably better than echocardiography.

()CMR is the most accurate and reproducible method forthe measurement of cardiac volumes, wall thicknessesand left ventricular mass. It also reliably detects thick-ened pericardium and quantitates myocardial necrosis,perfusion and function. Quantitative biochemical infor-mation, especially on myocardial energetics, can also beobtained by magnetic resonance spectroscopy. Thisinformation is presently used as a research tool. Atthe present time, CMR is only recommended if otherimaging techniques have not provided a satisfactorydiagnostic answer. However, CMR is a powerful tech-nique and it is too early to define its role in patients

with heart failure[69]

. It is not difficult to imagine thatthe development of new, and presumably expensive,medicines that have the ability to retard the progression




8/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

or even reverse cardiac disease, will necessitate a greaterdegree of quantitation of cardiac dysfunction thancurrently is accepted.

,

Variability within different modalities has been referredto. These are less with RNA and probably least withCMR. Variability between modalities, however, is strik-ing but insufficiently appreciated. In the same individual,values of ejection fraction for example are systematicallyhigher with echo than with RNA and values for CMRseem higher still. This emphasizes the necessity for eachinstitution carrying out non-invasive assessment ofventricular function to know the normal values for theequipment in their hands since the absolute values ofejection fraction will be different for all three modalities.

Pulmonary function Measurements of lung function are of little value

in diagnosing chronic heart failure. However, theyare useful in excluding respiratory causes ofbreathlessness.

Epidemiological studies suggest that there is a strongassociation between chronic obstructive airways diseaseand ischaemic heart disease, one of the principal causesof heart failure[70]. Forced vital capacity (FVC) is also avalid marker for evaluation of severity (level) andtherapy in patients with chronic heart failure[71]. FVCand forced expiratory volume (FEV1) correlate with

maximum oxygen consumption (VO2 max.) in patientswith chronic heart failure. Peak expiratory flow rate(PEFR) and FEV1 are reduced by chronic heart failurebut not to the same extent as in symptomatic obstructiveairways disease[72]. Other parameters have no value indiagnosing or in grading disease progression in patientswith chronic heart failure[73].

Dyspnoea and fatigue are the main causes ofexercise limitation in patients with chronic heart failure.Respiratory muscle dysfunction may also play animportant role[74].

Exercise testing In clinical practice exercise testing is of limited value

for the diagnosis of heart failure. However, a normalmaximal exercise test, in a patient not receivingtreatment for heart failure, excludes heart failure as adiagnosis. The main applications of exercise testing inchronic heart failure are more focused on func-tional and treatment assessment and on prognosticstratification.

Recommendations for exercise testing in heart failurepatients have been recently released by the WorkingGroup on Cardiac Rehabilitation & Exercise Physiologyand Working Group on Heart Failure of the EuropeanSociety of Cardiology[75].

Accurate assessment of functional capacity requires

that the patient is familiar with what is required. Ideallythe test should be individualized and of sufficient dur-ation to achieve target end-points within 812 min.

Small increments in workload should be used betweenstages. The ramp approach (treadmill or bicycle ergom-eter) appears to facilitate these recommendations for

maximal testing. Oxygen uptake is a more stable andreliable measure of exercise tolerance than exercise time.A marked fall in oxygen saturation, PaO2 and arterio-venous oxygen difference, a reduced ventilatory reserve,a normal oxygen pulse and a normal ratio between VO2and workload suggest pulmonary disease.

In recent years, exercise testing has been used forprognostic purposes and exercise capacity is an import-ant component of the risk profile in chronic heartfailure. A peak VO2 18 ml . kg

1 min1 identifieslow risk patients, respectively. Values between thesecut-off limits define a gray area of medium risk

patients, without further possible stratification byVO2. The available prognostic data for women areinadequate. For submaximal testing the 6 min walktest may provide useful prognostic information whenwalking distance is


9/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

responding promptly to initial and appropriate treat-ment, dynamic mitral regurgitation in conjunction withvolume overload or exercise, when chronic lung disease

is a differential diagnosis and in patients with refractoryheart failure not responding to appropriate treatment.Routine right heart catheterization should not be usedto tailor chronic therapy.

Endomyocardial biopsy may be useful in selectedpatients with unexplained (myocardial ischaemiaexcluded) heart failure. Biopsy may help to differentiatebetween constrictive and restrictive aetiologies.

Natriuretic peptides Plasma concentrations of certain natriuretic peptides

can be helpful in the diagnostic process, especially inuntreated patients.

Several clinical and epidemiological studies have relateddecreasing cardiac, usually left ventricular function withincreasing plasma natriuretic peptide concentrations,raising the possibility of a diagnostic blood test forheart failure[4,29,78]. Most extensively characterized inthis respect are N terminal atrial natriuretic peptide(NT ANP)[29], brain natriuretic peptide (BNP) and itsprecursor, N terminal pro BNP (NT pro BNP).

Natriuretic peptides have been investigated in heartfailure, asymptomatic left ventricular dysfunction andacute myocardial infarction. These peptides may bemost useful clinically as a rule out test due to consistentand very high negative prediction values[3,78,79].

Patients suspected of having heart failure, especially inprimary care, can be selected for further investigation byechocardiography or other tests of cardiac function onthe basis of having an elevated plasma concentration ofa natriuretic peptide. In those in whom the concen-trations are normal, other causes of dyspnoea andassociated symptoms should be considered.

The added value of natriuretic peptides in this situ-ation has yet to be determined. Epidemiological studiessuggest that the negative predictive value remains highwhen individuals at high risk of left ventricular dysfunc-tion are targeted. Future clinical studies will establishthe optimal role of natriuretic peptides in the diagnosisof heart failure in screening strategies.

High levels of natriuretic peptides identify those atgreatest risk of future serious cardiovascular eventsincluding death[8082]. There is also recent evidencethat adjusting heart failure therapy in order to reducenatriuretic peptides levels in individual patients mayimprove outcome[83].

Recent developments include user friendly assays ofincreasing rapidity, suitable for use in real time clinicalpractice. Time and clinical experience will tell which willbe the most successful.

Other neuroendocrine evaluations

Other tests of neuroendocrine activation are notrecommended for diagnostic or prognostic purposesin individual patients.

Whereas there are no doubts about the importance ofneuroendocrine mechanisms in the pathogenesis of heartfailure, the role of neuroendocrine factors in the diag-

nosis is less clear. In large cohorts of patients, there isgood evidence that circulating levels of noradrenaline,renin, angiotensin II, aldosterone, endothelin-1 andadrenomedullin are related to the severity and prognosisof heart failure, but in individual patients these predic-tors are inaccurate and difficult to interpret. Diuretics,vasodilator agents, ACE inhibitors and beta-blockersalter plasma concentrations of neuroendocrine sub-stances in a complex fashion which limits diagnostic use.Plasma noradrenaline rises with age and healthy subjectsover the age of 75 years may have plasma concentrationsof noradrenaline in the heart failure range[84].

Holter electrocardiography (ambulatory ECG, long timeECG recording-LTER) Conventional Holter monitoring is of no value in the

diagnosis of chronic heart failure, though it maydetect and quantify the nature, frequency, and dur-ation of atrial and ventricular arrhythmias, whichcould be causing or exacerbating symptoms of heartfailure. Ambulatory electrocardiographic monitoringshould be restricted to patients with chronic heartfailure and symptomatic arrhythmias.

The high prevalence of ventricular ectopy and ventricu-lar tachycardia is well recognised, but it remains unclearwhether ventricular arrhythmias identify patients at highrisk of sudden death. In the GESICA trial, patients withnon-sustained ventricular tachycardia were found tohave significantly more severe heart failure, a higheroverall mortality, and a greater incidence of suddendeath[85]. However, multivariate analysis of CHF-STATand PROMISE studies support that ventricular arrhyth-mias are non-specific predictors of mortality. Thus am-bulatory electrocardiographic monitoring alone seemsnot to provide additional prognostic information[86].Furthermore, the finding of asymptomatic complexventricular arrhythmias on ambulatory electrocardio-graphic monitoring does not identify specific candidatesfor antiarrhythmic or device therapy.

Heart rate variability is a marker of autonomic balanceand is reduced in heart failure. The diagnostic andprognostic utility of this observation has beenextensively investigated[8789]. Correlation between timeand frequency domain HRV measures and clinical andhaemodynamic variables exists[90,91], and time domainparameters can predict survival independently fromclinical and haemodynamic data[86,92,93]. Although thesedata have recently been confirmed in a large, prospec-tive, multicentre study[87]. The value of this technologyin clinical practice still remains to be determined.

Requirements for the diagnosis of heart failure in

clinical practiceTo satisfy the definition of heart failure, symptomsof heart failure and objective evidence of cardiac




10/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

dysfunction must be present (Table 1). The assessmentof cardiac function by clinical criteria alone is unsatis-factory. Cardiac dysfunction should be assessed objec-tively. The echocardiogram is the single most effectivetool in widespread clinical use. A diagnosis of heartfailure also requires the presence of symptoms and/orsigns suggestive of the diagnosis and cannot be made by

any single laboratory test. Other conditions may mimicor exacerbate the symptoms and signs of heart failureand need to be excluded (Table 4). An approach to the

diagnosis of heart failure in symptomatic patients ispresented in Fig. 2, and should be performed routinelyin patients with suspected heart failure in order toestablish the diagnosis. Additional tests (Table 5) shouldbe performed or re-evaluated in cases where diagnosticdoubt persists or clinical features suggest a reversiblecause for heart failure. Coronary artery disease is a

common, and probably underdiagnosed, cause of heartfailure. If there is reason to believe that the patient willbenefit from revascularization then an angiogram should

Table 4 Assessments to be performed routinely to establish the presence and likely

cause of heart failure

AssessmentsThe diagnosis of heart failure Suggests alternative or

additional diagnosisNecessary for Supports Opposes

Appropriate symptoms + + + + +(if absent)

Appropriate signs + + + +(If absent)

Cardiac dysfunction on + + + + + +imaging (usuallyechocardiography)

(If absent)

Response of symptoms or + + + + + +signs to therapy (If absent)

ECG + + +(If normal)

Chest X-ray If pulmonary + Pulmonary diseasecongestion or (If normal)cardiomegaly

Full blood count Anaemia/secondaryPolycythaemia

Biochemistry and Renal or hepaticurinalysis Disease/diabetes

Plasma concentration of + + ++natriuretic peptides in (If elevated) (If normal)untreated patients(where available)

+ =of some importance; + + + =of great importance.

Table 5 Additional tests to be considered to support the diagnosis or to suggest

alternative diagnosis

TestsThe diagnosis of heart failure Suggests alternative or

additional diagnosesSupports Opposes

Exercise test + + + +(If impaired) (If normal)

Pulmonary function tests Pulmonary diseaseThyroid function tests Thyroid diseaseInvasive investigation and Coronary artery disease,

angiography ischaemiaCardiac output + + + + + +

(If depressed at (If normal;rest) especially

during exercise)Left atrial pressure + + + + + +(If elevated at (If normal; in

rest) absence of therapy)




11/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

be done. Figure 2 represents a simplified plan for theevaluation of a patient presenting with symptoms sug-gestive of heart failure. Table 6 provides a managementoutline which connects the diagnosis part of the guide-lines with the treatment section.

Treatment of heart failure

Introduction

Throughout the last decade, the therapeutic approachto heart failure has undergone considerable change.

Current treatment not only concerns symptomaticimprovement, but focuses increasingly on preventing thetransition of asymptomatic cardiac dysfunction tosymptomatic heart failure, modulating progression ofheart failure and reducing mortality. As this is likely tobe a slow process, the effect of novel preventive therapiesmay only become apparent after a time, in contrastto the often more rapid effects of pure symptomatictreatment. Thus, short- and long-term objectives withindividualized therapies should be identified. Importanttreatment targets include cardiac remodelling, neuro-endocrine and cytokine activation, fluid retention andrenal dysfunction. Accordingly, heart failure being acomplex syndrome, the therapeutic approach may needseveral strategies in combination with target differentmechanisms.

However, as the therapeutic approaches to heartfailure are multiple, including general measures, phar-macological therapy, mechanical devices and surgicalinterventions, they will not always be applicable in eachpatient. Adverse effects and interaction between differ-ent forms of treatment may preclude their use in some.Rapid deterioration of the clinical condition can requiremodification of the therapeutic approach.

There are regional differences in the approach toheart failure treatment in Europe. These differences areattributable to variations in aetiology and in health

resources. Of more importance, perception and accept-ance of the usefulness of and need to prescribe therapiesproven to be effective in large controlled trials by the

different physicians taking care of heart failure patients,is slow. Continuous education is clearly needed.

Aims of treatment in heart failure

The aims of heart failure management are those of thetreatment of any disease in general and consist of severalcomponents (Table 7).

Prevention of heart failure

The prevention of heart failure should always be aprimary objective. Many potential causes of myocardialdamage can be treated and the extent of myocardial

damage reduced. Examples include management ofrisk factors for coronary heart disease, treatment ofischaemia, early triage of acute myocardial infarction,prevention of reinfarction, accurate identification andaggressive treatment of hypertension and some causes ofspecific heart muscle disease and timely correction ofvalve disorders and congenital heart disease. However,primary prevention of cardiac dysfunction and heartfailure is a large topic, which falls outside the scope ofthe current Guidelines.

When myocardial dysfunction is already present, thefirst objective is to remove the underlying cause ofventricular dysfunction if possible (e.g. ischaemia, toxicsubstances, alcohol, drugs, thyroid disease). The second

objective of modern therapy is to modulate progressionfrom asymptomatic left ventricular dysfunction toheart failure. How to modulate progression fromasymptomatic left ventricular dysfunction to heartfailure is described on page 1550 under Treatment ofAsymptomatic Left Ventricular Dysfunction.

Management of chronic heart failure

The therapeutic approach in chronic heart failure due tosystolic cardiac dysfunction consists of general adviceand other non-pharmacological measures, pharmaco-

logical therapy, mechanical devices and surgery. Thecurrently available types of management are outlined inTable 8.

Table 6 Management outline

1. Establish that the patient has heart failure (in accordance with

the definition presented on page 1528 diagnosis section)2. Ascertain presenting features: pulmonary oedema, exertional

breathlessness, fatigue, peripheral oedema3. Assess severity of symptoms4. Determine aetiology of heart failure5. Identify precipitating and exacerbating factors6. Identify concomitant diseases relevant to heart failure and its

management7. Estimate prognosis8. Anticipate complications9. Counsel patient and relatives

10. Choose appropriate management11. Monitor progress and manage accordingly

Table 7 Aims of treatment

1. Prevention

(a) Prevention and/or controlling of diseases leading to cardiacdysfunction and heart failure

(b) Prevention of progression to heart failure once cardiacdysfunction is established

2. MorbidityMaintenance or improvement in quality of life

3. MortalityIncreased duration of life




12/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

The approach to the treatment of specific patientsubgroups, i.e. the elderly, or heart failure due topredominant diastolic dysfunction, is addressed in

special sections of these guidelines. The treatment ofacute heart failure, pulmonary oedema and cardiogenicshock will be presented in a future document.

Non-pharmacological management

General advice and measures

(Level of evidence C for all advice and measures unlessstated otherwise.)

Patients with chronic heart failure and their closerelatives should receive general advice (Table 9).

Patients are advised to weigh themselves on a regularbasis (once a day, twice a week) and, in case of a suddenunexpected weight gain of more than 2 kg in 3 days, toalert a health care provider or adjust their diuretic doseaccordingly, e.g. to increase the dose if a sustainedincrease is noted.

Sodium Controlling the amount of salt in the diet is aproblem that is more relevant in advanced heart failurethan mild failure. Salt substitutes must be used with

caution, as they may contain potassium. In largequantities, in combination with an ACE inhibitor, theymay lead to hyperkalaemia[94].

Fluids Fluid intake must be reduced in patients withadvanced heart failure, with or without hyponatremia.The exact amount offluid restriction remains unclear. Inpractice, a fluid restriction of 152 litres is advised inadvanced heart failure.Alcohol Moderate alcohol intake is permitted. Alcoholconsumption must be prohibited in suspected cases ofalcoholic cardiomyopathy. Light-to-moderate alcoholconsumption has been reported to improve prognosis in

patients with left ventricular dysfunction[95].

Treatment of chronic heart failure should include weightreduction in the overweight or obese. The subject isoverweight if his/her body mass index (i.e. the actualweight in kilograms divided by height in meters squared)lies between 25 and 30 and obese if it is >30.

Clinical or subclinical malnutrition is present in about50% of patients with severe chronic heart failure[4]. Thewasting of total body fat and lean body mass that

accompanies weight loss is called cardiac cachexia[96].Cardiac cachexia is an important predictor of reducedsurvival[97].

Table 8 Treatment options general advice and

measures, exercise and exercise training, pharmacological

therapy, devices and surgery

Non-pharmacological management General advice and measures Exercise and exercise training

Pharmacological therapy Angiotensin-converting enzyme (ACE) inhibitors Diuretics Beta-adrenoceptor antagonists Aldosterone receptor antagonists Angiotensin receptor antagonists Cardiac glycosides Vasodilator agents (nitrates/hydralazine) Positive inotropic agents Anticoagulation Antiarrhythmic agents

OxygenDevices and surgery Revascularization (catheter interventions and surgery), other

forms of surgery Pacemakers Implantable cardioverter defibrillators (ICD) Heart transplantation, ventricular assist devices, artificial

heart Ultrafiltration, haemodialysis

Table 9 List of subjects to discuss with a heart failure

patient and his family

General advice Explain what heart failure is and why symptoms occur Causes of heart failure How to recognize symptoms What to do if symptoms occur Self-weighing Rationale of treatments Importance of adhering to pharmacological and non-

pharmacological prescriptions Refrain from smoking Prognosis

Drug counselling Effects Dose and time of administration Side effects and adverse effects Signs of intoxication What to do in case of skipped doses Self-management

Rest and exercise Rest Work Daily physical activity Sexual activity Rehabilitation

VaccinationsTravelDietary and social habits Control sodium intake when necessary, e.g. some patients

with severe heart failure Avoid excessive fluids in severe HF Avoid excessive alcohol intake Stop smoking




13/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

Consider the possibility of abnormal weight losswhen(a) a body weight


14/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

left ventricular performance itself are important deter-minants of exercise capacity. Several small clinical andmechanistic studies and some randomized trials showed

that regular exercise can safely increase physical capacityby 1525%, improve symptoms and perception ofquality of life in patients with stable class II and III heartfailure (level of evidence B). No significant deleteriouseffects or significant deterioration in central haemo-dynamics have been reported with exercise training.

Standardized recommendations for exercise trainingin heart failure patients by the European Society ofCardiology have recently been published[100].

The exercise training can be performed by eitherinterval or steady state exercise, applying intensities of6080% of the predetermined peak heart rate. Intervaltraining methods may allow for more intense exercisestimuli on peripheral muscles than obtained during

steady state training, but without inducing greatercardiovascular stress. Exercise training should be per-formed in the following order: duration, then frequency,then intensity. Details are provided in Table 10.

Pharmacological therapy

Angiotensin-converting enzyme inhibitors ACE inhibitors are recommended as first-line therapy

in patients with a reduced left ventricular systolicfunction expressed as a subnormal ejection fraction,i.e.


15/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

The absolute benefit is greatest in patients with mostsevere heart failure. ACE inhibition markedly enhancessurvival in patients with signs or symptoms of heartfailure during the acute phase of myocardial infarction,even if the symptoms are transient[103]. In addition tothese effects of mortality, ACE inhibitors in generalimprove the functional status of patients with heartfailure. In contrast, only small benefits in exercise capac-ity occur. Moreover, whereas ACE inhibitors may pre-vent further deterioration of left ventricular functionand attenuate further cardiac dilatation, they do not

consistently reduce cardiac size[104,105]. Also, beneficialeffects on cardiac function may quickly diminish inmagnitude upon cessation of ACE inhibition[104,106].

ACE inhibitors should always be uptitrated to thetarget dose used in large controlled clinical trials, iftolerated, to reduce long-term morbidity and mor-tality. ACE inhibitors should not be titrated based onsymptomatic improvement.

Important adverse effects associated with ACEinhibitors are hypotension, syncope, renal insufficiency,hyperkalaemia and angioedema. Although cough mayoften be due to heart failure or concomitant diseases,e.g. respiratory disease, dry cough is a side effect of ACEinhibitors. Severe cough may lead to discontinuation ofACE inhibitor therapy. Some patients may toleratereinstitution of the ACE inhibitor after a drug-freeperiod. Substitutes for ACE inhibitors in this situationinclude angiotensin receptor antagonists or, if contra-indicated, the combination of high-dose nitrate andhydralazine. Observational data would argue againstusing an NSAID to suppress cough, as heart failure maydeteriorate. Sodium cromoglycate or thromboxane syn-thetase inhibition may suppress cough during ACEinhibitor treatment[107].

Changes in systolic and diastolic blood pressure andincreases in serum creatinine are usually small in normo-tensive patients. Moderate renal insufficiency and a

relatively low blood pressure (serum creatinine up to250 mol . l1 and systolic blood pressure as low as90 mmHg) are no contraindications to ACE inhibitor

treatment. Serum creatinine might increase by 1015%in patients with severe heart failure, irrespective ofbaseline serum creatinine[108]. In most of these patientscreatinine levels will either remain stable or decreasetowards pre-treatment values during continued treat-ment. It should be stressed that mortality is higheramong patients with elevated creatinine levels and thatthese patients in particular benefit from treatment withACE inhibitors. The risk of hypotension and renaldysfunction increases in patients with severe heart fail-ure, those treated with high doses of diuretics, elderly

patients and patients with renal dysfunction or hypo-natraemia. In addition, changes in serum potassium areusually small (02 mmol . l1). Mild hyperkalaemia is nocontraindication to use ACE inhibitors. However, serumpotassium levels >55 mmol . l1 are a contraindication.If potassium-sparing diuretics were prescribed to correctserum potassium levels they should be stopped duringinitiation of ACE inhibitor therapy.

Absolute contraindications for initiation of ACEinhibitor treatment are bilateral renal artery stenosis andangioedema during previous ACE inhibitor therapy.

The effect of ACE inhibition in heart failure has beendocumented in target doses that are usually higher thanthose used in clinical practice. Furthermore, in theATLAS trial morbidity expressed as hospitalizations forheart failure, was less in patients with a higher than alower dose regimen[109]. Target maintenance dose rangesof ACE inhibitors shown to be effective in various trialsare shown in Table 11.

Recommended initiating and maintenance dosages ofACE inhibitors which have been approved for thetreatment of heart failure in Europe are presented inTable 12. The dose of ACE inhibitor should always beinitiated at the lower dose level and titrated to the targetdose. The recommended procedures for starting an ACEinhibitor are given in Table 13.

( 13)Until further trials are completed, the dose of the chosenACE inhibitor should be titrated up to the maximum

Table 11 Doses of ACE inhibitors shown to be effective in large, controlled trials of

heart failure or left ventricular dysfunction

Studies of mortality

Drug Target dose Mean daily dose

Studies in chronic heart failureConsensus Trial Study Group, 1978[192] Enalapril 20 mg b.i.d. 184 mgCohn et al. (V-HeFT II, 1991)[142] Enalapril 10 mg b.i.d. 150 mgThe SOLVD Investigators, 1991[193] Enalapril 10 mg b.i.d. 166 mgATLAS[109] Lisinopril High dose: 32535 mg daily

Low dose: 255 mg daily

Studies after MI LV dysfunction with or without HFPfeffer et al (SAVE, 1992)[194] Captopril 50 mg t.i.d. (not available)AIRE[103] Ramipril 5 m b.i.d. (not available)TRACE[101] Trandolapril 4 mg daily (not available)

LV=left ventricular; MI=myocardial infarction; HF=heart failure.




16/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

target dose used in clinical trials. Careful attentionshould be given to the locally approved prescribinginformation when initiating therapy.

Regular monitoring of renal function is recom-mended: (1) before, 12 weeks after each dose increment,at 3 months, and at 6 monthly intervals; (2) whentreatment is changed, which may affect renal function;(3) in patients with past or present renal dysfunction orelectrolyte disturbances, more frequent measurementsshould be made.

Care should be taken in patients with a low systolicblood pressure or a serum creatinine above250 mol . l1. Patients with a systolic level below

100 mmHg should have therapy initiated under special-ist medical care. Low blood pressures (


17/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

inhibition or, in severe heart failure despite thecombination ACE inhibition and low-dose spirono-lactone (level of evidence C).

Potassium supplements are less effective in thissituation (level of evidence B).

Most patients on diuretics for heart failure will also betreated with an ACE inhibitor. Until recently the com-bination of potassium sparing diuretics and ACE inhibi-tors was regarded as potentially dangerous. One small,controlled study suggested that the administration ofspironolactone at dosages that result in diuresis andnatriuresis, i.e. 50100 mg, to patients who are notresponding to loop diuretics and ACE inhibition,may result in rapid weight reduction without hyper-kalaemia[114]. However, at present potassium-sparingdiuretics such as triamterene, amiloride and relativelyhigh dosages of spironolactone should only be con-sidered if there is persisting diuretic-induced hypo-kalaemia despite concomitant ACE inhibitor therapy, or

in severe heart failure, despite concomitant ACE inhibi-tion plus low-dose spironolactone (level of evidence C).Similar restrictions also pertain in cases of intolerance to

ACE inhibition and replacement therapy with angio-tensin receptor blockers. Oral potassium supplementsare less effective in maintaining body potassium storesduring diuretic treatment[115]. In general, the use of allpotassium sparing diuretics should be monitored byrepeated measurements of serum creatinine and potass-

ium. A practical approach is to measure serum creati-nine and potassium every 57 days after initiation oftreatment until the values are stable. Thereafter,measurements can be made every 36 months.

Beta-adrenoceptor antagonists Beta-blocking agents are recommended for the treat-

ment of all patients with stable, mild, moderate andsevere heart failure from ischaemic or non-ischaemiccardiomyopathies and reduced left ventricular ejec-tion fraction, in NYHA class II to IV, on standardtreatment, including diuretics and ACE inhibitors,unless there is a contraindication (level of evidence

A). In patients with left ventricular systolic dysfunction,

with or without symptomatic heart failure, following

Table 14 Diuretics

Initial diuretic treatment

Loop diuretics or thiazides.Always administered in addition to an ACE inhibitor.

If GFR


18/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

an acute myocardial infarction long-term beta-blockade is recommended in addition to ACEinhibition to reduce mortality (level of evidence B).

The first recommendation is based on data obtainedfrom large and small studies including over 15 000patients[116127]; the second on the recently publishedCAPRICORN study with carvedilol[195]. In severallarge, randomized, placebo-controlled mortality trialscarvedilol[121,124,196], bisoprolol[125] and metopro-lol[126,127] have been associated with a long-term reduc-tion in total mortality, cardiovascular mortality, suddendeath and death due to progression of heart failure inpatients in functional class II-IV. In these studies, beta-blocking therapy also reduces hospitalizations (all,cardiovascular and heart failure), improves the func-tional class and leads to less worsening of heart failurethan placebo. This beneficial effect has been consistentlyobserved in subgroups of different age, gender,functional class, left ventricular ejection fraction andischaemic or non-ischaemic aetiology (level ofevidence A).

Although a reduction in mortality and hospitalizationhas been demonstrated with several beta-blockers inchronic heart failure, a class-effect has not been estab-

lished. In one large trial, no benefit on survival wasobserved with bucindolol (BEST)[128]. Accordingly, onlybisoprolol, carvedilol and metoprolol can be recom-

mended at present. A direct comparison of differentbeta-blockers is currently being evaluated in theCOMET trial (metoprolol vs carvedilol). In smaller,controlled studies beta-blockade has been shown toimprove ventricular function[120,129,130]. In contrast,exercise capacity usually does not improve.

Further data are needed to establish the effects ofbeta-blocking agents in certain demographic groups,such as elderly subjects (>75 years), certain racial sub-sets and atrial fibrillation. In SENIORS the effect ofbeta-blockade (nebivolol) in the elderly patient with

heart failure is investigated.

As beta-blocker action may be biphasic with long-termimprovement, possibly preceded by initial worsening,beta-blockers should be initiated under careful control.The initial dose should be small and increased slowlyand progressively to the target dose used in the largeclinical trials. Up-titration should be adapted to indi-vidual response. Beta-blockers may reduce heart rateexcessively, may temporarily induce myocardial depres-sion and precipitate heart failure. In addition, beta-blockers may initiate or exacerbate asthma and induce

peripheral vasoconstriction. Table 16 gives the recom-mended procedure for the use of beta-blockers in clinicalpractice and contraindications. Table 17 shows the

Table 16 The recommended procedure for starting a beta-blocker

1. Patients should be on a background therapy with ACE inhibition, if not contraindicated.

2. The patient should be in a relatively stable condition, without the need of intravenous inotropictherapy and without signs of marked fluid retention.

3. Start with a very low dose (Table 17) and titrate up to maintenance dosages shown to beeffective in large trials. The dose may be doubled every 12 weeks if the preceding dose was welltolerated. Most patients can be managed as out-patients.

4. Transient worsening failure, hypotension or bradycardia may occur during the titration periodor thereafter Monitor the patient for evidence of heart failure symptoms, fluid retention, hypotension and

bradycardia If worsening of symptoms, first increase the dose of diuretics or ACE-inhibitor; temporarily

reduce the dose of beta-blockers if necessary If hypotension, first reduce the dose of vasodilators; reduce the dose of the beta-blocker if

necessary Reduce or discontinue drugs that may lower heart rate in the presence of bradycardia; reduce

dose of beta-blockers if necessary, but discontinue only if clearly necessary. Always consider the reintroduction and/or uptitration of the beta-blocker when the patient

becomes stable.5. If inotropic support is needed to treat a decompensated patient on beta-blockade, phosphodi-

esterase inhibitors should be preferred because their haemodynamic e ffects are not antagonizedby beta-blocker agents.

The following patients should be referred for specialist care: Severe heart failure Class III/IV Unknown aetiology Relative contraindications: bradycardia, low blood pressure Intolerance to low doses Previous use of beta-blocker and discontinuation because of symptoms Suspected asthma or bronchial disease

Contraindications to beta-blockers in patients with heart failure Asthma bronchiale Severe bronchial disease Symptomatic bradycardia or hypotension




19/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

titration scheme of the drugs used in the most relevantstudies.

Aldosterone receptor antagonists spironolactone

Aldosterone antagonism is recommended inadvanced heart failure (NYHA III-IV), in addition toACE inhibition and diuretics to improve survival andmorbidity (level of evidence B).

Although spironolactone was developed as a diureticagent at a higher dose level, it is now understoodthat aldosterone has an important role in the patho-physiology of heart failure. It promotes vascular andmyocardial fibrosis, potassium and magnesiumdepletion, sympathetic activation, parasympatheticinhibition and baroreceptor dysfunction[131,132]. ACEinhibitors insufficiently suppress circulating aldosteronelevels[133].

The RALES mortality trial showed that low dosespironolactone (12550 mg) on top of an ACE inhibitorand a loop diuretic markedly and progressivelyimproved survival of patients in advanced (NYHA classIII or IV) heart failure, irrespective of aetiology[134]. Atthis dose, spironolactone is believed not to have anappreciable diuretic effect. Both death from progressiveheart failure and sudden cardiac death were reduced inRALES and although only 11% received a beta-blocker,the mortality reduction was significant in this pre-specified subgroup. Whether an aldosterone antagonistcould be useful in patients with class II heart failure orasymptomatic left ventricular dysfunction remains to beestablished.

Administration and dosing considerations areprovided in Table 18.

If painful gynecomastia develops (10% in RALES),spironolactone may need to be stopped. The new selec-tive aldosterone receptor antagonist eplerenone, with alower affinity for androgen and progesterone receptorsthan spironolactone, may reduce the risk of gyneco-mastia, but needs further evaluation. Ongoing trialswill assess the effect of eplerenone on morbidity andmortality.

Angiotensin II receptor antagonists Angiotensin II receptor antagonists (ARBs) could be

considered in patients who do not tolerate ACE

inhibitors for symptomatic treatment (level ofevidence C).

However, it is unclear whether ARBs are as effectiveas ACE inhibitors for mortality reduction (level of

evidence B). In combination with ACE inhibition, ARBs may

improve heart failure symptoms and reduce hos-pitalizations for worsening heart failure (level ofevidence B).

Side effects, notably cough are significantly less thanwith ACE inhibitors[135]. In the majority of studies,mainly carried out in hypertension, the side effect profileof ARBs is comparable to placebo. The ELITE study,comparing the safety and tolerability of losartan withthat of captopril in elderly patients with heart failure,found no difference in the incidence of renal dysfunction

between the two drugs after 1 years follow-up[136].Monitoring of renal function is as essential withARBs as with ACE-inhibitors. ARBs studied or underinvestigation in heart failure are given in Table 19.

ARB ACE

Thus far, ARBs have not been shown to be superior toACE inhibitors, although side effects may be less withthe ARB[137]. In Elite II a possible negative interactionof losartan and beta-blocker therapy was observed.Further studies are ongoing.

ACE ARB

Angiotensin II can be produced from non-ACEpathways, and during angiotensin II receptor blockade

Table 17 Initiating dose, target dose and titration scheme of beta-blocking agents as used in recent large, controlled

trials

Beta-blocker First dose (mg) Increments (mg . day

1) Target dose (mg . day

1) Titration period

Bisoprolol[125] 125 25,375,5,75,10 10 weeksmonthMetoprolol tartrate[119] 5 10,15,30,50,75,100 150 weeksmonthMetoprolol succinate CR[126] 125/25 25,50,100,200 200 weeksmonthCarvedilol[121,196] 3125 625,125,25,50 50 weeksmonth

Daily frequency of administration as in the trials referenced above.

Table 18 Administration and dosing considerations with

spironolactone

1 Consider whether a patient is in severe heart failure (NYHA

III-IV) despite ACE inhibition/diuretics2 Check serum potassium (


20/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

levels of angiotensin II rise, thus competing with theantagonist. This has led to the hypothesis that thecombination of an ACE inhibitor and an ARB may bebeneficial. In VAL-HeFT, patients were randomized to

placebo or valsartan on top of standard therapy, whichincluded an ACE inhibitor in nearly all. The resultsshowed no difference in overall mortality, but a reduc-tion in the combined end-point all-cause mortality ormorbidity expressed as hospitalization because ofworsening heart failure[138]. In patients who alsoreceived a beta-blocker a trend towards a negative effectof the ARB was observed. Whether a possible inter-action truly exists needs to be clarified in further studies.In the ongoing CHARM-add on study, patients withheart failure and left ventricular dysfunction arerandomized to placebo or candesartan cilexitil on top ofan ACE inhibitor.

Cardiac glycosides Cardiac glycosides are indicated in atrial fibrillation

and any degree of symptomatic heart failure, whetheror not left ventricular dysfunction is the cause,in order to slow ventricular rate, thereby improv-ing ventricular function and symptoms (level ofevidence B)[139].

A combination of digoxin and beta-blockadeappears superior than either agent alone (level ofevidence C)[140].

In sinus rhythm, digoxin is recommended to improve theclinical status of patients with persisting heart failuresymptoms due to left ventricular systolic dysfunctiondespite ACE inhibitor and diuretic treatment (level ofevidence B). Insufficient data are available for patientswith heart failure due to left ventricular systolic dysfunc-tion and sinus rhythm on the combined treatment ofACE inhibition, beta-blockade, diuretics and, in severeheart failure, spironolactone.

Digoxin and digitoxin are the most frequently usedcardiac glycosides. They have identical pharmaco-dynamic effects, but different pharmacokinetic profiles.Elimination of digoxin is renal. In contrast, digitoxin ismetabolised in the liver and is less dependent on renalfunction, potentially useful in renal dysfunction and inelderly patients.

In the DIG trial in 6800 patients with an ischaemicand non-ischaemic cardiomyopathy and mild-to-moderate heart failure, long-term digoxin did not

improve survival and a small decrease in the risk ofdeath from heart failure was offset by an increase in therisk of death from other causes[141]. Thus, the primary

benefit and indication for digoxin in heart failure is toreduce symptoms and improve clinical status, andthereby to decrease the risk of hospitalization for heartfailure without an impact on survival.

Contraindications to the use of cardiac glycosidesinclude bradycardia, second- and third-degree AV block,sick sinus syndrome, carotid sinus syndrome, WolffParkinsonWhite syndrome, hypertrophic obstructivecardiomyopathy, hypokalaemia, and hypercalcaemia.

The usual daily dose or oral digoxin is 0250375 mg ifserum creatinine is in the normal range (in the elderly

06250125 mg, occasionally 025 mg). No loading doseis needed when treating chronic conditions. The treat-ment can be initiated with 025 mg bid. for 2 days. Renalfunction and plasma potassium should always bemeasured before starting treatment. In renal failure, thedaily doses should be reduced accordingly. As thedigoxin clearance closely approximates to creatinineclearance, the latter should be measured or calculated bythe Cockroft and Gault formula, which is provided inTable 3.

Vasodilator agents in chronic heart failure

There is no specific role for vasodilators in thetreatment of heart failure (level A), although theymay be used as adjunctive therapy for angina orconcomitant hypertension (level of evidence C).

In case of intolerance to ACE inhibitors ARBs arepreferred to the combination hydralazinenitrates(level of evidence A).

-

Vasodilator agents may be used as adjunctive therapy inthe management of heart failure. In previous guidelines,the hydralazine-isosorbide dinitrate combination wassuggested as an alternative when ACE inhibitors arecontraindicated or cannot be tolerated[2,142]. Recent datawould suggest that angiotensin-II type-1 antagonists arethe preferred medication in that situation (level ofevidence B, see page 1545).

Relatively high doses of hydralazine (up to 300 mg) incombination with high-dose isosorbide dinitrate (up to160 mg) without ACE inhibition may have somebeneficial effect on mortality, but not on hospitalizationfor heart failure[143]. At these doses, the combi-nation increased exercise performance more than withenalapril. There is no evidence of proven benefit wheneither nitrates or hydralazine are used alone in additionto current therapy.

Nitrates may be used for the treatment of con-

comitant angina or relief of acute dyspnoea. Earlydevelopment of haemodynamic tolerance (tachyphy-laxis) to nitrates may occur with frequent dosing (every

Table 19 Currently available angiotensin II receptor

antagonists

Drug Daily dose (mg)

Losartan 50100Valsartan 80320Irbesartan 150300Candesartan cilexetil 416Telmisartan 4080Eprosartan 400800




21/34


200

1Th

eEur

opeanS

ociety

ofCardi

ology

46 h), but is less with intervals of 8 to 12 h[144] or inconjunction with ACE inhibitors or hydralazine[145].

- There is no evidence to support the use of alpha-adrenergic blocking drugs in heart failure (level ofevidence B).

In general, calcium antagonists are not recommendedfor the treatment of heart failure due to systolic dysfunc-tion. Diltiazem and verapamil-type calcium antagonistsin particular are not recommended in heart failuredue to systolic dysfunction, and are contraindicated inaddition to beta-blockade (level of evidence C).

Newer calcium antagonists (felodipine, amlodipine) in

addition to baseline therapy including ACE inhibitorsand diuretics do not provide a better effect on survivalcompared to placebo [146,147] (level of evidence A).

As long-term safety data with felodipine andamlodipine indicate a neutral effect on survival, theymay be considered as additional therapy for con-comitant arterial hypertension or angina.

Positive inotropic therapy Inotropic agents are commonly used to limit severe

episodes of heart failure or as a bridge to hearttransplantation in end-stage heart failure (level ofevidence C). However, treatment-related complica-tions may occur and their effect on prognosis is not

well recognized. Repeated or prolonged treatment with oral inotropic

agents increases mortality (level of evidence A). Currently, insufficient data are available to recom-

mend dopaminergic agents for heart failure treatment.

Intravenous inotropic therapy is used to correct thehaemodynamic disturbances of severe episodes of wors-ening heart failure. The agent most often used in thissetting is dobutamine. However, its use has been insuf-ficiently documented in controlled trials and the effectsof dobutamine on prognosis are not well characterized.Problems related to use of dobutamine are tachy-phylaxis, increase in heart rate and often an inadequatevasodilatory effect. Similar problems are present withother cAMP-dependent inotropes, i.e. phospho-diesterase inhibitors such as amrinone, milrinone orenoximone.

In acute heart failure, intravenous milrinone does notreduce the number of hospitalizations or cardiovascularevents, but leads to a higher incidence of treatment-related complications, such as atrial fibrillation andhypotension as compared to placebo[147]. In studies withoral treatment, milrinone, enoximone, vesnarinone andamrinone invariably increase arrhythmias and mortality.

For acute worsening of heart failure, the short-term

administration of levosimendan, a new inotrope withcalcium-sensitising properties, appears to be saferthan dobutamine[148]. Also, in acute heart failure after

myocardial infarction, levosimendan improved symp-toms and halved mortality during the first 72 h, adifference in mortality which was maintained over the

next 6 months[149]. However, this effect on mortalityrequires further confirmation in formal trials.

Currently, there are insufficient data available to recom-mend oral dopamine analogues for the treatment ofheart failure. The dopaminergic agent ibopamine,which also has sympathomimetic properties, is notrecommended for the treatment of chronic heart failuredue to systolic left ventricular dysfunction (level ofevidence B)[150].

Intravenous dopamine is used for the sort-termcorrection of haemodynamic disturbances of severeepisodes of worsening heart failure. At low dosages(35 g . kg1 . min1) it may improve renal bloodflow. Its use has not been evaluated in prospectivecontrolled trials and its effects on prognosis are not welldefined.

Antithrombotic agents There is little evidence to show that antithrombotic

therapy modifies the risk of death, or vascular eventsin patients with heart failure other than in the settingof atrial fibrillation when anticoagulation is firmlyindicated (level of evidence C)[151], and prior myo-cardial infarction when either aspirin or oral anti-coagulants should be used as secondary prophylaxis.

Patients with heart failure are at high risk of thrombo-embolic events. Ischaemic heart disease is the common-est cause of heart failure and coronary vascularocclusion is the commonest vascular event in this popu-lation. However, patients with heart failure are also atgreatly increased risk of stroke and other vascularevents[152]. Many cases of sudden death may be precipi-tated primarily by vascular, rather than arrhythmicevents[153].

The reported annual risk of stroke in controlled heartfailure studies is between 1 and 2%, and the risk ofmyocardial infarction from 2 to 54%, respectively. Theannual risk of stroke in the Stroke Prevention of AtrialFibrillation study (SPAF) was 103% in atrial fibrillationpatients with definite heart failure and 177% in thosewith recent heart failure[154].

Oral anticoagulants reduce the risk of stroke in heartfailure patients with atrial fibrillation[155]. However,there is a lack of evidence to support the use ofantithrombotic therapy in patients in sinus rhythm, evenif they have had a previous vascular event or evidence ofintra-cardiac thrombus. There is little evidence to sug-gest that patients with layered left ventricular thrombusare at increased risk of thromb

guidelines for the diagnosis and treatment of chronic heart failure

Documents